本文为加拿大多伦多大学（作者：George Tin Lam Lau）的硕士论文，共99页。
本文提出了一种无人驾驶边境巡逻车的在线路径规划算法。在这个追逃游戏中，无人驾驶汽车必须在多个入侵者到达他们安全的目标安全屋之前逮捕他们。问题的公式化是基于Isaacs的目标防御问题，但扩展到多个逃跑者的情况。所提出的路径规划方法是基于快速探索随机树（RRT）的，能够在数秒内产生轨迹，捕获2或3个逃逸者。仿真结果表明，所得到的轨迹接近基于非线性规划的数值最优控制解所产生的最优解。在无人地面车辆上进行了实验，验证了所提出的在线路径规划算法在实际应用中的可行性。
This thesis presents an online pathplanning algorithm developed for unmanned vehicles in charge of autonomousborder patrol. In this Pursuit-Evasion game, the unmanned vehicle is requiredto capture multiple trespassers on its own before any of them reach a targetsafe house where they are safe from capture. The problem formulation is basedon Isaacs’ Target Guarding problem, but extended to the case of multipleevaders. The proposed path planning method is based on Rapidly-exploring randomtrees (RRT) and is capable of producing trajectories within several seconds tocapture 2 or 3 evaders. Simulations are carried out to demonstrate that theresulting trajectories approach the optimal solution produced by a nonlinearprogramming-based numerical optimal control solver. Experiments are alsoconducted on unmanned ground vehicles to show the feasibility of implementingthe proposed online path planning algorithm on physical applications.
- 引言
- 文献回顾
- 问题建模
- 算法
- 仿真
- 实验
- 结论
附录A SPSE目标边界
附录B C++ SP2E算法
附录C 用于速度预测的卡尔曼滤波
更多精彩文章请关注公众号：