Parking Trajectory Planning Using Multi-resolution State Roadmaps
Parking Trajectory Planning Using Multi-resolution State Roadmaps
复制标题
使用多分辨率状态路线图进行停车轨迹规划
DOI：
10.1109/tiv.2017.2769882
复制
发表时间：
2017
影响因子：
8.2
通讯作者：
Tatsuya Suzuki
中科院分区：
文献类型：
--
作者：
Yuichi Tazaki;Hiroyuki Okuda;Tatsuya Suzuki
提出了一种基于多分辨率路线图的自动泊车轨迹规划方法。建议的路线图生成算法作为输入手动设计的停车场的几何信息，包括停车位布局，障碍物配置，和一组几何特征称为指南。它构造了一个路线图，通过划分在多个分辨率的指南，直到它达到足够的覆盖范围内的一组安全的轨迹，满足车辆运动学，曲率限制，和碰撞避免。使用此路线图，一个复杂的停车轨迹组成的前向和反向运动可以计算与小的在线计算成本。所提出的方法进行了评估，在数值模拟和真实的实验。
This paper presents a trajectory planning method for automated parking that makes use of multiresolution roadmaps. The proposed roadmap generation algorithm takes as an input manually designed geometric information of a parking lot consisting of parking space layout, obstacle configuration, and a set of geometric features called guidelines. It constructs a roadmap by dividing the guidelines in multiple resolutions until it achieves enough coverage over the set of safe trajectories satisfying vehicle kinematics, curvature limit, and collision-avoidance. Using this roadmap, a complex parking trajectory composed of both forward and reverse motions can be computed with small online computation cost. The proposed method is evaluated in both numerical simulations and real experiments.