Inaugural Lecture Series: Provost Research and Innovation, Professor Warren Manning video transcript
The title screen appears. A blue background is shown with the University of Derby three hills logo in white in the middle of the screen. The University of Derby's name in white is directly below. White text on the screen reads below:
Professorial Inaugural Lecture Series: Out of Control: The Evolution and Future of Motoring and Mobility by Professor Warren Manning
The title screen fades out and a live video of Paul Lynch on a webcam is shown. He is a man with grey hair. He is wearing glasses, a black suit jacket and a white shirt, with a blue flower-patterned tie.
[Paul] Good evening, I’m professor Paul Lynch, chairman of the university professorial council and on behalf of the university professorial council, I’d like to welcome you to Professor Warren Manning's inaugural lecture.
The lecture is entitled "Out of Control: The Evolution and Future of Motoring and Mobility." I would now like to introduce the university's Vice-Chancellor, Professor Kathryn Mitchell, who will formally introduce you to Professor Manning. Vice-chancellor.
The live video of Paul disappears, and a live webcam video of Kathryn Mitchell appears. She is a woman with short brunette hair and red-framed glasses. She is wearing a white turtleneck top with a black bead necklace.
[Kathryn] Thank you very much Paul and good evening to everyone. And, I’m very much delighted to be introducing warren this evening. So, I’m going to be very formal about it. I’m quite sad I’m not wearing my robes in person but it's fantastic that we're still going ahead with all of our inaugural lectures.
So, Professor Warren Manning graduated from the University of Salford in 1993 with a first-class honours degree in mechanical engineering. And, before I go on to the rest of his personal history, I do need to acknowledge that Warren comes from Wigan, which is very close to my hometown of Saint Helens and we share a very early connection. Warren's secondary headteacher was somebody called Dennis Lavelle and he was a great friend of my mother's. Both believers in the power of education as being significant for social mobility, and I’ve spent quite a lot of today trying to re-contact with him because I do know that Dennis would be really delighted to listen to this inaugural lecture and I’m going to share it with him in the next week.
So, from Salford, Warren joined The University of Leeds to undertake a PhD in intelligent control systems; investigating how fuzzy neural networks can improve the control of satellites with flexible arrays. After graduating from his PhD in 1997, he spent a brief while working on vibration control of aircraft rotors and smart controller flexible beams, before starting a research career in vehicle dynamics and control in 1999. I think warren was incredibly privileged because working under the tutelage of the internationally renowned vehicle dynamicist Professor Dave Crolla. He studied how active dynamics control systems can improve vehicle ride and handling using collaborations with; the Rover group, BMW, TRW and Red Bull Racing, to help develop their technology and analytical tools. And, I think many of the advances that we saw were very much under that research and Dave Crolla was world-famous in how that changed some of the ways that we have the transport of today. During that time, Warren built collaborations with human factor experts in the Institute of Transport Studies to investigate the effects of automated systems on driving ability.
After a move to Manchester Metropolitan University, Warren focused on the development of both student education and research, becoming the discipline lead for the mechanical engineering. And, following that position, he moved to De Montfort University and became the head of engineering and sustainable development. And, actually what I would say now is that this is the wealth for me about the importance of being an academic and being able to really define and hone your skills as an academic and a researcher. So, from that move with the rich history in engineering, he moved to be head of Leicester Media School, deputy dean of the faculty of technology and finally the PVC dean of the college of business and law.
In 2017, I am incredibly delighted that Professor Manning joined the University of Derby as PVC dean of The College of Engineering and Technology, coming back to his roots and I think actually, the experiences that he gained from those previous roles have really proved fruitful for the University of Derby. With his position now as provost of innovation research since 2019. What I believe that those roles have meant is that we now have a strong and passionate leader leading innovation research at the University of Derby.
So, that's the beginning and that's the person. But, this evening's lecture for me is very interesting and something probably that I don't know a lot about, but what it does do is look at the evolution of modern vehicles and how innovation research communities have responded to the various imperatives to improve performance. It will discuss the future of motoring and how ready the industry is responding or able to respond to the future demands for motoring, whilst highlighting key strategic priorities for Derby and particularly, how we can support our students through this work in their research activity.
For over 50 years, researchers, scientists and engineers have found new ways of improving the dynamic performance of vehicles to meet customer needs. Safety, comfort, handling, emissions, fuel usage and carbon consumption continues to drive significant changes. And, for many of us sitting around, that changes the way that we actually will consider how we use mobility in the future.
One of the major drivers for the innovation and research community is to develop generations of active systems for anti-lock braking systems for autonomous vehicles, and for many of us again, that was a major breakthrough and that was actually something that probably made us much safer on our roads. Warren's lecture will explore the context and reasoning behind these changes and how research communities have tackled conflicting priorities in consumer and policy demands. It will explain and discuss the major technologies that have shaped performance across the passenger and motorsport areas, using his own personal experiences and contributions in this field of study.
The lecture will draw on personal experiences to discuss how the outcomes of this work go beyond the research and business of the community to impact the student body, both in and outside the curriculum. The presentation will proceed to explore the future of this sector, particularly the need to decarbonize and the challenges facing self-driving vehicles. It is with my greatest pleasure that I present Professor Warren Manning for his inaugural lecture.
The screen changes and Kath Mitchell has disappeared off the screen. The next screen is split into two. On the right, is Professor Warren Manning, a man with white hair and a white beard. Wearing glasses and a black suit jacket, white and blue striped shirt and a red tie. On the left, the first slide of the presentation is shown. The text on the slide reads ‘Professor Warren Manning Inaugural Lecture’. The background image is of the inside of a car radiator fan, with a red inner piece, going into a black fan. The University of Derby three hills logo in white is placed in the top left corner, and text reading ‘derby.ac.uk’ in white is in the bottom right corner.
[Warren] Many thanks, Kath. And, I will probably mention Dennis Lavelle who was quite an important part of my development when we go through the presentation. Could we move to the next slide, please?
The slide changes. A white background appears with black text reading ‘Out of Control’. The University of Derby 3 hills logo is in black and in the top right corner. This will appear on all of the following slides, along with black text in the bottom right corner, reading ‘derby.ac.uk’. The live video of Warren Manning is still being shown on the right side of the screen and will continue to do so throughout the presentation.
[Warren] I just want to thank everyone first of all for giving their time up this evening to attend this lecture. We've got people from within the university, people I know I’ve invited externally from the university, some of them who know a lot more about this subject than I do so I’ve got a careful filter on the questions as they come in tonight. But, the major subject of my talk really is around control, you know I’m a control engineer I’m not a control freak. But, I am particularly interested in how we've developed since the mid-90s; the vehicle improved its behaviour, improved its safety, improved its comfort and handling but we're now moving to bigger things such as improving the environment that vehicle works in and how it can benefit society as a whole, particularly through decarbonisation. But I’ll probably be pointing out throughout the lecture that a lot of those advances have taken the driver out of the loop and some of the future that we need to engage with, whether that's through different forms of mobility or engagement with the motor car itself, will require further sacrifices in some respects from some of us who are pretty strongly engaged in the driving culture.
But, before I go on to my talk I just want to give some thanks out for those people who've helped me get to the position I’m in today. So, if we could go to the next slide, please?
The next slide has a white background and black text that reads:
Dedications and Thanks
Mum and Dad
Mandy, 5 kids, and 6 Grandkids for putting up with me and giving me the opportunity to have a career
Professor Martin Levesley at Leeds for my first Post-Doctoral position
All the students and PhD students I’ve worked with who made undertaking research and “designing” race cars an interesting experience
Professor Dave Crolla for the inspiration and confidence to see a level playing field, know “carthorses” from “racehorses” and find the fun in mundane
The University of Derby and Professor Kathryn Mitchell CBE for both the support and opportunity
[Warren] So, clearly my family have been a huge support to me. My mum and dad probably didn't have the benefits of education that I did, I was the first in my family to go to university. When I said I was going to become an engineer most of my family thought I’d be fixing the car and the engine part of the car you know? It was a very new step for me and actually going to university itself was a huge experience, I didn't have a clue what a first-class or 2-1 or 2-2 degree meant until I was in my final year at university. But I am going to ... although I haven't put Dennis Lavelle on this, Dennis Lavelle was an important part in my career. He taught me A level maths, after probably a bit of a dressing down over something that we did he brought me into his office and just said "You know you you're going to get three straight A's? That's what your capabilities are." And, although I didn't live up to it and was far from it, that seed was planted and actually it says something about learning, that there's a non-linear aspect and I probably really excelled at my learning, not at college because it wasn't the right time for me, but at university. And, he showed the benefits university can have in terms of social mobility. But also, in terms of my wife Mandy, five kids, six grandkids for putting up with me moving around the country, going to different jobs and giving me the opportunity to have a career because I couldn't do what I do today without the support. Professor Martin Leversly at Leeds gave me my very first postdoctoral position. I have to say my PhD was helpful, but it wasn't the greatest experience. I didn't think that I made the progress I should have done and I really stepped into research when Martin took me under his wing and gave me a lab job to do which involved me spinning the router very quickly at 200 hertz with a gas mask on and a tape recorder, and it was an interesting step into the world of research, but from there I was able to start a career in research at the University of Leeds, where much of my work took place.
