摘要
传统的快速扩展随机树(RRT)算法具有探索能力强和收敛速度快等显著优点,但是由于算法采用随机采样作为路径搜寻手段,导致RRT算法的规划性能十分不稳定,且在复杂环境下尤为明显。针对这个问题,借鉴启发式算法的思想,提出动态规避算法,通过引入启发式约束采样策略,适当增加算法的指向性,并且使规划的航迹更符合无人机的飞行轨迹,然后采用动态步长规避策略,改善算法的探索和避障能力。再对规划成功的路径进行优化,获得相对平滑的航迹。最后通过MATLAB仿真对比实验对算法进行分析,对比实验的结果表明算法在障碍物密集的区域内,可以有效增强算法的收敛速度和稳定性,可见动态规避算法可以有效改善算法的稳定性,而且还拥有相对优秀的避障性能。
The traditional Rapidly-exploring Random Tree( RRT) algorithm has significant advantages such as strong exploration ability and fast convergence speed,However,because the algorithm uses random sampling as a path search method,the planning performance of the RRT algorithm is very unstable,and it’s especially obvious in complex environments. To solve this problem,drawing on the idea of heuristic algorithms,Proposed dynamic avoidance algorithm,introducing heuristic constraint sampling strategy,appropriately increase the directivity of the algorithm,and make the planned trajectory more in line with the flight trajectory of the drone,then adopt dynamic step avoidance strategy to improve algorithm exploration and obstacle avoidance ability,then optimize the planned route to obtain a relatively smooth track,finally,the algorithm is analyzed through MATLAB simulation and comparison experiment,the results of comparative experiments show that the algorithm can effectively enhance the convergence speed and stability of the algorithm in an area with dense obstacles,it can be seen that the dynamic avoidance algorithm can effectively improve the stability of the algorithm,and it also has relatively excellent obstacle avoidance performance.
作者
陈良剑
赵文龙
娄嘉骏
CHEN Liangjian;ZHAO Wenlong;LU Jiajun(School of Infomation Engineering,NCHU,Nanchang 330063;Ningbo Water Meter(Group)Co.,Itd.,Ningbo 315032)
出处
《现代计算机》
2021年第22期72-76,80,共6页
Modern Computer
基金
江西省研究生创新专项资金项目(No.YC2019018)。
关键词
航迹规划
快速扩展随机树算法
无人机性能约束
启发式算法
动态规避算法
Trajectory Planning
Rapidly-Exploring Random Tree Algorithm
UAV Performance Constraints
Heuristic Algorithm
Dynamic Avoidance Al-gorithm
