James Day attends the Connected and Automated Vehicle Engineering conference to gain insight on the progress of the UK’s autonomous trials and how legislation is moving
From September 25-26, Hotel La Tour, Birmingham, hosted the Connected and Automated Vehicle Engineering conference for 2017. The event featured speakers involved with many of the UK’s most high-profile autonomous vehicle trials, along with a number of manufacturers and Government organisations.
The event was chaired by Alan Stevens, Chief Scientist and Research Director at the Transport Research Laboratory (TRL).[wlm_nonmember][…]
By subscribing you will benefit from:
- Operator & Supplier Profiles
- Face-to-Face Interviews
- Lastest News
- Test Drives and Reviews
- Legal Updates
- Route Focus
- Industry Insider Opinions
- Passenger Perspective
- Vehicle Launches
- and much more!
He explained how the conference had secured speakers from each of the three major Government-funded autonomous vehicle projects, introducing each of them in turn. Each project makes some use of a vehicle which could be considered an autonomous minibus, with much of the initial research examining ridesharing vehicles.
The first day also provided interesting insight on legislation and regulation, both international and UK-based, which is moving to make autonomous technology a reality.
Autodrive
Mike Waters, Head of Policy and Strategy at Transport for West Midlands (TfWM), gave a presentation on Autodrive – an ongoing project in Coventry and Milton Keynes.
The autodrive project was a response to a Government driverless cars competition, with the specification of a car and one other form of development.
Mike highlighted the benefits of testing in the two different locations, with Milton Keynes more representative of urban environments worldwide, while Coventry has the medieval street layout more commonly seen throughout the UK. The three-year project is split into three primary components – city transport, M1 passenger vehicles and low speed autonomous pods, and Mike described the project as ‘large and ambitious.’
A video was shown, highlighting testing at Horiba-Mira’s proving ground in the Midlands, which included the vehicle identifying and stopping at traffic lights, moving out of the path of emergency vehicles and avoiding collisions at complex road junctions.
“Our work is quite focused on the vehicle, and we have quite limited vehicle to infrastructure connectivity in this project,” Mike said. “Trials are getting progressively more complex and we’re looking at the impact of the vehicles on road user attitudes and congestion.”
An interesting statistic which Mike brought up from a global survey on autonomous vehicles was that 23% of people expected that they would call an autonomous vehicle to pick them up from a bus stop. “There’s learning to do about where is technology will be used, and who will buy it,” Mike added.
Greenwich Gateway
Representing the high-profile Greenwich Gateway project was Richard Cuerden, Chief Scientist for Engineering and Technology at TRL. He described the upcoming trial as a first-mile and last-mile public transport system.
“The idea of the pod is to really explore how people would or would not want to use that sort of vehicle,” Richard said.
“Before you get there you must do a safety case and be really sure you’ve designed something robust enough.”
Richard showed the many forms of onboard safety technology which keep the vehicles and other road users safe, including cameras, lidar and ultrasound to detect obstacles at various distances, as well as a safety steward who is currently required to be onboard monitoring the vehicle. A GPS system monitors the vehicle’s location.
The idea of transport utopia vs dystopia was brought up once again by Richard. He asked the question of whether a lack of regulation could result in an inhuman city, with little communication between people, high congestion and unemployment, due to the majority of people travelling around in small limited occupancy autonomous vehicles.
“The promise of accessible, safe, assisted travel is what we’re trying to deliver,” he said. He showed an image of what was essentially an autonomous mobile coffee shop to show the potential the technology has to revolutionise travel in a positive way, should infrastructure be put in place to support it.
Richard also highlighted research on how people react to an autonomous vehicle. Through work with a simulator, it was found that drivers did drive with slightly more risk when more autonomous vehicles were on the road, by pulling out into slightly smaller gaps from junctions, but they did not ‘bully’ the autonomous vehicles.
Bristol Venturer
The third of the major autonomous projects is Venturer, which is taking place in Bristol. This projects representative and the third speaker of the day was Tony Pipe, Deputy Director of the Bristol Robotics Laboratory at the University of West England.uch of the work at Venturer has focused on control handover (where the autonomous vehicle put the human driver back in control) and integration with other road users.
Studies were carried out in both simulators and real vehicles. Reaction times were found to be slightly faster in the real world, since people felt more relaxed and safe in the simulator, though findings were broadly parallel.
“Although we’re all focusing on urban areas rather than motorway driving, there are publications on an international scale for motorway or highway handover of driving,” Tony said.
“For motorway style driving, although the vehicles are driving faster, a pedestrian will not suddenly appear as it is illegal for them to walk along the motorway. It’s less likely that someone will need to react very quickly during the handover and the autonomous vehicle can take more time over it to ensure the driver has the necessary awareness of what is going on. You may well not have that time in an urban environment.
“We found that in those urban environment, it may never be a good idea to give control back to a human being. Although people took back control more quickly, it still took several seconds for them to effectively assume control.
“Some of the things people were doing as they retook control were not safe, such as slowing rapidly to half of the vehicle’s speed when it was driven autonomously, or swerving, and people were not very aware of their surroundings for around five seconds.
“This was not an emergency handover. This was the vehicle asking them to take control at a random point, and the driver knew it was going to happen. An emergency handover would be even more questionable.”
Connectivity trials
Mike Waters provided a second talk on connectivity projects across the West Midlands.
The first was the UK CITE project, which he said was likely the largest connectivity project running in the UK, covering a 42 mile real-world test environment.
“We cover all road types, including some challenging mixed-use environments with no white lines, through to the M42, M40 and M6 corridor,” he said.