A huge thank you to all the students and PhD students I’ve worked with, I tended to do my research in collaboration with PhD students and most of my publications will have the PhD students as the first author. And that was an incredible sort of cultural research experience in terms of working with them, seeing them develop, seeing them go on to fantastic jobs, even the ones who interestingly were poached halfway through their PhD's and went to work particularly in the form of one industry and didn't quite make the PhD, but probably went further and faster than some of the others. So, the whole culture of research I was able to engage with for a good 10-15 years at Leeds. Which I think was a great foundation for me and for what I could do in the future.
But the person who was probably the tipping point for me in my career was Dave and Dave very much took someone who understood research but showed them that the world out there in terms of research can be a level playing field if you know how to play the game. It showed me very much the difference between cart horses and racehorses and life and those of you who know Dave will know exactly what I mean by that. But I think we have more fun in research than most other research groups I’ve ever come across and it just made the whole culture of research - which was high performance but usually enjoyable at the same time. I think some of my colleagues who I know are probably here today who worked with Dave as well would all say that he was the tipping point in their career, in taking them on to excellent things and I think for all of us who've been successful in research, people like that are key, but also the University of Derby who have given me a massive opportunity here in seeing the work we've done in college, in the past as a PVC dean and giving me the opportunity to lead the innovation and research agenda and particularly Kath, who I can't thank enough for her introduction but also the opportunity and for taking me on board at Derby and actually supporting me and taking risks along the way in getting me to where I am today. Next slide, please.
The next slide has a white background and black text that reads:
Main research themes
Automatic vehicle control requires an integrated approach to traction, steering, suspension, and braking systems
Driving systems need to adapt to the environment to deal with a wide range of weather and road conditions
Understanding the interface between the driver and the vehicle is critical to the implementation of automated driving
[Warren] So, there are three broad themes that came out of a piece of work that really started with the work that I did with Martin Leversley, Dave Crowler and then Andy Plummer at the time at Leeds. There was a paradigm about how we control vehicles, it looked to set a road map for how you might start looking at the safety of vehicles, the comfort of vehicles, the handling behaviour. The kind of classic vehicle design parameters that we've seen evolve all the way from the 50s to the present day. But really with an eye on the future and potentially using those systems that can evolve in that way and that we can create a market for to then become the systems that drive cars of the future in an automated way. So, systems for steering, for suspension, for braking and for driveline that’ll improve comfort, handling and safety. They are all critical in their own way to moving towards self-driving cars, connected autonomous vehicles and the paradigms that we put in place under something called "intelligent vehicle motion control," were really important because they allowed us to step through that road map stage by stage and what we'll talk about today are some of the first steps, how we moved from vehicle control to look at the wider environment and then studying the impact of moving to autonomous vehicles on the driver and the vehicle system as a whole. Next slide, please.
The next slide has a white background. One-half of the slide is covered by an angled image of the University of Derby’s Friar Gate Campus and the sky. Black text on the left side of the slide reads ‘The vehicle and the driver. DA Crolla’
[Warren] So, the vehicle and driver. Those of you who have worked with me in the past. Dave and the team will recognize some of the greatest hits probably that are coming out through these slides in terms of some of the ways that we describe vehicle and driver behaviour. The key point is that it's one system; the vehicle is the bit that we can engineer, and the driver is the bit that's incredibly uncertain in that system, although very clever and very adaptable but also quite challenging too. So, can we go to the next slide, please?
The next slide has a white background with a 2x2 blue table in the middle. The black title text at the top reads ‘Driver – Vehicle properties’.
The first column in the table reads:
consistent - overall road, weather conditions
predictable - to establish confidence in its response
conventional - we have built-in expectation levels
controllable! - need to retain steerability at all times
The second column in the table reads:
sophisticated - immense capability for processing information and exerting accurate control
lazy! - prefer minimum effort
adaptive - if necessary
prefer “responsive” system - e.g., contrast control of a car and a boat
[Warren] So, if you look at a comparison between the two, what we need from the vehicle, and this applies as we move through the transition on powertrains to electric vehicles and the move to autonomous driving, we need consistency and that's difficult because, you will know, most of you are drivers so you know the conditions in which we drive and change all the time. It needs to be predictable, so we've got confidence that when we get behind the wheel the vehicle responds in the way we expect it to be. Strangely, vehicles haven't changed significantly, they tend to have a wheel, two or three pedals and four wheels to move you from A to B and that conventionality is really difficult to change even small changes in the steering system can have a big effect. Most vehicles when they're approved tend to be approved by one or a small number of test drivers who absolutely understand what you might call the DNA of a particular automotive company and will feedback pretty quickly if the conventionality isn't where it needs to be, that's changing now but that's been the tradition. It obviously needs to be controllable, and you would expect all vehicles to be stable but there are conditions that we can control and the designer control that makes some vehicles not stable so, pretty straightforward stuff.
Drivers are a different kettle of fish; absolutely sophisticated, huge capability for processing information, unbelievable in terms of being able to control but tend to be quite lazy. Most drivers tend to prefer minimal effort, I think concentration levels can increase in a particular handling manoeuvre if someone wants to do that but on the whole on a journey, as we'll see in some of the work that that concentration piece can change. "Adaptive if necessary" you can move from a high grip scenario to a very low grip scenario and drivers will respond and they will have had the training, but drivers do prefer quite a responsive system and interestingly a lot of the other vehicle dynamic developments that took place in the 90s and 2000s and are probably still ongoing now are about providing vehicles that feel responsive, that control more like a sports car than trying to steer a boat around the corner, so we'll pick up some of that in the presentation. So, next slide, please.
The slide changes to a white background with a black title that reads ‘The vehicle dynamics problem’. Three images are spread across the middle of the page. Image one is of a red car with the title ‘Ride’ in black below the image. The words ‘Bounce’ ‘Pitch’ and ‘Roll’ are labelled off the image, as well as text that reads ‘road disturbance inputs’.
The second image is of a grey box with a yellow border and 4 small black rectangles inside. The title below reads ‘Handling’. Blue arrows pointing off the image are labelled ‘Forward motion’, ‘Rotation’ and ‘Side motion’.
The third image is of a pink vertical rectangle with a small red rectangle inside, representing the side profile of a car wheel. The title below reads ‘Tyre’. Three black arrows pointing up, down and to the right are labelled, from the top, ‘Traction force’, ‘Cornering force’ and ‘Braking force’.
[Warren] So, just to identify the problem. There are a few key motions that get us from A to B. If you boil the motion problem down of motoring to a specific small number of things, then really you need to generate forces at the tyres that move you from one position to the next in a controlled way.
The key issue is, these motions are coupled so we look on the left ride is incredibly important, you want to make sure you're comfortable and as we drive down the road the vehicle is constantly pitching, rolling, bouncing up and down and it's disturbed and then when you go into a corner those motions are exacerbated, they become bigger, particularly the roll behaviour of the vehicle and as that happens the handling of the vehicle changes.
So, if you look at the handling piece what we're trying to achieve is maintaining the forward motion, keeping the rotation at the correct levels and keeping what we call the side slip behaviour of the vehicle correct too and all that happens between the tyre and the road. What would be wonderful is if the tyre was what we call a linear system, so if we increase acceleration then the grip increases proportionately, but it's not. The tyre is incredibly complex, it's non-linear in its behaviour, tyre forces saturate as you'll know if you brake hard and the vehicle goes into a skid but also the vehicle riding handling problem is a coupled problem. So, as soon as we corner then the vertical loads change and that contact patch you see in dark red in the centre of that tyre shifts and moves around, so it's hugely difficult and it's usually difficult because of the relationship between the tyre and the vertical forces and other forces acting.
