The findings of the project, which has been carried out in collaboration between the Section for Control and Communication at Linköping University and Volvo Car Corporation, are now being made public for the first time in a dissertation by Jonas Jansson, Collision Avoidance Theory-With Application to Automotive Collision Mitigation.
The system is mainly based on two components: a radar sensor that measures the distance to vehicles in front of the vehicle and a computer that makes decisions and applies the brakes if necessary. This happens less than a second before an inevitable collision-if the driver hasn’t already acknowledged the warning signals from the system. If the speed is reduced by between zero and 15 kilometers per hour, then the risk of injury is decreased by 15 percent, calculated on existing collision statistics for tailgating accidents. Since such collisions account for a third of all collisions leading to injuries, this means a total reduction of severe and fatal injuries of five percent.
Working together with Volvo, the Swedish researcher Jonas Jansson has developed two demonstration cars that have been driven on test tracks with inflatable balloons as figureheads. The results show that by jamming on the brakes the system manages to reduce the speed as expected, and in certain cases considerably more than 15 km/hr by the time of impact.
The cars have also been tested in regular traffic with the automatic braking disconnected. Normal driving in traffic is full of distracting signals that might trigger false alarms in the system, which obviously must be avoided. Complemented by another target-monitoring sensor, the system has been made robust.
The technology is based on statistical risk measurements that, on the one hand, factor in inevitable faulty measurements from the sensors and, on the other, take into account several vehicles in the vicinity of the car and their possible maneuvers. The brakes are applied only when the risk of collision is virtually 100 percent.