Advanced Driver Assistance Systems

Advanced Driver Assistance Systems (ADAS) is used to describe active safety systems on a vehicle that can identify safety-critical situations and take action, either automatically or by sending a warning to the driver. 

ADAS systems have developed fast, thanks in part to sensing technology; cameras, radar and laser technology referred to as lidar. This technology is also helping the development of automated and driverless vehicles.

A limited amount of ADAS is mandated through European law. Mostly it is not mandated. ADAS are rapidly being developed and fitted by vehicle manufacturers, often to high-end vehicles, and further regulation is being considered by the EC. 

Existing systems mandated on new vehicles 

Some mandatory requirements under European law for new vehicles are described below. 

Autonomous Emergency Braking Systems (AEBS) senses the chance of a collision ahead and automatically brakes to mitigate or avoid it. EU General Safety Regulation 661/2009 requires medium and heavy commercial vehicles to be fitted with it.

Lane Departure Warning Systems (LDW) senses an unintended lane departure and provides a warning to the driver. EU General Safety Regulation 661/2009 requires medium and heavy commercial vehicles to be fitted with it.

Electronic Stability Control (ESC) detects and reduces loss of traction (skidding) through automatic braking of specific wheels and/or engine braking. EU General Safety Regulation 661/2009 requires all vehicles to be fitted with it.

Automatic Braking Systems (ABS) on cars monitors the speed of wheels and applies the brakes independently to individual wheels in order to prevent them from locking.

Controlling speed and emergency braking: opportunities for ISA and AEBS regulation

There is a review of General Safety Regulation 661/2009 underway by the EC, presenting opportunities for more ADAS to become mandatory. Prior to this review, the EC commissioned TRL (the UK's Transport Research Laboratory) to review ADAS, consult with stakeholders, and make recommendations regarding the value of systems on a safety and cost-benefit basis. TRL's 2015 report[i] made a number of recommendations including: 

Intelligent Speed Assistance/Adaptation (ISA)
TRL recommended Intelligent Speed Assistance/Adaptation (ISA). ISA identifies if a vehicle is exceeding a speed limit and can: control the vehicle to below a limit (mandatory ISA); or enable the driver to control whether the system can restrict their vehicle speed and/or the speed it is restricted to (voluntary ISA); or warn the driver if they are exceeding a limit (advisory ISA). The speed limit information is either received from a digital road map, which requires reliable information about posted speed limits from GPS, or from transponders in speed limit signs (known as a ‘beacon system’). All three forms of ISA are considered to be life-saving, but mandatory ISA more so than voluntary or advisory ISA. 

AEBS
TRL recommended AEBS on all vehicles inclusive of cars, and Emergency Brake Light Display. This system automatically deploys rapidly blinking brake lamps in case of hard braking, to warn following drivers.

Support for ISA and AEBS from academics and practitioners
In 2014, TRL carried out in-depth investigations for Highways England into most of the fatal crashes on England’s strategic road network (all England’s motorways and most of its A roads), using crash investigation teams. TRL estimated how many deaths in these fatal crashes would have been prevented if certain ADAS systems had been mandated. TRL concluded more than a third (34%) of deaths studied could have been prevented if AEBS had been mandatory on all vehicles, and one in seven (14%) could have been prevented if advisory ISA had been mandatory.[ii]

Other transport academics have also estimated reductions in deaths through fitment of ISA. It has been estimated that nearly one in three fatal crashes could be prevented by Intelligent Speed Assistance.[iii]

After a successful trial on its bus fleet, Transport for London is requiring ISA to be installed in its new buses by 2017.[iv]

Lane keeping and lane change warning systems

In its report for the EC, TRL recommended an automatic system to keep vehicles in their lane. It also recognised the potential safety benefits of systems that warn of hazards when a vehicle is lane changing. 

Lane Keeping Systems
Unlike lane departure warning systems (LDW), which warn a driver of imminent lane departure, a lane keeping system (often referred to as Lane Keep Assist (LKA)) corrects the line of travel of a vehicle automatically, steering it back into the correct position. TRL said in its report[i]: "The effectiveness of LKA is considered to be greater than LDW because the system takes action to prevent the departure event, providing of course it is active and the speed of the vehicle means the system is active. Stakeholders consulted by TRL highlighted that the effectiveness of this system could be dependent on lanes being clearly marked.

