AVs and Smart Cities: Insights from the ServCity Project

Welcome to London. The British capital is a hub for business and innovation and was once again ranked as the smartest city in the world in the IESE Cities in Motion Index. So there's no surprise that in such a vibrant city where Mobile has been a focus for decades and even centuries with the world's first metro being built here that there would be autonomous vehicle projects we've come to talk to James long head of technical consulting at the smart mobility living lab in London which has just concluded a three -year AV research project so we are a real-world testbed to enable new mobility new technologies, new services to be developed, validated and ultimately trialled and deployed in the real world. The Service City project was a really fantastic activity. It was a part government funded and part industry funded project with a variety of partners. So it was led by Nissan and involved TRL, the Transport Research Laboratory in the UK, Connected Places Catapult, Hitachi and University of Nottingham. And it was a collaborative research and development project to explore... how autonomy can be utilised in complex urban environments. A big part of this project was getting the vehicle to be able to operate appropriately in autonomous mode in the city, but also where it might take input from the roadside infrastructure. So on the vehicle they had a selection of different sensors, so they had lots of LiDAR cameras, radar, to be able to enable it to get an understanding of its environment. They had lots of... compute to be able to create its local maps and then Nissan's own automated driving system within that to determine how it should progress along the streets and what it should respond What it was then taking from the infrastructure was we used all of the Smart Military Living Lab's 300 roadside cameras and we used a subset of those to give inputs to the vehicle. So our roadside infrastructure, we've developed all of the underlying fibre network. So we have a private fiber network that we can use, which feeds all of our roadside sites. To then interact with the vehicle, we're sharing it over short -range communications using DSRC. We're using cellular V2X and we're also using an internet -based server -to -server interface. We're using 4G for the purpose of this activity because it wasn't the primary focus for the particular environment. initiative but certainly the same can be applied through 5G. So we were looking at a situation where you had a big long bus stop that was beyond the line of sight, it was around the corner, if you ended up in that bus stop you find it quite hard to get out of it. So we had a camera that was looking and watching that bus stop to determine when buses were present, sending that to the Nissan vehicle much much earlier so it could get it in real time, get the information and incorporate it into the vehicle. operation. So it was the vehicle and the infrastructure working hand-in-hand. Autonomy is going to be one part of many many different services that that infrastructure could support. I think one of the activities that we've seen a lot of, particularly in developing this facility, was the importance of very configurable architecture for the... roadside infrastructure so whether we're looking at lamp columns that have got power network connectivity for example to be able to take on a variety of different devices so we're looking at situations where we can detect traffic movements where we can start to pull out air quality measurements so I think what we've what we've seen through this project is understanding how you can take not just the vehicle in its own right but how do you address a system as a whole how do you start to bring together what's happening with users what's happening with vehicles but what's happening with the city and that could be a direct input to the vehicle it could be something that says how do I route this most effectively what's the fleet management how does this work with the traffic management in the city so I think there's lots of different ways that this can start to interact so whether it's as we've done where it's providing shared situational awareness and direct inputs to the vehicle, so you might have all sorts of different sensors, detection software that sits with that. It could be around the connectivity of another use case to enable the wider rollout of different systems. It could be about how that works with advanced traffic management solutions. Whether or not it's in the sensor industry, whether that's roadside or on the vehicles themselves, whether it's the connectivity, whether it's people who are trying to maintain streets in different ways. It's underpinned also by all the simulation companies. The supply chain is enormous around automated vehicles, but the potential impact that that can have is also substantive. So I think what we're seeing progressively more of are... organisations who are developing and validating their solution, not just physically in the real world, but also digitally in a simulation environment. And so that digital twin is becoming a really important part. So lots of organisations have looked at synthetic scenarios for how they simulate their vehicles. In my conversation with James, he told me that the UK has taken a proactive stance in the world. world of AVs and we discussed how some companies are choosing to have all their technology on board to be a closed market. But AVs communicating with road infrastructure seem like an opportunity for smart city industry players.