Driving a safety 'brake-through'

  • 10-Jan-2013 05:15 EST
Thatcham 12-12 - Mazda CX-5 steering robot 4.jpg

A steering robot is fitted to a Mazda CX-5 by Thatcham Research for accident assessment using autonomous emergency braking (AEB).

Autonomous emergency braking (AEB) is set to eclipse the airbag, antilock brakes, and even electronic stability control as one of the most significant developments in automotive safety, believes Andrew Miller, Research Director at Thatcham Research in the U.K. AEB is expected to be standard equipment on all new cars in Europe within a decade: “This means that car manufacturers are facing not only complex design and engineering challenges to integrate the technology but must meet standards that have yet to be formally agreed.”

In Europe, the effort to get AEB technology quickly and widely adopted by vehicle manufacturers is being driven by the safety organization EuroNCAP. From 2014, unless a vehicle achieves maximum—or very high—marks in every category, it will become practically impossible for it to reach the top 5-Star EuroNCAP rating if it does not feature an AEB system.

As a leading member of the Research Council for Automobile Repair (RCAR) group of international research centers and chair of the AEB group, Thatcham is heading the development work in AEB’s place within EuroNCAP. As recently reported by AEI, EuroNCAP is placing emphasis on AEB’s role in making very significant advances in vehicle safety.

Ramping up AEB

European OEMs and relevant suppliers have comprehensive AEB programs in place or in development. Latest additions include the 2013MY Volkswagen Golf Mk. VII, which includes City Emergency Braking as standard on many versions, and the new Ford Fiesta, which achieves a segment first with Active City Stop. Volvo introduced AEB as City Safety Braking in 2007 on the XC60, and it is now fitted to a majority of its vehicles. Although it has not been confirmed, Volvo's AEB is likely to be extended to cover reversing and to reduce risk of accidents involving crossing traffic.

Mercedes-Benz is also expected to cascade full AEB technology through its car range in the near term. Its current radar-based cruise control-linked Distronic system is capable of braking the car to a standstill.

AEB systems use laser, radar, and cameras, or combinations of these technologies, to detect slow moving, rapidly decelerating, or stationary vehicles ahead, warning the driver and applying appropriate deceleration through autonomous braking, either avoiding a collision altogether or mitigating the severity. The technology has the potential to achieve similar results when pedestrians—including small children—are involved. Data generated by Thatcham Research indicates that in the U.K. alone, when widely adopted, AEB has the potential to prevent some 2700 pedestrian casualties annually.

But the challenge for motor manufacturers and suppliers is that development of AEB is extremely rapid, stated Miller, with systems integration complex but vital. R&D is seeing new technology, such as stereoscopic cameras with enhanced resolution and improved processing power constantly emerging. Some existing technology may become quickly obsolete: “These factors—advances and resultant obsolescence—make engineering a system at the right performance/price point extremely difficult, especially as industry testing standards have yet to be fully defined.”

Real-world data from the U.S. (Insurance Institute for Highway Safety) found a 27% reduction in third-party crashes involving the Volvo XC60, one of the first production vehicles to be fitted with the technology. Using data from this and other real-world studies, combined with first-hand research experience, Thatcham has worked with the Association of British Insurers (ABI) to implement a favorable change in the insurance group rating for vehicles with AEB as standard equipment. This means potentially delivering reduced insurance premiums to their end-users.

Target practice

Thatcham has now developed its own test scenarios and equipment, based on the AEB Group’s work. Says Miller: “The AEB Group is probably doing the most innovative work in this area, and it looks as though its proposals will form the framework for EuroNCAP’s rating system.

“AEB will make its first appearance in EuroNCAP’s 2014 ratings. This will involve a new protocol for assessment of systems for low speed (up to 50 km/h), rear-end longitudinal car-to-car collision avoidance or mitigation—the City Test.”

Also next year, an AEB Inter-Urban Test (IUT) for mid-to-high speed (up to 80 km/h) rear-end collision avoidance or mitigation will be introduced, which will assess the AEB function and also the benefits of forward collision warning (FCW). In 2016, AEB systems will be assessed for their performance in pedestrian collisions.

Miller revealed that Thatcham has already worked on the physical test targets for car-to-car rear-end impacts and car-to-pedestrian collisions. The resultant targets are lightweight and reusable but designed to accurately represent the necessary sensor attributes of real cars and of humans.

The car target is a robust inflatable structure that is capable of being impacted at speed without suffering damage itself, nor damaging the test vehicle. It is representative of the shape and size of a typical European vehicle and includes key features such as license plate, lights, and rear windshield. It also contains elements that provide the sensors on the car under test with the correct visual properties and radar reflectivity. The City Test involves a stationary target only, but for the IUT there are also moving vehicle situations, with the target towed on a framework by a vehicle, enabling realistic but safe testing at speeds of up to 80 km/h.

