On the 7 May the iMobility Forum Working Group on Automation for Road Transport held a workshop dedicated to the subject with speakers from across the globe presenting the progress made and challenges still to overcome. Laura Johnstone reports.
Automation and automated vehicles have long been a source of fascination since as far back at the 1930s when American industrial designer Norman Bel Geddes first showcased the General Motors-sponsored Futurama at the 1939 New York World’s Fair.
Revellers marvelled at his vision of the year 1960 the most futuristic aspect of which was an automated highway system. Bel Geddes imagined a future where humans and thus human error would be eliminated from the driving process where bottlenecks would be a thing of the past and where city traffic would be greatly accelerated. Of course Bel Geddes’ predictions did not come true (although by 1960 General Motors had developed an automated highway prototype) and even today 70 years on we are some way from reaching his safe efficient fast and comfortable automated utopia.
But the buzz surrounding automation has continued to amplify thanks to technological developments rendering automated vehicles entirely feasible as well as the publicity surrounding initiatives such as the DARPA Grand Challenge the Autonomous Audi TTS Pikes Peak and most notably the Google Car. Many vehicle manufacturers and suppliers are known to be investing in applications which are set to steadily increase the level of automation available to the driver in the coming years.
Automation will help to solve several societal challenges – first and foremost in reducing the number of road accidents but also in increasing the efficiency of road networks and reducing fuel consumption – and research into the topic is currently being funded by national and supranational authorities.
In Europe the European Commission has funded a number of projects over the last 20 years from Chauffeur to Cybermove and Cybercars. More recently there have been three projects in particular working on different aspects of automation: Citymobil which focused on achieving more effective urban transport through testing and evaluating new solutions based on advanced city vehicles HAVEit which aimed to improve traffic safety and efficiency through the development and validation of next generation ADAS as well as defining different degrees of automation and developing an appropriate vehicle architecture suited to highly automated driving and SARTRE which developed vehicle platoons for public highways.
The follow up to CityMobil launched last September with some of its aims including the definition of common technical specifications for cybercar manufacturers the undertaking of detailed studies on implementing cybercar systems in 12 cities and the definition of a European legal framework for certification of automated transport systems. The project includes 45 partners and will run for just over three years. In addition supervised automated driving also featured as one topic in the last call for proposals of FP7 under DG CONNECT cementing the commitment of the European Commission to investing in and developing automation for road transport.
Japan meanwhile is using cooperative systems to overcome the unique challenges stemming from its densely populated immense urban areas with the ITS Spot Service the world’s first infrastructure for vehicle cooperative systems which offers drivers dynamic route guidance driving safety support and supports electronic toll collection. Furthermore cooperative vehicle longitudinal control is being used in conjunction with the ITS Spots to reduce traffic congestion in uphill highway areas (the so-called “sag effect”).
The Japanese Ministry of Land Infrastructure and Tourism has also established its ‘Autopilot System Council’ which is working towards the deployment of automated driving on highways from cooperative vehicle control and adaptive cruise control towards steering assistance. The systems envisaged will utilise and further develop existing technologies with the autopilot system estimated to be put into place in the early 2020s.
In the US the Department of Transportation (DoT) is currently drafting its vehicle automation programme plan which aims to develop a multi-modal research plan to focus on accelerating public benefits whilst diminishing public risk. The draft schedule for this plan is due to be released in summer of this year following collaboration with a range of external stakeholders.
An important part of this plan is the identification and categorisation of the issues and challenges inherent to automation. Here the Transportation Research Board (TRB) ITS Committee will play an important role. TRB Committees are made up of volunteers from government industry and academia who come together to address shared interests in transport. Last summer the TRB held a workshop on automation which brought together experts from around the world where the state of the art of many aspects of automation were presented and specially designed breakout sessions were held to facilitate debate and discussion. This year’s TRB workshop will take place in July in California and the discussion at this event will be considered in the final DoT vehicle automation programme plan which is set to be released in September.
It is clear that all three regions envisage automation as part of the future of road transport and whilst there has been great advancement in the research and development of automation applications and services over the last few decades – so much so that we can confidently say that automation technology in itself is almost ready to be deployed – there remain several stumbling blocks before anything resembling Bel Geddes’ Futurama comes to pass.
So if automation in its varying degrees is technically feasible why aren’t we all spending our daily commute reading checking emails or watching films behind the wheel?
At the iMobility Forum Automation workshop two issues were raised repeatedly – human factors/ user acceptance and legal issues.
Regarding human factors and user acceptance there are currently many unanswered questions that remain particularly regarding whether the driver will understand the new system and be able to work with it whether he trusts the new system believes in its reliability and is able to accept handing over control to the vehicle.
Furthermore as we rise through higher degrees of automation how can we be sure that the driver understands how much attention he has to pay to his surroundings? And can the driver handle potential changes in workloads for example switching from underload to excessive overload in a critical situation?
The CityMobil project included small-scale studies which examined situational awareness and driver reaction to unexpected traffic situations whilst driving in “normal” semi-autonomous and fully autonomous vehicle simulators both in urban environment and in dedicated eLanes. Vehicle manufacturers are also working on their own human-machine interface (HMI) solutions. However not enough work has been carried out at this point to give a sufficient answer to these questions and it is clear that there needs to be a coordinated concerted effort to ensure that the driver’s needs in terms of safety efficiency and comfort are met.
