IM@UCL: The Podcast - Transcript - Episode 0

Cassidy Martin 0:04

Hello and welcome to IM@UCL: The Podcast, a podcast about the research at UCL that will revolutionise the future of driving. My name is Cassidy Martin, and I will be your host on this journey of self-driving discovery.

I have always had this wide range of interests. Which has made my life interesting but sometimes difficult for honing in on a specific occupation. Anyway, we won’t focus on that. So during my undergraduate degree, unsurprisingly, I took a variety of courses spread out amongst different departments. And at one point I was taking two courses at the same time, that were somewhat related but in different departments with a different focus. The courses were ‘Primate Conservation’ - focused on the conservation of non-human primates - and ‘Environmental Justice’ - focused on how the most marganilised in society are the most vulnerable to environmental problems, anotherwords humanitarian focused. In the conservationist class humanitarian efforts were criticised for not considering conservationist efforts in their plans and implementation and vise versa. So although these two groups, conservationists and humanitarians, shared this common issue, environmental change, instead of working togther, they mainly just fought for funding and for what they thought was best for who they prioritised regardless of the negative consequences on others.

I did enjoy both of these courses, and understood the importance of fighting for both sides and spent a lot of time defending the other side in discussions in each course. It drove me nuts because I kept thinking if conservationists and humanitarians could just put aside their differences, take the time to listen and learn from each other's perspective they could see how much they really have in common and work together as a much stronger unit.

That’s what I love about IM@UCL. IM@UCL is taking a shared issue and working on it with input from an incredibly diverse group of people from a plethora of backgrounds and I have so much respect for that.

If you don’t have an engineering or mechanics background, don’t worry (I don’t either) and have made sure that this podcast is accessible to those of you, who like myself, are not familiar with the field’s academic know-how. And in reality, as you’ll soon discover while listening to this series, and really what this series is all about, research into self-driving vehicles goes beyond these disciplines and includes research from urban planning to psychology to…well, almost any research area you can think of. Because going from a society of manual to semi-autonomous (i.e. somewhat self-driving) to autonomous (i.e. self-driving) is a very complex process and by having that variety of input and know-how helps to make that transition as smooth and SAFE (emphasis on the safe) as possible.

To get us started, I sat down with the director and founder of IM@UCL to give you a bit of a background and understanding of what IM@UCL’s facility and research is all about. This is…

Bani Anvari 3:41

…Bani Anvari. I am a lecturer in intelligent mobility at the Centre for Transport Studies.

Cassidy Martin 3:48

Intelligent mobility is all about figuring out how to optimise the movement of people and goods. And Bani’s research specifically focuses on…

Bani Anvari 3:58

…exploring driver and pedestrian interactions with semi and fully autonomous vehicles in different contexts. 

Cassidy Martin 4:07

The car industry is slowly making cars more capable of driving on their own. For instance, in recent years, self-driving  features such as self-parking and automated emergency braking have become the norm. And this is ultimately a good thing because features like these can help prevent things like accidently bumping into the car next to you while trying to squeeze into a tight parking spot. We’ve all been there. And one of the ways researchers are trying to ensure these ever-increasing numbers of features do lead to a safer, more efficient driving experience, is by looking into…

Bani Anvari 4:44

…infrastructures which allow for connectivity and exchange of information between different cars.

Cassidy Martin 4:51

But getting cars to connect with each other is complicated. Especially when not every car is at the same level of self-driving capability.

Bani Anvari 5:02

If we have an ideal case, where we have 100% of the cars, our streets connected and autonomous, we have full information about where all the cars are, how they're travelling, where their destinations are, how fast or slow they're travelling at times, the occupancy level of the cars and so on,  it gives us a lot of capabilities in terms of improving the efficiency of getting A to B because we can make sure that cars are driving with the specific average speed which allows minimising stop and go behaviour which results in pollutions. They can keep a safe distance from other cars, and can better plan the route for different vehicles to get to the final destination. And generally not only optimised for a single vehicle, but actually optimised for the whole system and for the benefit of a higher number of people. But before we get there, there's this phase where we have a mixture of cars that are autonomous and connected, and cars that are still from old times and they don't have those capabilities and so on. So we need to deal with this phase where we have some information, but we have a lot of missing information, and how can we optimise the movement of people or goods with the data that we have? And how can we best make use of the data that are missing?

Cassidy Martin 6:31

That tricky, in-between phase is the crux of Bani’s academic work. And in order for Bani to fill in that missing information, she first needs to understand how people operate. And as she puts it…

Bani Anvari 6:43

…go beyond the current human autonomous system control, engineering solutions, and revolutionise and extend the complex interaction between humans and highly automated vehicles. And this can happen by linking mind to intelligent machines, so the mind of the human to the intelligent machine.

