引入必经点约束的智能汽车全局路径规划

Global Path Planning of Intelligent Vehicle with Must-Pass Nodes

目前对于智能车全局路径规划的研究多数只针对从起点到终点的情况。针对该问题,本文中融合改进A*和模拟退火算法,设计了一种引入必经点约束的全局路径规划算法。首先,基于A*算法计算关键节点间的最短路径并保存。然后,基于启发式算法中的模拟退火算法对过必经节点的全局路径进行迭代随机优化。接着,基于真实高精度地图对算法的有效性以及时间复杂度进行实验分析。结果表明,设计的算法在求解质量和求解速度方面都有较好的表现。最后,通过实车实验,进一步验证了算法的有效性和适应性。

At present,most of the research on the global path planning of intelligent vehicles only focuses on the situation from the beginning to the end.To solve this problem,this paper combines the improved A* and simulated annealing algorithm,and designs a global path planning algorithm that introduces in the constraint of mustpass nodes.Firstly,the shortest path between key nodes is calculated and saved based on the A* algorithm.Then,based on the simulated annealing algorithm in the heuristic algorithm,the global path through the must-pass nodes is iteratively and randomly optimized.Then,the validity and time complexity of the algorithm are tested and analyzed based on the real high-precision map.The results show that the designed algorithm has good performance in solving speed and solving quality.Finally,the effectiveness and adaptability of the algorithm are further verified by the real vehicle test.