“Our philosophy with this project is that we don’t yet know the answer to which connective technology is the right one. We need to do much more real-world testing where things can and do go wrong.”
UK CITE has examined a number of technologies, including mobile phone-based connectivity, and vehicles communicating with each other. Mike said the purpose of the communication dictates which form of technology is the most appropriate.
He explained: “We also think the right choice will change over time, as technology is used more and adapted, and new vehicles are introduced. We’re focused on having a balance of technologies which can be tested.”
The other project mentioned by Mike was an ‘on-the-cheap’ freight project which has received less than £300k of investment. It is designed to test technology which is not supported by substantial roadside infrastructure as can be found on the high-tech M42 corridor.
The system uses a combination of WiFi and 4G to attempt to tell lorries to speed up or slow down, or adjust signals. It uses mostly existing architecture with an open API output, making it easier for third-party systems.
“The point here is not to be scared of the technology and assume everything requires a large amount of infrastructure and expense,” Mike said, displaying an image of a very basic electric circuit made with Lego which could control a set of traffic lights. “We can adjust existing, simple infrastructure ourselves for a fraction of what others can.”
Mike added: “It’s all great in theory, but it needs to work in the real world which is why this extensive testing is necessary.”
Improving highways
Phil Proctor, Head of Highways England’s technology team, began by speaking about the safety improvements he believes autonomous vehicles could make to the motorway network, reducing the number of people killed or seriously injured. He said that even the relatively low-level autonomous technology seen on vehicles today, such as autonomous emergency braking, was substantially reducing the number of accidents the vehicles equipped with it were involved in.
Moving on from safety, Phil said the Highways Agency network is expecting a significant growth in traffic, on top of the 48% increase between 1993 and 2013, adding to the £4bn it estimates that congestion is costing the UK economy.
Autonomous vehicles are often spoken about in the same breath as connected vehicles, and Phil touched on this subject too. He noted that many of the signs used to inform road users on motorways were autonomously controlled, and the information could potentially be sent to connected vehicles to help them anticipate out-of-sight hazards.
“Autonomous vehicles need to react to what the signs are saying, because there will be a mixture of autonomous vehicles and human drivers on the motorway,” Phil said. “They need to cope with these systems to improve the flow of traffic.
“We’re about to start a two-year project with autonomous trucks to understand how to implement this technology safety on the highways road network.”
A word from the UN
An update on the legislation surrounding autonomous vehicles formed another part of the programme.
First to speak on this subject was Francois Guichard from the United Nations Economic Commission for Europe (UNECE), who spoke to the conference’s attendees through Skype.
He said the UNECE had a mandate to review the 10kph limit on self-steering systems, which it was looking to remove for inter-urban journeys subject to technical requirements. Once this has gone through, manufacturers may be able to work on lane-keeping systems for highways which can work without the driver touching the wheel.
Francois raised some issues with the identity of autonomous vehicles. He showed a standard road car, which had no identifying features to show whether or not it was autonomous, along with a Navya autonomous minibus, which because of its low capacity may be treated as a road car in many countries and would therefore require some changes in how it would be assessed for safety. A small, remotely controlled delivery vehicle was shown, with the question asked of whether it would require regulation and what that would be, and a small autonomous two-seater vehicle was displayed as it does not easily fit into the classification of passenger car or quadricycle, and was therefore difficult to assess.
“Currently we don’t have clarity of the technical requirements of these vehicles, especially for safety,” Francois said. “We need to have clear definitions in the requirements for the level of technology in automation and the autonomous capability. Only by solving these issues can we bring these vehicles safely into traffic. This is crucial for manufacturers, Governments and users.”
Government insight
Ian Forbes, Head of the Centre for Connected and Autonomous Vehicles, provided an update more centred on the UK and the Government’s autonomous strategy.
He said the strategy recognised that change is coming, and the Government needed to be clear on what it wants from the change, as well as adapting in response to new information.
Ian explained: “In the top of our minds we have road safety, improving the efficiency of our road network, improving travel experiences – particularly for those who find it difficult to get around our transport system, and we want to ensure UK businesses are at the cutting edge of new vehicle technology.
“Ministers in the UK have been clear that they want the country to be as open as possible when it comes to the development and deployment of automated vehicles.
“On the regulatory side, in order to achieve that objective we have to look at vehicle safety, driver licensing, insurance and so on. It’s a big undertaking across many areas of the regulatory framework, so we need to be selective about where we start.”
Ian highlighted that in 2015, the UK was one of the first countries in the world to publish a code of practice for testing these vehicles on public roads.
The first area of focus for the UK Government has been insurance. Autonomous vehicles are expected to shake up the industry dramatically as it moves away from insuring drivers, to insuring vehicles.
Ian explained: “The insurance industry’s response to our consultation said that the change to the liability framework was fraught with difficult consequences, but it argued that there was an easier and more effective way for use to achieve our policy objectives.
“Surprisingly, they asked for the Government to force them to pay out on claims. Their alternative proposal was that when someone takes out an insurance policy to insure a vehicle that can drive itself, part of that policy should also cover automated driving.
“They then suggested that when there is a crash involving an autonomous vehicle driving in automated mode, the insurance company should have first instance liability – when an innocent party experiences harm, the insurance company should be compelled to pay the claim.
“In return, they suggested they should have strengthened rights to reclaim damages from the party ultimately responsible – whether that is the manufacturer, supplier or someone else. This means there would be no gap in coverage for accident victims. It would feel very similar to how it is at the moment, while complicated legal action would go to corporations, not individuals.
“It seemed like a far more sensible solution, so we adopted it, and the new Automated and Elecric Vehicles Bill will enter parliament soon to establish this framework.”[/wlm_ismember]