Predominantly what we're trying to do in the handling problem, particularly the control systems, is to vector those tyre forces to have the right cornering forces to pull the sideways, the right traction forces to take us forwards and the right braking forces where necessary. We can use all three to steer, decelerate, accelerate and move the vehicle sideways. And the principle of the control problem that we're going to look at how we use active control systems to control that behaviour at the tyre road interface. Next slide, please.
The next slide has a white background with a black title that reads ‘The driver dynamics problem’. There is an image on the left side of the slide of a man in a striped shirt and blue jeans, riding a bike. On the right side, the text reads:
Driver Dynamic Characteristics:
Response time delay
Preview control: looking ahead
[Warren] So, yes that is me on the bike unfortunately because of course that's the vehicle dynamics problem but there's a driver dynamic problem too. It's not a car, it's a bike but actually, most of the models that we use in vehicle dynamics are bicycle models and I think that particular driver is understeering quite significantly on the track there at Silverstone. Not sure how we got a bike on the track, but we did.
So, the driver is different, but we can model the driver mathematically and this is done more than you think, there's a huge piece of work done particularly in the formula one industry looking at how we simulate the vehicle and the driver on the track and how you can optimize track performance by looking at particular vehicle setups. In simple terms, drivers have a response time delay which, strangely, various different tests will tell you different things but can be anywhere between 0.5 seconds and three seconds which is quite scary. 0.5 Seconds at 70 miles per hour is about 16 meters in terms of the distance that you will travel before you put any reaction in place. Three seconds it's scarier than that, it's about 93 meters in total and people have done quite significant tests at places like Mira et cetera, understanding what that response time delay looks like. There is a correlation between age and response time and that's one of the challenges in terms of mobility for the future in keeping an ageing generation mobile when we know that some of these things start to drop off. There's a neuromuscular delay, so as soon as you decide to take action it will take time for the steering wheel to move under your grip or for you to press the pedals so all those things are difficulties in terms of driving but what drivers do to compensate is use something called look ahead or preview control and what we found in some of the studies that we undertook that we use in the driving simulator that we'll look at later, is that you can improve the control performance by thinking ahead into the bend, for instance, understanding the amount of deviation that you predict is going to take place into the future and then taking action on that with your control action at the current time.
This is used significantly again in things like lap time simulators to try and improve the performance of vehicles going around the lap. One of the trends that you pick up is if you're in a... let's use the Buick as an example because it's a classic example of what people are used to, an old American car that handles like a boat, then you probably need to look a long way down the road so the relationship between the preview control and the dynamics of the vehicle is quite strong. If you're in a Ferrari Monza, then your look ahead distance can be much shorter. So, something to try for drivers in the future, next time you go around the bend think ahead see how much it improves the driving performance. So, could we go to the next slide, please?
The slide changes to a white background with a black title that reads ‘Driving and vehicle: one system’. In the middle of the slide is a diagram that shows the ‘desired path’ arrow pointing towards a yellow box labelled ‘Driver dynamics’. An arrow from this box points to the other yellow box labelled ‘vehicle dynamics’, and an arrow comes out of this box, pointing right, labelled ‘actual path’. Another arrow is pointing down towards the second box and is labelled ‘disturbances’. From the ‘actual path’ arrow, a ‘feedback information’ arrow is drawn to the ‘driver dynamics’ box.
[Warren] So, we have to look at both pieces and the driver dynamics and the vehicle dynamics work together to get us from the desired path that we want to move through to the actual path of the vehicle and for much of the evolution of the car we've focused on engineering the vehicle dynamics. You can go back and you can look at key cars through history and think about how they have or haven't improved that behaviour, sports cars, in particular, will have a massive improvement on the vehicle dynamics by just setting the weight distribution correctly, picking the right tyres and looking at the suspension so we can compensate for some of these aspects. What we probably started to do, and I would argue that some of this came into play with simple things such as anti-lock braking systems which I think have been around now probably for the best part of 30 or 40 years. We started to think about well actually, can we compensate for some of the driver behaviour? Do we want to leave the whole of the safety of the vehicle, the whole of the handling of the vehicle and in the future the whole performance in terms of what comes out of the tailpipe or other systems, the energy demands? Do we leave those all in the hands of the driver which has been historical, or do we start to interfere and intervene a little bit more with the driver dynamics? And that was very much the principle behind the intelligent vehicle motion control, it led to a huge amount of activity in the UK, both within businesses but also in the research fields during the period that we undertook this research. So, next slide, please.
The next slide has a white background. One-half of the slide is covered by an angled image of the University of Derby’s Friar Gate Campus and the sky. Black text on the left side of the slide reads:
Integrated Vehicle Control:
[Warren] So, we're going to look a little bit at the vehicle control work that we undertook; I’ll draw some examples out from cars that you may have driven, or you may currently be driving in terms of some of the challenges that we have in putting those controllers in place. So, next slide, please.
The slide changes to a white background with a black title at the top that reads ‘Active vehicle control’. There are 4 images of different cars in each corner of the slide. In the middle is a diagram that reads ‘Desired handling, stability and comfort’ with an arrow pointing towards a white block with black borders and text. The text in the block reads ‘Active steering’, ‘active braking’, ‘active driveline’ and ‘active suspension’. 4 double-sided arrows point to both this block and another white block that is labelled ‘vehicle’. An arrow comes out of this block with text that reads ‘actual handling, stability and comfort’.
[Warren] So, if you look there's two cars there on the left, one's the Citroen BX I think and the other one's a Ferrari Monza. Different cars historically have been developed for different reasons. Citroen cars in France tended to be developed from a comfort perspective, to look at more ride behaviour because they were designed to go across rutted tracks and very difficult road surfaces more than handling behaviour and anyone who's driven certain cars - I once had the joy of a Citroen AX, I’m six foot eight so you can imagine me in a citron AX - You go around the corner, and it rolls like anything but the ride is absolutely fantastic. And then on the flip side sports cars have fantastic handling behaviour but the ride behaviour isn't particularly good, you can probably feel every bump and there's certain surfaces you definitely wouldn't drive over.
So, active vehicle controls came in to really start off to address some of these compromises that we had between ride and handling as well as to address safety concerns, so we probably saw the first systems coming to production cars. They're now mandatory, as we know active braking systems are a must on all new vehicles and interestingly stability systems which correct in case, you're in a corner manoeuvre. So, they came in first and they took away some of the driver control which on the whole is positive because we'll see in a minute when we look at grip that it's very difficult to hit peak braking force and to control slip in a manoeuvre so that kind of intervention was positive. Most of us rarely experience it but when you do experience it and you feel the pulse on your foot it is a very odd experience knowing that the car is taking over in terms of what you're doing. So, that was the first step and those systems allowed us because they were interfering with the brakes effectively and doing that independently of the driver. We then could do that independently at each wheel so we could break one wheel or another wheel and that allowed us to provide some steering control of the vehicle, so the active braking moved from a stability piece that was about making sure you could steer while you brake and perhaps reducing stopping times to - if you're in what we call the classic moose test (I’ll talk about the vehicle that's rolling there at the moment) or the elk test as it's called - You'll be driving down the country line and the moose walks out and you have to steer quickly to avoid it, that is an actual manoeuvre. Then the braking system comes in to stabilize that steering behaviour and we'll look at how significantly we can improve that. So, all these things are in vehicles now, they work, which means that if you have to make a sudden manoeuvre with the steering wheel it is going to put that correction in place.
So, that was really positive but then we started to understand that active steering systems started to come onto vehicles to improve feel, active driveline systems such as on some of the Subaru cars to improve the stability but without interfering with the speed, an active suspension system so the vehicle on the top right of your screen is a Land Rover Discovery, it's got an ace based system on it which means it's got traction control which means you can go around the corner pretty quickly and it will help you and support you in cornering and it will avoid you skidding as well when you accelerate quickly, but it's also got an active suspension in there broadly because it's got enough space to put one in. And if you have ever been in one of those cars, and you can literally go around a roundabout in 40 miles an hour and it feels like you're driving a go-kart and the handling is absolutely amazing because you've got the active suspension keeping it level and you've got the traction control support in the cornering enhancements.
But what we started to find was that these systems were interfering with each other, so there are certain actions that you might take to improve what we call the understeer oversteer of the vehicle and the driveline would do one thing such as perhaps increase the traction at a particular wheel and the suspension system would do another thing with the same objective in mind, but it would actually start to lift the wheel where you were putting the increase in traction because there was no coordination between the two. So, this was a problem that they were finding so TRW brought this problem to us because they were developing both systems, Land Rover and Rover Group at the time were interested in the problem, BMW when they assimilated Land Rover, we were there at that time equally we're interested in the problem, was a big challenge in the sector. So, we changed the paradigm or the model a bit. So, could we go on to the next slide, please?
