Chen, S., Huang, L., Bai, J., Jiang, H. et al., "Multi-Sensor Information Fusion Algorithm with Central Level Architecture for Intelligent Vehicle Environmental Perception System," SAE Technical Paper 2016-01-1894, 2016, doi:10.4271/2016-01-1894.
Intelligent vehicles can improve traffic safety and reduce damage caused by traffic accidents. Environmental perception system is the core of the intelligent vehicle which detects vehicles and pedestrians around the ego host-vehicle by using vehicle environmental perception sensors. Environmental perception system with the multi-sensor information fusion algorithm can utilize the advantages of each environmental perception sensor and detects targets with higher detection probability and precision. Most of the published papers are based on the sensor level fusion architecture which is not stable and robust in detecting target. This paper presents a multi-sensor fusion algorithm with central level architecture, which can improve the target detection probability compare to these with the sensor level fusion architecture. The fusion algorithm with central level architecture detects the target based on targets’ longitudinal distance, lateral distance and speed measured by the environmental perception sensor. At first, each environmental perception sensor measures the distance of the closest targets. Then, multi-sensor information fusion algorithm detects the nearest target in the whole region. Finally, the target is tracked by Kalman filter. The performance of the multi-sensor information fusion algorithm is evaluated by using a test vehicle. Three different types of the vehicle environmental perception sensors are mounted in the front of an intelligent vehicle, which are a monocular camera, a long-range and a short-range millimeter wave radar. Multi-sensor information fusion algorithm implemented in AutoBox to detect the targets in front of the intelligent vehicle. The experiments show that this multi-sensor information fusion algorithm with central level architecture can improve target detection probability.