可达集约束下的路径规划RRT算法

The RRT Algorithm for Autonomous Vehicle Path Planning Under Set Constraints

针对目前RRT路径规划算法存在不可达现象及安全性置信度低的问题,提出了以可达集修正约束条件的自主车辆路径规划算法。即以多车道,多车辆的复杂情况作为道路环境,使用后向可达集演变范围修正RRT算法的约束条件,采用自主车辆当前状态越靠近高等级安全区域斥力越大的思想,解决自主车辆RRT路径规划算法不可达问题,并且从理论上保证了该算法安全性的置信水平。仿真表明,该方法不仅提高了势场RRT算法的安全性,同时也可以推算自主车辆轨迹跟踪的过程中安全状态的可达范围,预测自主车辆未来一段时间间隔内安全状态。

In view of the current RRT path planning algorithm, which has the problem of non-accessibility and low safety confidence, the autonomous vehicle path planning algorithm with the constraint of up to set correction is proposed. That is to say, with the complex situation of multi-lane and multi-vehicle as the road environment, the constraints of the RRT algorithm are corrected by using the backward-up-to-set evolution range, and the idea that the current state of autonomous vehicles is closer to the high-grade safety area is used to solve the problem that the RRT path planning algorithm of autonomous vehicles is not reachable, and the confidence level of the safety of the algorithm is theoretically guaranteed. The simulation shows that this method not only improves the safety of the potential field RRT algorithm, but also can calculate the range of safety status in the process of track track tracking of autonomous vehicles and predict the safety state of autonomous vehicles in the future interval.