The slide changes to a white background with a black title at the top that reads ‘Integrated vehicle control’. There is a black line diagram in the middle of the page that is the same as on the previous slide, just with an extra block between the text ‘desired handling, stability and comfort’ and the ‘active’ block. The new block is labelled ‘master control’.
[Warren] So, what we did was brought in some quite advanced control techniques to take that desired steering angle that you put into the car to interpret that, to look at the stability of the vehicle and to have a master controller. We use something called sliding mode control which for the mathematicians minimizes something called a layer function to get you from A to B and what that produced was four generic wheel talks for each of the wheels to say, "this is what we need the wheel to be doing," forging the two generic steer angles or one depending on whether you had front and active rear-wheel steering and four generic tyre forces for the suspension. But it produced those forces to ensure that you were getting the optimal performance out of the vehicle that you wanted and then we could manage those across the active steering systems, active braking, drive iron and suspension systems to improve performance. Can we go to the next slide, please?
The next slide has a white background and has a black title at the top that reads ‘High “g” handling test’. In the middle of the page is a black line graph. The line on the graph is a wiggly line that gets larger in the middle to end, between the times 5-15 seconds.
[Warren] So, we looked at the effects of both of these systems, the combined control was where we were basically switching in and out the systems in a very rough and ready control manner and the integrated control was where we're looking at the overall vehicle performance.
Now, interestingly the integrated control in this graph is slightly better, I should probably explain what's happening here, we've got a driver going along at about 100 kilometres per hour and then gradually increasing the steering angle putting what we call a sine wave into moving it to the left and to the right back to the left, making it bigger and bigger every time and taking it to a point where it's quite a high and intensive manoeuvre. So, this is getting sort of in the region of 1g lateral acceleration, if any of you've gone past 0.4 g in a car that would be quite surprising, 1g is quite high unless you're in the motor industry and you do some racing so, quite a severe manoeuvre. The integrated control is better which is good so we showed that we could reduce the amount of lateral acceleration, I wouldn't say significantly but appreciably, but the integrated control was looking at the overall control of the vehicle, it wanted to keep that lateral acceleration correct, but if we look at the next slide
The next slide has a white background and a black title at the top of the page that reads ‘Effect on vehicle speed’. There is a black line graph that shows vehicle forward speed in km/h from 40 – 110, against time in seconds from 0 – 15. One line shows the vehicle is combined control and the other shows vehicle with integrated control. Both lines decline from the speed of 100 km/h at 0 seconds. ‘Vehicle with combine control’ ended at 50km/h at 15 seconds, and ‘vehicle with integrated control’ ended at 60km/h at 15 seconds.
[Warren] this is where we started to get some interest because what the combined control was doing was probably overusing the braking systems, so you can see the solid line there on the combined control is decelerating the vehicle at the same time from 100 kilometres per hour down to 50 kilometres per hour and that is a problem for feel and people could perceive this, depending on the vehicle, but some were particularly poor at it, the Mercedes A-class that we showed you earlier was a real problem because they had to over-engineer the stability control on that after the test driver rolled it in front of a load of photographers. But it is a problem because you don't want to make a manoeuvre but necessarily slow down too significantly at the same time, particularly if you've got traffic behind you or you've decided to overtake a long row of traffic and jump in right at the last minute, and this can happen. With the integrated control, because it was combined in the driveline system, the braking system and the steering system, it was putting in the best efforts of the three systems in a coordinated way and it was keeping the vehicle forward speed at a higher level. This was jack, he now works for Bosch, we looked at just about every single combination of systems and we identified that there were quite significant benefits to adding this layer of control on top. It didn't mean that we had to implement new active systems because they're quite expensive and active steering or braking or driving or suspension systems are also expensive, it just meant we were interfering with the control paradigm, we were bringing in algorithms into the existing engine management and existing body management systems, which could be done at a relatively low cost. So, it was a real positive in terms of the work that we were doing. I think some of the challenges to bringing this to implementation, we did this work with Rover group for instance and some of the earlier work we did was with TRW, and it seemed that every time we work with a company they were assimilated by another company from around the world. But certainly, we were able to see the impact of some of this work going on to real cars, particularly the ace work that we did very early with Land Rover. So, next slide, please.
The next slide has a white background with a black title at the top that reads ‘The need for high-level supervision’. The black graph from slide 12 appears again, with an added block at the top labelled ‘Intelligent supervisor’. Two arrows pointing in both directions come off this block vertically towards the ‘master control’ block and the ‘active steering’ block.
[Warren] So we've made quite a lot of progress, by putting the master control in place and using some advanced nonlinear control techniques, we've shown that we can make some significant differences. But, the driving problem is it’s great when the road is dry, when you don't have side winds, when some of the vehicle parameters such as tyre pressures (and I’m sure we're all very good at keeping our tyres up at the right pressure) all these things change on the car and of course the master control unit and all the active systems don't necessarily understand these changes have taken place and you either have to create a robust control scheme which means that it operates under a wide range of values (and I’ll show you in a minute why that's really difficult to do) or an adaptive control system that measures the changes in the environment and updates the controller based on where you're operating at the moment. So, the intelligence supervisor was brought in and one particular piece of work that we did with Matt Wilkin was looking at how we would estimate... well it's a bit of a holy grail actually in vehicle dynamics which is the road surface coefficient of friction, it's the grip between the tyre and the road, it's incredibly important, it keeps most of you very safe when you're driving the car on a day-to-day basis. So, can we go to the next slide?
The slide changes to a white background with a black title at the top that reads ‘Adapting to the environment. Below, the left half of the slide contains a black line graph that shows the fiction coefficient between 0 – 1, against tire slip between 0 – 1. There are 4 lines on it that are labelled (from the bottom) ‘Ice’, ‘Snow’, ‘Wet asphalt’ and ‘Dry asphalt’. The right half of the slide reads:
“Grip” varies significantly depending on road surface and tyre behaviour
In wet and dry conditions peak grip needs careful control
Active control schemes underperform without access to an estimate of grip
Tyre quality remains in the hands of the driver
[Warren] So, these are very typical of what we call friction coefficient curves, they tell you broadly how much grip you're going to get depending on the conditions. They change actually, I would say if you look these up they change depending on the tyre fundamentally, the material the tyres are made of, the size of it and the pressures they change it, they change if you don't keep your tyre pressures at the right point so that's my public health warning for all of you. But the key point on these graphs is that they peak, particularly on dry asphalt and wet asphalt. The graphs peak and then drop down again so if you look at the far right of the dry asphalt or the wet asphalt graph, if you're there, you've locked the wheels and your car is sliding and you may think "well, I’ve done everything I can, I’m pushing the brake as hard as it possibly can" but it's no good because you're probably 20 per cent below the optimum level of grip for the vehicle.
Very good drivers, test drivers, formula one drivers etc can feel that level of grip, not all of them but some can, but your average driver would be struggling to feel that level of grip definitely in wet asphalt because it happens more quickly than on dry asphalt. So, things like anti-lock braking systems for instance came in because drivers couldn't reach the optimum but the system could measure this and it could keep us at the peak point, but the issue for the master control and the integrated control that we were looking at was...we were moving to a position where we were taking more and more of the driving behaviour out of the driver’s hands and that system didn't know if it was raining, didn't know if it was wet, it doesn't always have to be raining for you to have wet conditions, for instance, it didn't understand what the behaviour would look like.
So, Matt did a piece of work and interestingly used the former student cars that they've been working on at Leeds. Matt did a piece of work to try and estimate using extended Kalman filters what the forces were acting in the tyre, in the longitudinal lateral directions to see if we could estimate the grip. so, if we look at the next slide, please.
The next slide is a white background with a black title at the top that reads ‘Estimating tyre forces’. Below is a line graph that shows rear lateral force (kN) against time in seconds between 130 – 150. There are two lines, one black and one red, overlapping each other. Starting at 130 seconds, the line decreases until 140 seconds and then increases from here to 150 seconds.
[Warren] So, you can look at Wilkin, Allen and I think Matt's thesis is online as well but most of the work that we did we looked at experimental data that came off the car and then we looked at what the extended Kalman filter does and the extended Kalman filter basically takes the data from the vehicle that we can measure, things like the lateral acceleration, the roll behaviour, as much as you can get, and it pumps it into an algorithm that then looks at that data, estimates it, looks at other data but then produces out of that whole process an estimate of the things that you can't measure. There are some conditions that need to be in place that you check that's what we call an observable system, but this was really positive. It didn't capture everything, the high dynamics it doesn't tend to pick up but that's not a bad thing because filtering the high dynamic behaviour out is a positive because you don't want your control system responding to such high frequencies anyway.
