基于EKF的自动泊车系统位姿估计算法设计

利用车辆2个后轮轮速信号和方向盘转角信号,基于扩展卡尔曼滤波技术设计一种车辆位姿估计算法,并在veDYNA中仿真试验。仿真结果表明,算法的估计精度比较理想,可满足泊车系统对自身车辆定位的需求。以定位精度2 cm（1σ）的GPS信号为参考,实车试验结果表明,位置估计误差控制在3%以内,此精度的估计结果可以为泊车系统提供车辆定位信息,为增强泊车安全性奠定基础。

Design for Position Estimation Algorithm of Parking Assistance System Based on EKF

In this article,the signals of two rear-wheel speed sensors and a steering wheel sensor are used to design the position estimation algorithm of vehicle based on an extended Kalman filter technology,which is stimulated in the veDYNA test. The testing result shows that the estimation accuracy of the algorithm can reach an acceptable level and that the algorithm can meet the needs of vehicle location by its parking system. With reference to the GPS signal whose position accuracy reaches 2 cm（ 1σ）,the real car experiment result shows that the position estimation error can be controlled within 3% and the estimation result of accuracy can provide the vehicle location information for the parking system,laying the foundation for the safety of parking.