Results of a precrash application based on Laserscanner and short range radars

Abstract

In this paper, we present a vehicle safety application based on data gathered by a Laserscanner and 2 short range Radars that recognizes unavoidable collisions with stationary objects before they take place in order to trigger restraint systems. Two different software modules are compared that perform the processing of raw data and deliver a description of the vehicle's environment. A comprehensive experimental evaluation based on relevant crash and non-crash scenarios is presented.

360°-Environment sensing and signal processing for an automotive pre-crash application

“…for object classification and improved measurement accuracy, optical sensors like mono-and stereo cameras as well as laser scanners have been studied. Multi-sensor fusion systems using laser scanners and radar sensors are shown in [8] and [9]. The problem of short reaction times in pre-crash scenarios has been addressed by the author earlier in [10], where a Kalman filter is used in scenarios with short filter settling times, which outperforms a naive approach working on only the estimation means.…”