So, we took it to the next step and proved that you can move to systems that improve safety, improve handling, improve comfort and take you on that step to automated driving but also you can make them adaptable to some of the key road conditions and the key external conditions that the vehicle is operating. The other classic one that you look at is speed interestingly; speed's very difficult to measure which sound's bonkers because everyone's got speedometers but the instantaneous speed, if you imagine being in a skid situation it would probably tell you (because the gearbox is not necessarily moving, and the wheels aren't moving) that your vehicle isn't moving but you know you are moving because you're probably still travelling at 30 miles an hour. So, measuring speed while the wheel is slipping and spinning and skidding is very very difficult to do and you either have to get the friction force or the speed of the vehicle. So, we got to a good place in terms of that piece of work. So, next slide, please.
The next slide has a white background. One-half of the slide is covered by an angled image of the University of Derby’s Friar Gate Campus and the sky. Black text on the left side of the slide reads ‘Automated driving’.
[Warren] So, we'd really establish some of the foundations. When we first started all that work, the IBMC project, interestingly, we did press interviews and I think I went on Piccadilly Key 103 which for those of you know Manchester it is a radio station, strangely I wasn't in Manchester, but they wanted to interview and everyone was into this idea of snooze control and self-driving vehicles and being able to sleep on your way to work and that that was probably 21 or 22 years ago. The California path program in 1984 identified that California could see they had a massive problem coming in terms of traffic congestion, in terms of emissions, fuel usage, safety etc. It established platooning self-driving vehicles in 1984. The technology has never really been a problem, the problem really is in the human factors and the adoption of it and making it work. So, we did two projects, the first which is probably the most recent actually so we can move to the next slide.
The slide changes to a white background with a black title at the top that reads ‘Range extended engine management system’. Below is a diagram with four rows. In the first row, 8 orange circles are across the diagram, with 3 arrows pointing south-west on all of them. Different measurements are also shown in orange text, in measurements of W/m2. In the second row, 6 blue raindrops and 2 snowflakes are spread out across the row. These show blue percentage measurements of rainfall. The third row contains 8 sets of 3 green arrows. The first 5 sets are pointing east, and the remaining 3 are pointing west. These have measurements of kilometres per hour in green text to the right. Below this, there is a diagram of a black car travelling along a road of different angled variations, such as flat, uphill and downhill.
[Warren] It's some work that we started at DMU and we're picking back up again now with some of the work we've been doing with Ashika now at Leicester, but looking at extending... well at the time was looking at extending the range of vehicles, I think if you go out and buy an electric vehicle and it's in the Tesla three upwards range, the range is probably not a massive issue for you but certainly when we did some work at DMU on a Nissan leaf vehicle that we got in, it was a huge problem. One of our colleagues commuted quite a short distance from Oakham to Leicester every day and it would absolutely depend on the conditions in terms of the rain, how warm it was and the wind speed but also the profile of the road, going from certain areas as to whether he was confident that he could get home or not due to what we call range anxiety.
So, we started to look at if we could optimise over the trip, looking at the data now that was external from the vehicle to coming to the intelligence supervisor. Because you can measure some of them, you can understand trip time, what some of these parameters might look like it's not perfect data but it's better than nothing, but what you can certainly do is understand what the profile of the hills look like. You get advantages on downhills where you might use things like regen breaking and disadvantages on uphill’s, and what we said was (a bit like the look-ahead control from the previous work) if we can look ahead and we understand what the trip looks like we should be able to optimize the state of charge in the battery over that period. So, if we go to the next slide.
The slide changes to a white background with the same title as the previous slide. On the left half of the slide is a black line diagram. There is a larger block that contains the sections of a car mechanism, such as the wheels, power control unit, and generator, with arrows showing the flow of how it works. Above this are three blocks that link off. These are labelled: ‘Trip demand prediction’, ‘computational intelligence unit’ and ‘driver behaviour classification’. On the right half if the slide is black text that reads:
Broader energy demand usage
Fuzzy Logic Based CI unit
Driving condition metadata derived from Artificial Neural Networks
[Warren] So, the bottom box there is basically, the range-extended vehicles, it's got an electric motor control, you know, all the things you'd expect and unlike many has a range extender on there but we ought to use an artificial neural network technique so that broadly we could take quite a wide data set of environmental conditions and create what we call metadata, almost token data that might not mean something in terms of, it won't tell you what the temperature is but it means something to the computational intelligence unit, and then we used type 2 advanced logic decision making to then take that data and decide how it would change the state of charge in the battery. This was a different project for us because it was less of a research-based project where we would publish directly to the open journals because this was working directly with a couple of businesses, it was funded through ERDF, out of the transport net project at Loughborough University and actually, most of the work went straight back into the organization for them to use and evaluate further.
So, we're picking that up again now, we're looking at part the issue with this which is that artificial neural networks are okay in the vehicle world broadly until they start to intervene between the driver and the vehicle because there's a trust issue, because they're a bit of a black box. the concepts of explainable AI that we've been working on a little bit with Ashika on the project and we just published a paper on that. If we can get more confidence in that then we can get to a point where we can start to introduce this in earnest a little bit more. So, can we go to the next slide, please?
The next slide has a white background and a black title at the top that reads ‘Driver - Vehicle studies of automated driving’. On the left is an image of a white driving simulator in a warehouse, with a whiteboard to the side of it that reads ‘University of Leeds Driving Simulator’. On the right of the slide, the black text reads:
Advanced Driver Assistance Systems
Adaptive Cruise Control
Lane Keeping Systems
Lead Vehicle Deceleration
Car cutting in
Acceleration in Bend
[Warren] So, those were our steps really to look at automated driving and the advantages of it, how we protect the different systems and coordinate it and again, going back to some of the work we did at Leeds, it was interesting at the time and Leeds was in the top four in the country in engineering for its research, it was outstanding but the concept of working internally... beyond bioengineering which was amazing but working internally with other colleagues wasn't necessarily strong and I had a very strange look from the then guy who looked after PhDs when I said "well can we do a bit of work with the institute of transport studies?" to look at the human factors as well as the objective dynamic factors of bringing in these advanced driver assistance systems. And we got support from a guy called Robin Auckland to do a joint PhD, looking at the interface between the two and the timing was perfect because we then won a grant to bring in this advanced driving simulator. So, that pod that you see on the screen has got most of an x-type Jaguar inside it to make it a real experience for the driver, it also moves 10 meters forwards and 10 meters sideways to introduce cues (what we call them) but basically, it tricks the driver into thinking that the car is rolling, the car is pitching or the car is generating lateral acceleration to the side, so we can get the dynamic behaviour of the car for the first time really in some of these simulators as close as possible to the real thing and that allowed us to then investigate two key technologies at the time, one was adaptive cruise control which basically means that it keeps a fixed time distance or physical distance from the vehicle in front and then lane-keeping system on motorways which allowed drivers to...there was a safety piece first to make sure that you don't deviate from the lane when you shouldn't but actually underpin self-driving vehicles and being able to steer on the motorway going forward because these systems would actively provide some steering control.
And we were able in this to look at some of the key problems so the vehicle decelerations obviously if the car in front of you slows down or a car cuts in but one of the challenges around adaptive cruise control systems is what we call acceleration in bend. So, if you think about it, the car you're in is connected to the car in front, it's following it, it tends to use a lidar system to check what it is. If that car goes around the bend, it believes the car in front is accelerating and there were occasions where it would accelerate your car to follow it as you're moving into the bend so in certain scenarios in all of these it required an interface between the actual automated system being in control and then the driver coming in to correct the automated system when the control may not have been working effectively. We wanted to understand some of the human factors around that, what it felt like for the driver but also looking at some of the objective measures and whether the performance was strong or not. Next slide, please.
The next slide has a white background with a black title at the top that reads ‘effects of automated systems on driving’. On the right of the slide is an image of a 4-lane road with a white car driving on the outer lane. Orange and white traffic cones are lines along the outer lane going into the greenery. On the left side, the black text reads:
Where evasive action was required, the ACC improved the driver’s reaction times making collisions less likely
Drivers were able to counteract any possible lane deviation caused by ACC acceleration whilst cornering
The driver’s performance was reduced when the LKS ‘dropped out’ due to confusing lane markings when compared to an unassisted driving
Both systems were acceptable to the driver, ACC more so than LKS
The driver experienced increased comfort when driving with ADAS
The driver experienced reduced enjoyment when driving with ADAS
[Warren] So, on the right it gives you one of the examples, that's an example of the simulator. I helped to give Robin some feedback on this when he was undertaking the work and it was a really weird experience. It's immersive sitting in the cockpit of an actual car looking out onto a road that looks like a real road and we ran one test and then because it took time to turn the simulator off and back on again to restart the test, Robin asked me just to drive through the central reservation back through the oncoming traffic and start it again and it almost made me feel physically sick, it was that immersive, so it was a really good model of driving. In this particular test, you see out we've got a classic case of narrowing lanes because there's probably some road works going on and a situation where probably the automated driver isn't picking up what a real driver could do and there's a transition between the two controls.
