中文摘要：

为满足智能车在低速和高速运行时稳定和精确的轨迹跟踪,提出了一种基于几何模型的智能车轨迹跟踪算法。算法首先通过惯性导航系统的航向角参数,计算车辆纵向运动方向和轨迹跟踪点切线方向之间的切向角,再通过横向偏差角进行转向的偏差校正,实现轨迹的实时跟踪。以真实智能车在实际道路环境中对算法进行了20 km/h下的小曲率直道和大曲率路口弯道以及50 km/h下的小曲率直道的轨迹跟踪实验。实验结果表明,在不同的典型路况下,采用该算法的智能车能够实现稳定和精确的轨迹跟踪;与其他轨迹跟踪算法相比,该算法具有较好的性能。

英文摘要：

In order to achieve stable and accurate path tracking of the intelligent vehicle at low and high speed,a new geometric-based path tracking method is proposed in this paper. Firstly,this algorithm calculates the tangential angle between the direction of longitudinal motion and the tangential direction of the tracking point of the vehicle by using the parameter of azimuth angle of the inertial measurement unit mounted in intelligent vehicle. Secondly,steering deviation is corrected by deviation angle in order to realize path tracking in real time.The proposed algorithm installed in a real intelligent car has been tested on the straight road with small curvature and at the intersection with large curvature at the speed of 20 km/h in the real traffic environment. After that,the experiment has been also carried out on the straight line with small curvature at the speed of 50 km/h in the realtraffic environment. The experiments results demonstrate that the intelligent vehicle using this algorithm can realize stable and precision path tracking under different typical test conditions. Compared with other path tracking algorithms,the proposed algorithm has better performance.