Lane Change Assistance systems and blind spot detection systems
Lane change assistance systems use sensors, mostly radar, to monitor the area around a vehicle prior to a lane change manoeuvre, and issue a warning to the driver when it is unsafe to change lanes, usually due to the presence of other vehicles in the driver's intended lane of travel. More capable systems can warn of approaching vehicles, as well as vehicles that are closer and in the driver's blind spots. TRL points out that there have been "very few field tests" regarding the effectiveness of these systems in preventing crashes but there is "anecdotal evidence" that they improve driving. 

Detecting people outside the vehicle (pedestrians, cyclists and other vulnerable road users) 

There are a number of types of ADAS, most frequently available on high-end vehicles, that aim to detect people on foot and bicycles. These were also studied by TRL in its report for the EC[i]. TRL recommended one of these:

Reversing detection systems for cars
Cars are a hazard when reversing, particularly to short, crouching and slow moving people, especially very young children and the elderly. Larger cars have more driver blind spots. Solutions to aid detection of people behind cars include sensing technology that can send a warning sound to the driver. (Additionally, cameras linked to an in-car screen, and improvements to mirrors can help improve the driver's view.) The effectiveness of these systems is dependent on the driver's ability to use mirrors / cameras and respond to warning noises.  

TRL found other systems that detect people on foot and bicycles to be in various stages of development and of variable efficacy and cost-benefit at present, as listed below: 

Pedestrian and cyclist detection systems
These systems use infra-red sensing and other sensing equipment for example radar and cameras, to detect pedestrians and cyclists and then automatically brake. This is emerging technology that does not replace the need for driver attention. Systems on the market cannot detect some people in some situations. Volvo's 'Pedestrian and Cyclist Detection with Full Auto Brake' comes with a warning that it cannot identify pedestrians in certain types of clothing or carrying certain objects.[v] TRL highlighted the particular difficulty of detecting cyclists: "reliable detection of cyclists is considered to be more difficult and is some way from being ready on a mass-production scale." Combined with AEBS, these may become increasingly valuable as a safety measure, although TRL also points out that other mitigation strategies include "improvements with respect to the separation of vulnerable road users from traffic."

Night vision systems
Night vision systems use sensing equipment to detect people and then trigger a warning sound or visual display. Similar to reversing detection systems, their effectiveness is dependent on the driver's behaviour. TRL concluded: "The benefits of current systems, as well as possible negative consequences in terms of driver distraction and risk homeostasis are largely unknown. Overall there is not a consensus that night vision systems are a benefit."

Automatic cruising and emerging manoeuvring systems

Automatic/Adaptive Cruise Control (ACC) uses sensor technology to maintain a constant time gap from a moving vehicle ahead. Some systems on some vehicles have a ‘Stop & Go’ or ‘Queue Assist’ function for use in slow-moving traffic that can slow the vehicle to a complete stop and resume driving in traffic queues.

In its report[i], TRL identified research outlining benefits and risks from ACC, including that reduced driver workload could lead to inattention.

Tesla's 'Autopilot' system has an ACC system that is described by Tesla as an "assist feature" that requires a driver to "keep hands on the steering wheel at all times," and "maintain control and responsibility for your vehicle".[vi]  A driver died in the USA when their Tesla collided with a truck. Tesla said: "neither Autopilot nor the driver noticed the white side of the tractor trailer against a brightly lit sky, so the brake was not applied."[vi]

Some vehicle manufacturers at the forefront of developing driverless technology are introducing features that enable a driver to instruct their vehicle to automatically carry out a manoevre that is more complex than stop, go, or travel at a set speed or distance. This includes instructing a vehicle to: assist in changing lanes in moving traffic (detect when an adjacent lane is hazard free and move into it); self-park (detect and manoeuvre into a parking space); be summoned when parked (come out of a parking space to a driver who is outside their vehicle).  Manufacturers inclusive of Tesla and Mercedes are prominent in bringing this kind of technology on to the market. ) These rapidly emerging manoeuvring technologies were not studied by TRL in its report.[i]


End notes

[i]TRL, 2015, Benefit and Feasibility of a Range of New Technologies and Unregulated Measures in the fields of Vehicle Occupant Safety and Protection of Vulnerable Road Users
[ii] TRL, McCarthy and Barrow, Towards Zero Study on fatal collisions on the SRN during 2014, Report 2264 for Highways England
[iii] Lai, Carsten & Tate (2012) ‘How much benefit does Intelligent Speed Adaptation deliver: an analysis of its potential contribution to safety and environment’. Accident Analysis and Prevention, 48: 63-72.
[iv] Transport for London, Successful trials prove effectiveness of speed limiting technology on buses, 2016 
[v] Owner's manual online, Collision warning pedestrian detection
[vi] Tesla https://www.tesla.com/blog/tragic-loss

Last updated October 2016