基于样条理论的泊车轨迹生成与仿真研究

为了减少泊车时间并且达到准确泊车的目的,通过分析泊车轨迹曲线的设计要求并综合考虑B样条曲线特点,提出基于多约束函数的泊车方法。研究泊车过程中潜在碰撞点,以泊车轨迹连续性为目标,建立相对应的约束函数。然后利用matlab软件非线性约束优化函数求得轨迹控制点,得出轨迹曲线。将得到的轨迹曲线进行存储,最终实现对泊车轨迹库的构建。根据泊车环境、车辆的初始位置进行决策,选取轨迹库中合适的泊车轨迹曲线,最终实时生成泊车所需要的运动轨迹。仿真结果表明,这种方法响应速度快、实时性强、可靠性高,生成的轨迹曲线符合泊车环境,减少了泊车用时。

The Research on Generation and Simulation of Parking Trajectory Based on Spline Theory

In order to reduce the parking time and park more accurately, parking trajectory was analyzed and an approach for trajectory planning was proposed on the basis of the B-spline theory, and the parking path was realized by solving multi- constraints functions. Potential collision points were analyzed, and the constraint equations, the aim of which is to keep the continuity of the parking trajectory, were established accordingly. The trajectory curve can be depicted by obtaining the control points utilizing nonlinear constrained optimization function in matlab. The result curve was stored in order to construct trajectory library. According to parking environment and the initial position of the vehicle to make decisions, the appropriate trajectory curve in the library was selected, and ultimately the trajectory curve which meets parking requirement can be generated real- time. Simulation results have shown that the method was characterized by high response speed, strong real- timeliness and good reliability, the generated curve was easy to control. The trajectory curve met parking environment and parking time was shorter apparently.