Lasers Test Accuracy of Car Crash Systems

Michael A. Greenwood

For many drivers, it is their last best hope. The crash-warning systems that are increasingly common in higher-end cars sound an alarm if a collision is imminent, providing a few precious seconds for the operator to slam on the brakes or quickly change course to avoid an accident.

A laser-based system to test the effectiveness of crash-warning systems in cars and trucks has been developed by NIST. Courtesy of NIST.

The reliability of these systems is crucial, so the National Institute of Standards and Technology (NIST) in Gaithersburg, Md., wants to find out whether there is room for improvement.

The government agency devised a laser-based independent measurement system and is helping the US Department of Transportation to assess the performance of the collision-warning systems. Researchers in the automotive industry will be able to use the data to make improvements to warning systems currently under development.

Most collision-warning devices that are now in use rely on radar to detect objects within proximity of the front or rear of the vehicle, and they depend on cameras to detect whether a vehicle is drifting sideways.

The measurement system developed by NIST uses commercial laser scanners, with one mounted on each of the outside front corners of the vehicle being tested. The scanners emit pulses from four infrared laser diodes and measure the return time to gauge the distances of nearby objects. A 10-Hz rotating mirror sweeps the laser pulses around a 300° horizontal field of view. The four laser diodes are oriented at 1° increments to provide a 4° vertical field of view that enables the scanner to pick up obstacles when the vehicle pitches.

Additionally, the system uses four commercially available CCD cameras (each with IR LEDs for nighttime use) to record what is happening at various positions around the car. Two of the cameras are mounted with the laser scanners and are pointed to the side of the vehicle to record the distance to the lane boundaries. The third camera is mounted on top of the hood and is pointed forward to record the road ahead. A fourth camera, along with a microphone, is mounted inside the car to record the driver’s actions during tests.

In tests conducted so far in East Liberty, Ohio, and in Dundee, Mich., NIST used its system to measure ranges to targets at the time of warning to determine whether targets were in-path or out-of-path and to measure delays in warning times. The crash-warning systems performed well in most instances. However, the NIST system did find some delays in issuing warnings for high-closure-rate situations. Detection of vehicles in curves or during lane changes also was flagged by the NIST system.

The laser-based system can be mounted on cars or trucks with trailers and requires no modifications or connections to the warning system being tested. The NIST system can detect an object to within 33 in. at a distance of 197 ft and at speeds of up to 56 mph.