Pedestrian targets, which accurately represent the proportions of an adult male and a child—particularly difficult—are currently under development. Thatcham is also involved in research to identify the key attributes of pedestrians to ensure that the test devices represent real people and are sufficiently robust to allow repeatable testing. The dummy will be mounted on a mobile platform and propelled across the path of the test vehicle, which is moving at speeds up to 50 km/h, in varying scenarios where the pedestrian’s approach is either visible or obscured by a parked vehicle.

To maximize repeatability during drive-cycle testing, driving robots and vehicle dynamics instrumentation are used to control the vehicle. Says Miller: “These have been specified and sourced by Thatcham and include steering and accelerator robots working in a feedback loop to regulate the approach path and speed of both the car target and the vehicle under test. Braking robots control the target deceleration and react to the FCW in the test vehicle.”

Differentially GPS-corrected inertial measurement units are installed in the target and vehicle under test to measure their individual dynamics. A communications system between the vehicles generates real-time measurements of relative speed and position.

Overcoming differences

To ensure that performance comparisons between different cars and AEB systems are both accurate and repeatable, permitted tolerances are designed to be absolutely minimal. In a car-to-car rear-end test in which the target is towed along a straight path and decelerated to a halt, target tolerances are: speed ±1.0 km/h; lateral position ±0.10 m; yaw rate ±1.0 °/s; and the deceleration must ramp up and maintain within a specified corridor. The test vehicle approach tolerances are: nominal test speed +1.0 km/h; steering wheel velocity ±15 °/s; accelerator pedal position ±2%; lateral position ±0.10 m; yaw rate ±1.0°/s; and headway +1.0 m.

Matthew Avery, Thatcham’s Head of Research, explains: “There are substantial performance differences in the AEB systems from various manufacturers and, in addition, the technologies are advancing quickly. Our goal is to devise testing scenarios that are ‘technology blind,’ so that we look purely at performance. To achieve this, our tests have to be extremely accurate and repeatable. Before the test cycle begins, we have to ensure that each vehicle is prepared, fueled, calibrated, and put through a conditioning drive-cycle in exactly the same way.”

EuroNCAP’s City and IUTs have already been finalized, and it is expected that the pedestrian tests, still being formalized by the organization’s technical development assessment team, will use Thatcham’s research work and expertise as a key input to future collision avoidance and collision mitigation capability.

The City Test sees the vehicle under test approach the stationary target at speeds up to 50 km/h in 10 km/h increments. If an impact occurs, the test is repeated at a 5 km/h lower speed and then in 5-km/h increments to establish the performance curve. Full points are awarded by EuroNCAP for AEB collision avoidance, and in the case of mitigation, points are awarded proportionally to the speed reduction achieved. FCW is not considered.

The more complex IUT involves assessing the performance of the AEB and FCW systems in three test scenarios: with a stationary target, a slower moving target, and a decelerating target. The test against the stationary target assesses the performance of the FCW system over the speed range from 30-80 km/h by programming the robot to react to the collision warning and, after a delay of 1.2 s, brake the vehicle to simulate a real driver’s crash-avoidance action.

For the slower moving target test, the target moves at 20 km/h and the vehicle under test approaches at 50-70 km/h, giving 30-50 km/h speed differentials. In the decelerating target tests, both the target and vehicle under test drive in unison at 50 km/h with headways of either 12 or 40 m. The target vehicle then decelerates at either 2 m/s², typical of an everyday driving maneuver, or at 6 m/s² in an emergency. The performance of the AEB and FCW systems are assessed in the slower moving and decelerating target tests. If a crash occurs, the test is repeated at a 5 km/h lower speed and then in 5 km/h increments to establish the performance curve, with points awarded similarly to the City Test.

The pedestrian assessment—still under development—is currently performed with the vehicle under test moving at up to 60 km/h in 10-km/h increments, again with 5-km/h steps being included to establish the performance curve. When the test is formally established, it will involve both obscured and unobscured pedestrian targets being moved into the path of the vehicle from the nearside.

Within the next decade, the incorporation of AEB technologies on cars of all classes will become the norm and another step will be taken toward what was once seen by many as a fanciful dream: accident-free road travel.

Share
HTML for Linking to Page
Page URL
Grade
Rate It
3.35 Avg. Rating

Read More Articles On

2017-03-04
Euro NCAP will establish a separate category for autonomous vehicles, but there is not likely to be one for cars that are claimed to protect all occupants from serious injury or death.
2017-01-07
Motion sickness in autonomous vehicles is the new "elephant in the room," with engineers suffering during autonomous-driving simulator runs. Researchers are working to solve this nasty issue.
2017-01-13
Range anxiety is not just affecting EV drivers on the road; it is also a significant hurdle for Formula E teams on the track. U.K. simulator specialist rFpro says its technology can help.
2017-01-09
CEO John Krafcik told the Automobili-D audience in Detroit that Waymo is building its own hardware suite with a fully top-to-bottom, full-stack approach. The classic auto industry vertical integration includes all vision sensors, radars and LiDAR, along with related “AI compute” artificial-intelligence platform.

Related Items

Technical Paper / Journal Article
2010-09-28
Training / Education
2010-03-15
Training / Education
2013-02-20
Technical Paper / Journal Article
2010-10-19