Cassidy Martin 7:07

You're hoping to fuse human mind and automated? What does that mean?

Bani Anvari 7:14

So in order to be able to have autonomous vehicles that can replicate human driving behaviour, they need to be able to replicate how we make decisions in our mind. So replicate, for instance, capabilities of predicting what is coming next, and then reacting to it based on our prediction. In order to do that, we need to get familiar with how the human mind works, how it makes decisions, how the conscious and unconscious part of our mind works and how they fuse for the decision-making process.   

Cassidy Martin 7:55

And to be able to research the human decision-making process while driving, you need the right kind of equipment, something that UCL previously did not have. This is what sparked Bani’s desire to create a research facility at UCL.

Bani Anvari 8:10

For doing experiments, and for testing my ideas, I require a car simulator. At UCL, we didn't have a car simulator. We were always relying on collaboration with institutions, which have a car simulator. And of course, the car simulator allows people to not only do research within a specific field and dimention but also within other fields, like looking into decarbonizing through improved vehicle technology, or looking into influencing driving behaviour. So, I was seeing  this missing element at UCL when there was a call toward getting a capital grant for building facilities. And so I came up with this idea that maybe we should look into developing a car simulator at UCL. But maybe not doing it by myself, but doing it with a group of people who are interested and who can make use of this facility for their research. So I spoke with a number of people, especially people who I thought are relevant for my research area and research areas that I thought would be explored using these facilities, and got the confirmation to be involved and be on board for writing this proposal. And thankfully, we are successful in securing the money.

Cassidy Martin 9:39

Yay!

Cassidy Martin 9:45

Getting people on board and securing the money was one thing, but the real challenge came when building the actual facility.

Cassidy Martin 9:53

Did you face any, like, particular challenges when you were trying to build it?

Bani Anvari 9:58

It has been challenging throughout. But at the same time very interesting and very exciting because we learned a lot throughout the whole process. It has been the first time for all of us doing such a thing. There were a number of hardwares that we purchased, but a car simulator is not something you can just go and purchase. You need to purchase different parts - softwares, hardware - and then try to integrate them. Then you need to make sure that all of these systems are speaking to each other. All of these fields were quite out of the comfort zone of everyone within the team. And we had great support from Ansible Motion, from specific people like Kia and Andy from Cave, and a lot of other engineers at UCL who supported us to make these installations and throughout. Every day, we learned something new because something was not working. We had to go through the whole process from the start to get to know what the problem was. Sometimes we don't even know what the problem is, so we were turning it off and turning it on and suddenly everything was working! And we were like, ‘okay, thank God, this is working’. Really, really challenging. But at the same time, we learned a lot. So it has been a very enjoyable journey.

Cassidy Martin 11:21

The newly completed facility is known as Intelligent Mobility at University College London, or IM@UCL for short. And comes with a host of technological equipment that researchers can utilise.

Bani Anvari 11:34

We have a driving simulator, made by Ansible Motion, in kind contribution from a car manufacturer with a 180 degrees field of view in front of the car, as well as psychophysiological and behaviour monitoring technologies...

Cassidy Martin 11:53

...like the EEG cap. EEG caps are small painless sensors that are attached to the scalp….

Bani Anvari 11:59

…so you can see which parts of the brain or which regions of the brain get activated at different times, and which frequencies you have in different situations. I mean, the human brain is very complex. And when you're making a decision, more or less, a lot of different areas of the brain get activated, and so on. But understanding human minds allows us to get inspired for building and developing these intelligent machines. And also, successful cooperation between humans and machines. Not only that, but also we’re looking into where the driver is looking at. So we put eye trackers on, and we combine these things. We look into where they're looking, and then how they're making decisions in their head. And what are they doing in terms of driving. Are they braking? How are they changing the speed? How are they changing the gear? How are they changing the routes that they're travelling? And so on.

Cassidy Martin 12:56

Okay. And then, when you gather this information, you get the data - what the eyes are doing, what regions of the brain are lighting up and so forth -  is that information used to design how the automated vehicle moves? Is it using that same kind of thinking for that? Or is it creating safety features based on how a person typically reacts?

Bani Anvari 13:28

Both of them, so getting inspired from how people make decisions, but also looking into how different feedback is, feedback that the autonomous vehicles’ providing, is actually  impacting the driver. And are they really doing what we are aiming at? So for instance, there have been accidents, where there was visual and audio feedback, and still the driver was out of the loop of driving. And within the seven seconds that the drivers were informed of taking back control, he was not successful - to take back control - because he was completely out of the loop of driving. So the feedback system was not successful, necessarily. Also, the driver was not fully informed about the capabilities of the car, what it can or cannot do, and who is responsible for what while driving. But actually understanding how we can make this cooperation between the human and the car successful requires us to understand how different feedbacks are impacting people and people's situational awareness in different scenarios.

