基于多卡尔曼滤波器的自适应传感器融合

准确估车前目标的位置和状态是车辆安全系统的一个重要组成部分,利用多传感器可以提高对目标测量的可靠性。给出了一种用于汽车防碰撞系统中估计车前目标位置的多传感器融合算法,根据车辆与目标之间的一步预测距离,从多个预置的卡尔曼滤波器中选择一个合适的滤波器来实时处理传感器的量测数据。该算法不仅可以有效地处理具有不同测量范围的传感器数据,同时还可以通过滤波器切换的方式,处理具不同数据速率的传感器数据。Monte Carlo数字仿真证明了该算法的有效性。

Adaptable sensor fusion based on multiple Kalman filters

To estimate accurately the position and state in front of a vehicle is an important section for vehicle safety system, and using multi-sensor can increase the reliability of measurement. A multi-sensor fusion strategy to estimate the position of targets in front of a vehicle is presented, which is used in a collision avoidance system. The multi-sensor data is fused with a Kalman filter, which is selected from a bank of Kalman filters according the prediction distance between vehicles and targets. The algorithm can handle data of sensors, which have different measure range, and can work asynchronously with different sensor data rates through a filter switching process. It is proved that the algorithm is effective by Monte-Carlo simulation.