Developing a HMI which takes into account human factors and meets driver’s needs although it requires some work is perfectly doable. The second big issue in road transport automation however is not so easy to solve. It is an issue which is raised time and time again and yet no one so far has come up with a definitive answer on how to get round this obstacle – the legality of automation.
The common starting point for this discussion is generally the 1968 United Nations Vienna Convention on Road Traffic specifically articles 8.1 “every moving vehicle or combination of vehicles shall have a driver” 8.5 “every driver shall at all times be able to control his vehicle or to guide his animals” and 13.1 ‘every driver of a vehicle shall in all circumstances have his vehicle under control so as to be able to exercise due and proper care and to be at all times in a position to perform all manoeuvres required of him”.
The Vienna Convention has been ratified by 70 countries including most of Europe but not the US or Japan. The US is however a signatory of the predecessor to the Vienna Convention – the 1949 Geneva Convention on Road Traffic which also contains some of the above articles. The terms of these conventions therefore render automated driving at a certain point where the vehicle has more control than the driver not legal.
The Vienna Convention has been amended on several occasions over the decades (most recently in 2006) so in theory it is possible to change it again to allow highly to fully automated driving. However it is often argued that amending the Vienna Convention to allow automation in road transport would be a long process; far longer than it will take for vehicle automation technologies to be ready for launch on the market. Therefore alternative solutions should be sought.
You may have noticed above the rather clunky use of “not legal” as opposed to “illegal”. This is a subtle yet important nuance as there are some who believe that the Vienna convention can be interpreted in a way that allows for automated driving.
Firstly what constitutes “control”? Does this mean that the driver must be in control in the sense of physically driving and controlling the vehicles’ functions or could we interpret that to mean the driver must monitor the automated functions and be ready to take control when something goes wrong?
Secondly “the driver must be in control of the vehicle” but does that mean he must be inside the vehicle? Could the driver in fact be outside keeping control remotely or from a control centre?
Such are the grey areas put forward during the Automation Workshop summed up with the rather nice phrase: “it’s not allowed… but it’s not forbidden!”
In 2011 Nevada famously made automated driving legal within the state. The legislation enacted by the State is a good example of how this grey area can be exploited; Nevada’s regulations state that vehicles may only operate in autonomous mode if they have a compliance certificate which leads to the important phrase “the autonomous vehicle may be operated in autonomous mode without the physical presence of the operator only if the certificate of compliance certifies that the autonomous vehicle is capable of being operated in that manner”. So with some cautionary previsions driverless vehicles are legal in Nevada.
For the moment the State’s Department of Motor Vehicles will only accept licence applications for testing purposes and state on their website that autonomous driving is not available to the general public but this is nonetheless a big step forward for the deployment of autonomous vehicles as not only does the example of Nevada (and Florida and California with more States to follow) show how automated and driverless vehicles can be deployed within the existing legal framework but these states also provide a huge testing ground for measuring and evaluating open issues such as user acceptance.
In Europe there are also some provisions being put into place which rely on creative interpretations of the Vienna convention; for example the idea that automated vehicles could be classified as trains where the vehicle’s trajectory can be considered as a virtual (rail)track. Specific tests have also been permitted on public roads; for example last year the SARTRE project successfully drove a platoon of vehicles down a motorway just outside Barcelona in the midst of normal traffic. The CityMobil project also demonstrated small fully automated vehicles inside the pedestrianised city centre of La Rochelle in France. Although such initiatives are very positive no European region or country has yet provided carte blanche for the testing of automated vehicles as we see in the US.
Almost 60 years after Bel Geddes’ predicted we would be zipping along automated highways his vision is edging towards realisation. The technical barriers preventing us from driving more safely cleanly and efficiently with the aid of automation applications and technologies have either been overcome or are very close to being overcome. However it is clear that non-technical barriers particularly those outlined above will continue to slow down the implementation of automation until a concerted common effort is made to resolve them.
1960 has long been and gone; let’s hope that by 2060 society as a whole can look back at many happy years of automated driving.
By Laura Johnstone
iMobilty Forum Automation for Road Transport WG: http://www.icarsupport.eu/esafety-forum/esafety-working-groups/automation/
iMobility Forum Automation for Road Transport Workshop: http://www.icarsupport.eu/events/workshop-automation-in-road-transport/
CityMobil project http://www.citymobil-project.eu/
HAVEit project http://www.haveit-eu.org/
SARTRE project http://www.sartre-project.org/en/
ITS Spot Services http://www.mlit.go.jp/kokusai/itf/kokusai_itf_000006.html
TRB ITS Committee https://sites.google.com/site/trbitscommittee/
UN Vienna Convention on road traffic http://www.unece.org/fileadmin/DAM/trans/conventn/crt1968e.pdf
Nevada Legislature http://www.leg.state.nv.us/Session/76th2011/Reports/history.cfm?ID=1011
State of Nevada Department of Motor Vehicles http://www.dmvnv.com/autonomous.htm
Link to original Article
Original Publication Date: Tue 19 Mar 2013