Cassidy Martin 14:38

And as pointed out earlier, Bani is not the only one benefiting from the creation of this facility.

Bani Anvari 14:43

So it's a research facility, but it’s a research facility that has been initiated through putting the ideas of a number of early career researchers together. And these early career researchers are coming from different backgrounds. Backgrounds from engineering, neuroscience, psychology, architecture, policy making. And also they're all coming from different faculties and departments such as faculty of Engineering Sciences, Brain Sciences, and The Bartlett, and so on. Also this is the first London based, fully immersive, driving simulator. So having access to this in London provides a lot of different opportunities for us. Also, it enables the research alignment with the UK government's strategic challenges and fundings that are available. So we are aiming to leverage UCL's cross disciplinary and cross departmental faculty research. And with the combination of this facility and that background, I'm hoping that we see more and more exciting research happening at UCL.

Cassidy Martin 15:52

Yeah, absolutely. And then I guess also, you now become the facility where other people want to come and work.

Bani Anvari 16:00

Exactly. It is also opening up new opportunities for collaboration.

Cassidy Martin 16:05

Yeah. And so you've mentioned that it's a team of early career researchers, can you tell us about why you chose early career researchers?

Bani Anvari 16:13

I mean, early careers are fresh out of their studies so they have the latest knowledge. And they're very keen to look into problems and they are probably keen to push for disruptive solutions, therefore, stepping out of their comfort zone. Not everyone, but the possibilities and the potential is there. Also, they have their whole career ahead of them. And therefore they are more motivated to put together proposals using this facility so that they can explore their ideas - some of them blue sky ideas, some of them more practical, higher, TRL. So that's why early careers seem to be ideal, in this case. Saying that, we also have an advisory panel, which contains a number of people from different departments, more senior people who can advise us through this journey. Again, from different backgrounds and from different departments. When we, for instance, for building this facility, we also had discussions with them, gathered their ideas, before making decisions about what to purchase, and also, the research areas that we possibly can explore using this facility.

Cassidy Martin 17:29

Yeah, so you're getting the best of both worlds, right? Because then you're getting the early career researchers, and they’re maybe a little more, some of them, maybe a little more idealistic, I guess. And then you've got the people who've been doing it for years who say, ‘okay, yeah, this is gonna work.’ Cause you have that kind of practicality, kind of like, ‘okay, yeah, you should get this, and if you want to do this, then get this.’ So you have both, I feel like. Which is a good thing.

Bani Anvari 17:53

Yeah, I mean, that was the goal. So I hope that, yes. That’s exactly, that was the idea. And I guess we have a very good combination of people on board. And so far, so good. So…

Cassidy Martin 18:07

Yeah, from the people that I've met, they’re lovely. They’re great. I think you've got a wonderful team.

Bani Anvari 18:12

Thank you!

Cassidy Martin 18:15

So, what are you most hoping to achieve with IM@UCL?

Bani Anvari 18:19

I hope IM@UCL and the research that we are doing at IM@UCL is going to inspire people. Inspire people to come to STEM and be within STEM. And also get inspired for future research within this field or any other field, doesn't matter, or other applications. So it can be that we are doing research that people can see can be beneficial for another field. So I would be very happy to see if we manage to get to that stage that we have other people coming to us, telling us that, ‘oh, you inspired me to stay in, or do research, or study this field or do research in this field, or you have. Your research has inspired us to go and do something different within our field or within another field, which can be beneficial for people.’

Cassidy Martin 19:13

I don’t know, I find it inspiring, just like talking to you. So you've inspired one person!

Cassidy Martin 19:24

As you’ll soon find out, I’m not the only one who has been inspired by Bani and the creation of the IM@UCL facility. Over the next 6 episodes, you will meet early career researchers at UCL whose work will be integrating the use of this unique facility.

Thank you for listening to IM@UCL: The Podcast. If you would like to learn more about the research at IM@UCL, you can check out their website at www.ucl-intelligent-mobility.com and/or subscribe wherever you are listening to this podcast so you can be notified when new episodes come out each month.

This episode was produced and hosted by myself, Cassidy Martin, with music from Blue Dot Sessions. It was brought to you by IM@UCL which is part of UCL PEARL in Daganham and supported by UCL Minds - bringing together UCL knowledge, insights, and expertise through events, digital content, and activities that are open to everyone.

A special thank you to Bani this month for sharing her time, knowledge, and insight.

I hope you enjoyed listening to this podcast and feel like you learned something new, like I have with everyone I have interviewed in this series. Take care and I’ll see you again next month, same time, same place. Cheers!

Back to IM@UCL Podcast Home