Truck active reversing control strategy based on modified particle filter and multi-sensors environment perception

Abstract

Due to the limit of the drivers' sight sheltered by the large body and the complexity of driving, heavy-duty trucks usually cause serious reversing traffic accidents. In this study, the authors present a new strategy to settle this problem by using the multi-sensors system, modified unscented particle filter, and an active safety control strategy. First, multi-sensors including binocular cameras, ultrasonic sensors, and laser rangefinders are used to perceive the environmental information to help drivers to decrease the blind area and obtain the distance between obstacles and the truck. Second, a federal information fusion mechanism based on the unscented Kalman filter is used to fuse the information so that the proposed system can eliminate spurious data and improve the robustness. Next, a framework combining low-rank representation with unscented particle filter that is used for target tracking is proposed. Third, a speed control method is introduced to realise the active speed control of a truck. Lastly, the experiment and the results prove that the proposed algorithm has higher accuracy than others. Moreover, the proposed system can not only help drivers decrease the burden while reversing but also increase the active safety ability to reverse, which has great significance for heavy-duty trucks.