[1]孟 丽,程 金.基于改进RRT算法的机器人路径规划[J].工业仪表与自动化装置,2025,(04):79-83.[doi:10.19950/j.cnki.CN61-1121/TH.2025.04.014]
　MENG Li,CHENG Jin.Robot path planning based on improved RRT algorithm[J].Industrial Instrumentation & Automation,2025,(04):79-83.[doi:10.19950/j.cnki.CN61-1121/TH.2025.04.014]
点击复制

基于改进RRT算法的机器人路径规划()

《工业仪表与自动化装置》[ISSN:1000-0682/CN:61-1121/TH]

卷:
期数:
2025年04期
页码:
79-83
栏目:
出版日期:
2025-08-15

文章信息/Info

Title:
Robot path planning based on improved RRT algorithm
文章编号:
1000-0682（2025）04-0079-05
作者:
孟 丽; 程 金
济南大学 自动化与电气工程学院，山东 济南 250022
Author(s):
MENG Li;  CHENG Jin
(University of Jinan, School of Electrical Engineering, Shandong Jinan 250022, China)
关键词:
深度强化学习算法; RRT算法; 路径规划; 轨迹优化
Keywords:
deep reinforcement learning algorithm;  RRT algorithm;  path planning; trajectory optimization
分类号:
TP18 TP242
DOI:
10.19950/j.cnki.CN61-1121/TH.2025.04.014
文献标志码:
A
摘要:
该文提出了一种采用深度强化学习算法改进RRT算法的路径规划方法。首先，在随机节点的筛选中采用强化学习方法对扩展树的扩展过程加以约束，避免过多随机点；其次，构建状态价值函数评估当前状态和当前动作的优劣，从而衡量策略函数的效果。最后，采用分段二阶贝塞尔曲线进行路径平滑处理。实验结果表明，该算法在路径长度上缩短了45.4%，在规划时间上缩减了63.9%，有效节点占比提高了40%，而且在凹凸弯曲的地图环境条件下可以明显看出路径曲线的连续性增强，轨迹优化能力明显提升，验证了该算法的可行性。
Abstract:
In this paper, a path planning method using deep reinforcement learning algorithm to improve RRT algorithm is proposed. Firstly, reinforcement learning is used to restrict the expansion process of the extended tree to avoid too many random points in the selection of random nodes. Secondly, the state value function is constructed to evaluate the current state and the current action, so as to measure the effect of the strategy function. Finally, the segmented second-order Bessel curve is used for path smoothing. Experimental results show that the proposed algorithm reduces the path length by 45.4%, the planning time by 63.9%, and the proportion of effective nodes by 40%. In addition, it can be seen that the continuity of the path curve is enhanced and the trajectory optimization ability is significantly improved under the condition of the curved map environment, which verifies the feasibility of the proposed algorithm.

参考文献/References:

[1] 裴以建,杨超杰,杨亮亮.基于改进RRT*的移动机器人路径规划算法[J].计算机工程,2019,45(05):285-290+297.

[2] LIANG C, ZHAHG X, WATANABE Y, et al. Autonomous collision avoidance of unmanned surface vehicles based on improved a star and minimum course alteration algorithms [J]. Applied Ocean Research, 2021(113):102755.
[3]孙振博,王明伟,李姝,等.基于改进蚁群算法的焊接机器人路径优化[J].组合机床与自动化加工术, 2024(02): 57-60.
[4] 潘成胜,周敏,杜秀丽,等.基于改进遗传算法的装备保障效能评估指标约简[J].计算机仿真,2021,38(7):223-229.
[5] 刘澳霄,周永录,刘宏杰.基于改进人工势场法的医疗配送机器人路径规划[J].计算机应用研究, 2024, 41(03): 842-847.
[6] 张建冬,王东,马立东,等.基于改进RRT算法的移动机械臂路径规划[J].电子测量技术,2021,44(23):48-53.
[7] 田箫源,董秀成.基于改进DQN的移动机器人避障路径规划[J].中国惯性技术学报,2024,32(04):406-416.
[8] 綦慧,胡天莹.基于改进鲸鱼算法的机械臂轨迹优化[J].工业控制计算机, 2022, 35(01): 91-94.
[9] 顾海蛟,宋宇.基于改进RRT算法的垃圾收集车路径规划研究[J].长春工业大学学报,2021,42(03):285-288.
[10] 巩浩,谭向全,李佳欣,等.基于改进RRT算法的移动机器人路径规划研究[J].组合机床与自动化加工技术, 2024(1):19-24.
[11] 陈锦涛,李鸿一,任鸿儒,等.基于RRT森林算法的高层消防多无人机室内协同路径规划[J].自动化学报,2023, 49(12):2615-2626.
[12] Muhammad M, Samsani S S .Socially compliant robot navigation in crowded environment by human behavior resemblance using deep reinforcement learning[J].IEEE Robotics and Automation Letters, 2021, 6(3):5223-5230.

相似文献/References:

备注/Memo

备注/Memo:
收稿日期：2024-11-21第一作者：孟丽（1998—），女，汉族，山东聊城人，研究生，研究方向为控制工程。
更新日期/Last Update: 1900-01-01