What we found was that broadly the automated system was good and interestingly they were evasive in longitudinal type actions such as bringing the braking in to intervene and things like cutting in, drivers were fine but where lateral manoeuvres were needed such as steering, such as in this case as you see on the screen with the lane dropout it, confused drivers to come in halfway through and provide some support and performance was reduced quite significantly overall when drivers had to intervene with the real system and there were mixed messages coming back from the drivers. We were fortunate to work with the team at the ITS because they understood how to do proper studies involving test groups etc and they were able to identify through their studies that both systems were acceptable. I think they put a questionnaire in place, did all the correct things in terms of working that out. But, with ACC the longitudinal were more accepted than the lane-keeping.
Drivers interestingly had increased comfort, I think they felt that watching the vehicle go through the motions of things they'd have to do themselves, I think they could sense that that was better, but the driving enjoyment was quite significantly reduced. I think there was there was a nervousness about that and there was a feeling of that loss of control. And I just want to pause on that point before we move to the next slide because the central argument of this talk is around who should be in control of the vehicle going forward, is it the driver?
The next slide has a white background. One-half of the slide is covered by an angled image of the University of Derby’s Friar Gate Campus and the sky. Black text on the left side of the slide reads ‘Impact on student education’.
[Warren] Is it the automated systems? And what happens at the interaction between the two? And the reality is that we know that platooning we know that self-driving vehicles and an autonomous approach has a massive improvement on congestion, can significantly reduce emissions and fuel consumption and massively increase safety. The question is the steps that we need to get there and we're going to come on to that at the end of the presentation when we talk a little bit about the future.
But I just wanted to touch briefly on something that became incredibly important to me, particularly when I moved from Leeds as a lecturer to Manchester Metropolitan University, a very different style of university at the time and had the crazy idea of starting an automotive engineering program and introducing formula student racing. The formula I knew a lot about and the latter I knew very little about, it just seemed like designing and building cars to me and that looked pretty straightforward. But that was probably my biggest step towards becoming a dean and what I do now because I started to really understand the impact of student education, the impact of research sorry on student education but no more so than when... you know one of my students first in the family, had really struggled with aspirations about what comes next, and I convinced him to apply for a job (I think was with JLR at the time) and he was convinced that he wasn't going to get it and he got the job and came back and said there were students from Bath, a top engineering university, Cambridge etc. He was the only one interested from a Pels 92, but he walked into the job. He walked into the job because he had the industry-standard knowledge and we picked up that industry-standard knowledge (particularly around vehicle dynamics) through working with those industries in the past and it was a no-brainer to me to teach students things that were advanced, and they were difficult, but you could see they would be transformational too. So, if we can go onto the next slide.
The slide changes to a white background with a black title that reads ‘Formula student racing’. There is an image on the left side of a white racing car being driven around yellow and green cones, in front of a building. On the right, the black text reads:
Focus is on design and the application of theory
Worldwide competition and “exhibition”
Process is as important as performance
[Warren] So quickly, this is one of the Manchester met cars because... I think you know for me this is where the most has happened. Those of you who've lived in Manchester can pick up that we always had 161 on the front of our cars for obvious reasons but it really was a focus on design and the application of theory and it's a worldwide competition, so people come from all over the country to compete and for students, it's like a big exhibition of their capabilities, their knowledge, their learning and what they've done on their programs of study.
I think we're pretty strong actually in the UK but some of the teams that come from Germany, some of the teams that come from America bring some amazing innovation that inspires students to go back for the next year and innovate even further. But I think the crucial thing for me is that what I saw during this journey was students absolutely taking the stuff they pulled off the curriculum and making it happen but also students constantly learning in a very nonlinear way in terms of learning because there's a need... so you would have first-year students doing things that are masters level and PhD and beyond because that was part of their project, they needed to develop a suspension system, they needed to learn how to do things. It drove a completely different type of behaviour that I probably haven't seen anywhere else in terms of inspiring students to get from A to B. The race is important don't get me wrong, but you tend to find when you get to the there's only a small subset of students who actually want to get in the car and race it because of the pressure etc, but the process of getting there for the students is key. Can we go to the next slide?
The next slide has a white background with a black title that reads ‘students in innovation and research’. On the left, an image of 4 students in blue overalls working on a mechanical engine. On the right, the black text reads:
[Warren] I saw, as I said, student knowledge increasing dramatically not alongside a bog-standard curriculum of doing this and that etc. The experience of teamwork was huge, it wasn't always nice, it wasn't always easy, it was a classic "we're working together to get things done." Students come in at seven in the morning and stay until people kick them out effectively. Constantly learning and constantly progressing all the time but also employability... this became a differentiator for job interviews for students, they could walk into a job interview, and they'd say, "well what experience have you got of actually taking your theory into practice?" And building a race car, designing it, having your component on the track, knowing that failure means that the whole team fails is huge for students and I became a bit of a convert to this because some of my colleagues from Leeds started this process and I picked it up at Manchester Met. So, it just told me that you could bring what you've taken from a research environment, you could bring it into the classroom and if you do it in the right way... and in this case a very practical way, it can have a dramatic effect on students and their performance and their future opportunities in life. Next slide, please.
The next slide has a white background. One-half of the slide is covered by an angled image of the University of Derby’s Friar Gate Campus and the sky. Black text on the left side of the slide reads ‘Mobility and motoring?’.
[Warren] So, what comes next then? And I think we are moving to more of a mobility agenda and motoring, and I think that certainly what everybody wants in some respects, but does everybody want it? And are we all ready to change our culture from a motor into mobility? I think this is the question that we have at the moment. So, could we go on to the next slide?
The next slide has a white background with a black title that reads ‘Broader considerations’. On the right, an image of a big crowd of people walking through a street. On the left, black text that reads:
Population growth and diversity
Benefits of the current transportation system, its technology and availability
Geographical imbalance between jobs and population growth
Reduction in Greenhouse Gases
[Warren] There are some broader considerations first, I’m a big fan of what California has been doing on his past projects but if you read it, it's scary, not just the population growth but the diversity of people, particularly an ageing population who need mobility. We did some work with a driving assessment centre in Leeds, and it was an eye-opener really because... Basically, for people of a certain age, if they're in a crash or a collision or any sort of situation then they'll get sent to a test centre they'll undergo some quite rigorous testing, and it will determine whether they'll keep their license or not. That's quite a massive change, particularly for people who are living by themselves or who haven't got access to mobility, removal of the car is a huge life-changing effect. So, mobility has to change for the population that is moving now but another major challenge is that our current transportation system... it has links to it. Jobs, not just getting to jobs but actual jobs in creating the systems that we've got, the whole way it works is that there's a culture, there's probably an economic model that better people than me know about and there's a way of life that works around it and there are huge benefits from the way that we move around now. One of those is that it's accessible to nearly all, you can never say all people have but most people can have a car and it can get them from A to B to live on a day-to-day basis. I think that will be one of the biggest challenges for future mobility systems if we make the transitions that we need to get the reduction in greenhouse gases that we want... and to deal with energy security and other significant challenges. And of course, linked to that is that we've got population growth in areas we've got jobs growth in areas but they're not always in the same places; mobility is crucial for getting people from A to B. So, next slide, please.
The next slide has a white background and a black title at the top that reads ‘Evolving world of mobility’. On the right, a range of images will appear. On the left, black text that reads:
Larger system (e.g., grid) optimisation
Platooning –e.g., freight
[Warren] So, if we just click and bring up the first image. You know the... Sorry, Anna can you just click on the screen? I’m not I’m not seeing the Kinto image. We'll move on... oh, there we go, great.
The first image appears. There is a woman in business attire walking past a car looking into the distance, holding her phone and a coffee cup. In the middle of the image, there is the word ‘Kinto’ large and in white.
[Warren] We'll probably get all five now in a row, this is the problem with putting some animations in. I think what Toyota's doing with mobility is amazing and I think it's brilliant that Chris and Tony Walkington at the University of Brussels are working with Toyota to bring the Kinto platform to the university. It's a game-changer, the whole concept of ownership is being challenged now by looking at leasing models, car-sharing models, pooling hailing models where there's a BMW situation where you can, through an app, hail a car and it will just come past within 250 yards of where you live and at the right time, you'll need to get on it. It's massively being driven by data and AI and we're seeing the ownership of mobility, I think shifting to either a different model or to a different space and maybe it's a must-do because... do you leave the control of the bigger pieces, (such as reduction in greenhouse gas emissions) do you leave that in the hands of the driver driving into individual vehicles or do you bring a more connected platform? Can you shift to a different form of mobility that we'll need for the future that's more inclusive as well? Because driving isn't (as I said for the ageing driving perspective) always inclusive. Can we just click to get the next picture up?
The image on the slide changes to a blue box with a green supply and demand diagram inside, in the shape of weighing scales. Both supply and demand sides have factory icons on the weighing scale. In the middle, is the balancing point, which is labelled above as ‘Frequency stabilization’.
[Warren] And then, there's the larger system for everything that we do. We're all using the internet now and dragging some energy out of the grid, we have to balance the energy that we use against the energy that can be supplied and interestingly electric vehicles can play a very important part in that because they are at a large scale, they're quite a significant energy store so if we can move to a system where data and AI are not just looking at the journey of the vehicle and taking over in some respects but also managing where we need the energy in each of those vehicles you can help to stabilise the grid and bring on more volatile energy supplies into the grid by using the existing store of energy elsewhere to optimise it. Next slide, please.
The image changes to a photograph of 3 lorries driving down a motorway, equal distance apart. The road is marked out by traffic cones. Alongside the road, are greenery areas with many trees.
[Warren] From the California path in the 1980s to the modern-day, the concept of platooning is quite significant. I’ll use a very parochial example of driving down the A1 and every time a lorry pulls out then the whole system tends to stop just because of the nature of the road with lots of hills, particularly on motorway driving but managing vehicles in a train. I’ve heard people refer to the distance between the vehicles as being a virtual towbar almost but managing that automatically can have huge benefits. Congestion quite significantly, because you can control at a master level the elbow behaviour of the traffic. You could interfere and slow down the vehicles, increase the speed of the vehicles etc but also between vehicles too; those trucks, the distance between them has an aerodynamic effect so you can reduce emissions in that way. But also, the way we drive has an impact on how emissions so, safety, congestion and emissions, we know all benefit significantly if we can move on motorways to self-driving vehicles where the platooning happens and basically you let the system take care of itself. For those of us who drive on motorways, you will know that that's quite a significant step-change because people like to have control of their speed and their positioning etc and the real challenge around platooning is how you change from one mode to the next because you can't have a half and half with people dipping into the gaps for instance between those lorries when they want to because it absolutely messes the whole system up. Next slide, please.
The image changes to an image of an electric scooter and an electric bike. They are both white and black and are placed beside a lake, with buildings in the background.
[Warren] And then modal shifts, what we call the first and last mile of mobility so you could imagine that we could improve the train networks we can improve the tram networks, the bus networks, all those things... But how do you get to the bus stop and back from the bus stop? How do you get to work from those points? And we've seen this already haven't we with electric scooters and things, so there are opportunities around modal shifts that are there today. And, what you really need, going back to the data and the AI is a... from getting up in the morning to getting into your seat at work (and we'll come to that in a moment) then you know it's really clear what that journey looks like and what each part of it is and that everything is in place at each step to make it convenient. Then the last one if we click again, please is the whole concept of connectivity in a different way.
The image changes to an image of 2 men from Tokyo with a robot in the middle of them, projecting a screen of a woman in business attire sitting down.
[Warren] For people probably at my age this seems absolutely crazy that you could have an interaction that was once a physical interaction with a robot with a picture of someone on the other side but actually, when they brought this into the Tokyo Olympics in 2020 it was about giving access to the spaces around the Olympic stadium to people who wouldn't normally be able to get that access, to wheelchair users and others. But the concept of mobility certainly (and this is another Toyota innovation that I think is really strong) is about being able to be part of a community but in a very virtual, telepresence sense. I think we probably learned through the pandemic that it wasn't great being outside that wider community but actually it did significantly reduce emissions, reduce fuel usage and had an advantage at the time. So, that's kind of the evolving world. If we go into the next slide...
The next slide has a white background. Along the top of the page are separate images of 4 cars. Along the bottom of the page are the images that were presented in the previous slide. In the middle, black text that reads:
Larger system (e.g., grid) optimisation
Platooning – e.g., freight
[Warren] it's all really positive and you can see across the bottom all those options, from the Kinto to the telepresence, you know being that you're thinking well this has got to be the way that we go forward but we are still incredibly attached to a culture of motoring, and I’ve got a couple more slides but talk a little bit about the latest trends in electric vehicle sales. It probably isn't indicating that the world has said driving a diesel motor car and... particularly I would say cars have got really big, haven't they? They're like big bricks driving down the road they're kind of crazy from the perspective of trying to save fuel. All these cars on the top I’ve not picked out any particular but just some examples. People are attached, people like ownership they like that direct control, that sense of feeling that you get through a steering wheel and knowing that you're in control of getting the car from A to B, the freedom of being off-grid and knowing that you could change your mind in a moment and do something completely different. There's a huge sense of identity that comes with motoring, the feel and the handling but also there's a culture behind it and I think it would be very very difficult for many of us who've grown up owning motor cars to say that it will be easy to move from that model to a different model that we might see on the left and if we move on to the next slide.
The slide changes to a white background with a black title at the top that reads ‘is there a shift in behaviour?’. Below is a 6x9 table that shows new car registrations between the year to date of 2021 and 2020, from Diesel, Petrol, MHEV diesel and more. A red border is placed around the ‘BEV’ row outlining the statistics.
[Warren] This data has a huge caveat with it because of course it's data that came during the pandemic in 2020-2021, we know that new car registrations dropped quite significantly during that time for obvious reasons. There's some light in the tunnel here, battery electric vehicles (BEV), market share in 2021 at the year to date in December we've gone up from 6.6 per cent to 11.6, going up by 5 per cent. But it just said that people going out buying electric vehicles in 2021, only 11 per cent of them could or would go down the electric route. Even though the technology in those vehicles is fantastic, they're great to drive, the acceleration feel is just wonderful (I think it is anyway) you know they're great cars. There's a total cost of ownership issue and I’ll come to that in a second, but we aren't seeing the transitions I think that we would be expecting and if you look at diesel and MHEV is what we call mild hybrid electric which means that there's a little bit of assist but it's not significant on the whole. You know diesel car sales have dropped but they were still 14.2 per cent of the market when diesel is the devil, and petrol car sales have dropped as well which is a good market share, but still, 58 per cent of people went out and bought a petrol car in 2021 and the SMMT who put this together are predicting that by the mid-20s that figure could be as high as in the 40 per cent bracket, but during that period it's probably never been a better time to buy an electric car. You have something called the plug-in car grant of about two thousand pounds I think it is now to help you to take the first step to buying it, you've got serious VAT reductions whether you're a company or an individual buying those vehicles and that's guaranteed probably for the next two or three years.
What the SMMT are saying is that three things probably need to happen if you want to get to a position where perhaps BEV ownership is around 87 per cent or above and that is that those tax breaks need to continue but that has a significant effect fiscally on the treasury because I think that money coming into the treasury from these vehicles is important and the grants need to continue but also the total cost of ownership needs to come down probably by the mid-twenties. So, the total cost of ownership of an electric vehicle... (which is compensated by the fact that it tends to be cheaper to get electricity than petrol) but the purchase price and the cost of ownership should probably be equivalent to a petrol vehicle by about 2025. And those are the conditions really to get us to a point where we can probably get to where we need to be in terms of our behaviour. The only other shift of course is that we just accept it's a must and we take the hit as a consumer, or legislation pushes us to a position where we probably don't have a choice and there's no sign of either of those two things significantly changing. So, we're just going to go to the last slide now...
The next slide has a white background with black text that reads:
Active systems have improved safety, ride and handling
Significant progress has been made on adapting the system to its surroundings
Connected and autonomous vehicles have demonstrated effectiveness, in field tests, since the 1980s
Data, AI, multimodal and first/last mile innovations are permeating the marketplace
BEV’s are growing in market share but for every BEV bought in 2021, five people bought a diesel or petrol. Infrastructure Incentives and reducing TCO are critical to the adoption
[Warren] and we've gone over a little bit but that's... Thank you for listening... So, final thoughts, I think we've seen over (certainly over my career) a number of years now that these active systems have improved what we wanted them to do, and they've probably created technology that we can use to move to autonomous vehicles in the future. We know the technology around autonomous vehicles works and we know the innovations are there now for electric vehicles to have a significant impact on reducing overall carbon emissions going forward. So, I don't think the technology drive is the issue and not necessarily the...Perhaps adoption is more of a critical issue. One thing I didn't mention on the last graph is the infrastructure is critical because all that growth in electric vehicle sales depends on a major assumption which is that the infrastructure will be in place and the infrastructure isn't in place at this point in time.
So, to come back to the central point of the argument, we've moved motoring to be able to control the driver quite significantly over this period, but I think there are steps that we need to take either through incentivization, maybe through legislation, maybe through understanding the behaviour change more to take more control to deal with some of the bigger issues around congestion, around emissions and to get a better sense of mobility. I know Paul's coughing there so he's moving me on a little bit but that's the end of my talk tonight. I’m quite happy to take any questions.
The slide presentation and Warren’s live video webcam disappears, and a live video of Paul Lynch appears.
[Paul] Thank you Warren, fascinating lecture. Sorry, I have a tickle in my throat post COVID it wasn't to move you on! Fascinating yes. We've got a lot of questions here, so I’ll just run through a few of them with you.
Do you think there's a connection/correlation between improvements in vehicle dynamics and race fatalities of road traffic accidents in the UK over the recent years?
Paul’s live video disappears, and Warren is shown.
[Warren] Oh, yeah, I think... I think bringing anti-lock braking systems in to be mandatory was a key move and then using that technology to improve what we call the lateral stability of the vehicle, so if you're heading for a crash and you want to steer out the way it's really difficult actually, but the vehicle will take over control, so I think those two steps... The ones that exist but probably haven't been mandated yet which I think will be really good to see an increase of is what we call collision avoidance assist systems which use the advanced driver assistance measurement between the vehicle, identifies when the collision's going to take place and probably says "there's no way the driver's going to get out of this so we're going to operate the braking system independently" and it has two effects, one is it can avoid the crash but secondly it can reduce the effects of the crash by reducing the overall deceleration felt by the driver so yes. Yes, would be the straightforward answer but I thought I would explain it as well.
Warren’s live video disappears, and Pauls is shown.
[Paul] Thank you for both the long and the short answer there. Is the lack of progress towards self-driving motoring, freight, trains, and lorries reducing admissions congestion etc surprising to you? Or do you think it will happen soon?
Paul’s live video disappears, and Warren is shown.
[Warren] So, there is the National Infrastructure Group. They are looking at bringing in platooning for freight and I think that's really important because it's probably easier to work with a sector such as freight than to work with the private market. The challenge is the infrastructure because the challenge is creating lanes and taking them out. It's really strange, isn't it? You're going to create congestion in some respects to reduce congestion in the long term because you're going to have to create lanes in which you can undertake this, but plans are in place to do that, and I think you'll see lorries that are tracking each other with the best aerodynamic package and getting us to the place that we need to be. I think once you see that, confidence will grow because people can see it, I think the concept then of getting on the motorway, joining the platoon and getting off three hours at the other end is quite appealing to me.
Warren’s live video disappears, and Paul is shown.
[Paul] Hopefully you can have a nice, gentle doze on the way to your destination then. Another question: manufacturers are bringing a wide range of battery-powered vehicles to market now, when do you think we will see hydrogen-powered vehicles come to market in numbers?
Paul’s live video disappears, and Warren is shown.
[Warren] So, they are kind of in the marketplace. Toyota have got a fantastic offer around hydrogen vehicles the challenge is that the infrastructure isn't in place yet and I think there is work that's being done within Derby/Derbyshire and beyond to look at how we could start to do that first nudge, to do that first change, to create an infrastructure that means people would be driving past a hydrogen filling station and think "you know what? I can go out and buy a Toyota hydrogen vehicle because I know there's support for it." So, there's a chicken and egg which is quite significant around the infrastructure that we need to get through. It needs public sector investment in my view to break that at the starting point. Once you create enough vehicles on the road that are hydrogen-powered, I think you'll then get the private sector bringing more infrastructure in place because they can see a demand for hydrogen. There are other issues about how we generate hydrogen in a clean way because most of it will be quite dirty probably to start off with but broadly it's the infrastructure issue.
Warren’s live video disappears, and Paul is shown.
[Paul] Thank you. There is a discussion surrounding cultural change being required for automation to be more acceptable. How do you think this should be or will be approached? Also, the potential change between ownership and option (e.g.) leasing your car or owning your cars.
Paul’s live video disappears, and Warren is shown.
[Warren] There are a number of factors at play and it's not about the technology, it's about the behaviour change. I think legislation may play a role, and there's a point where it has to come in but that is the stick rather than the carrot. I think, if you look at the adoption of battery electric vehicles, then incentivization schemes are good but they become problematic when they get to volume because they tend to be a cost from the public purse to get them from A to B but perhaps, they'll change the culture. It was probably deliberately provocative and not very scientific to say that we're all emotionally attached to the car, and we like the fact that we can get in it, put the heated seat on, get to work and get out at the other end. People need to see I think that the alternative works because I think it's a confidence piece and I think if the public sector piece can demonstrate confidence in the technology, I think people will start to shift and move.
Warren’s live video disappears, and Paul is shown.
[Paul] I think one final question picking up on some of your comments about students Warren, you talked eloquently about students working in teams towards solving particular problems and how that really enhanced the learning experience. So, it sounds like a great opportunity for teamwork, problem solving, application of learning theory etc, would you like to see national schemes like this to be developed for other disciplines? And what do you think such as initiatives could do for social mobility?
Paul’s live video disappears, and Warren is shown.
[Warren] So, I think you have to be careful about over-engineering the learning process. The reason formula student was successful was because it was a competition first and foremost that just required a huge amount of learning for people to compete and colleagues of mine who might be on the talk tonight who have been part of developing that in the UK and part of leading it through, will tell you that you can have the best car in the world going around the track but if it doesn't impress the judges with the designing it isn't going to work so it just tended to hit the right spots. It was massively supported by the industry, so you get your leading people who are on the trackside one weekend supporting... Sebastian Vettel or Lewis Hamilton and then you know for a weekend a year they're sat looking at students’ designs going "I wouldn't have thought to do it like that," they're quite critical as well you know and rightfully in a compassionate way incredibly critical of the students but I think if you'd have engineered it as a learning process it wouldn't have worked, engineered as a competition for that group of students it was incredibly effective and you put your own structures around how to compete well but actually the informal structures, the informal relationships, those micro-moments between students are as valuable as the broader structure that you put in place. I do think different disciplines could look at how you would bring motivational and competitional work into the classroom that has real meaning, that has almost a sense of loss for a student if they don't get a particular thing right, that's the important thing about performance; if your component fails on the track everyone goes home as a failure.
Warren’s live video disappears, and Paul is shown.
[Paul] That that's a really important lesson to learn in group work, isn't it? You're as strong as your weakest link in some ways.
Thank you very much indeed Warren, there are some other questions there which we will forward to so you can maybe look at those and get back to individuals, but I’d like to now introduce Professor Clare Brindley, the associate provost for research innovation at the University, who will give a formal vote of thanks for your lecture, Warren. Clare.
Paul’s live video disappears, and a live webcam video of Clare Brindley is shown. She is a woman with short brunette hair, wearing a light blue jumper and a necklace with different sized circles around it.
[Clare] Thank you, Paul, I want to thank Professor Warren Manning for his illuminating inaugural lecture and I’m sure the drivers in the room will now have a different understanding of what is happening when we get behind the wheel because Warren has been keen to emphasise that it's not just about cars, the future is mobility.
Thank you, particularly for the insightful questions that those that we haven't answered will get round to Warren answering in the next few days but particularly thanks to our external audience members because it's through this externality that is key to the university's applied research focus that Warren has clearly illustrated through this inaugural lecture. So, on behalf of everyone, Professor Manning, thank you very much for your inaugural lecture.
Clare's live video disappears, and Paul is shown.
[Paul] Thank you, Clare, thank you again Warren for a most interesting lecture and I hope to welcome all to the next order lecture next month. Thank you all very much indeed. Good night.
Paul’s live video disappears and a blue screen is presented with the University of Derby’s 3 hills logo in the middle of the screen, in white.
Inaugural Lecture Series: Provost Research and Innovation, Professor Warren Manning video