人工势场引导与ACC自适应调节的RRT*机械臂路径规划

doi:10.16182/j.issn1004731x.joss.26-0240

摘要/Abstract

摘要：

针对机械臂在三维密集障碍与非结构化场景路径规划中存在的采样盲目、收敛缓慢及路径冗余等问题，提出一种融合人工势场引导与累加器自适应调节的RRT*算法(artificial potential field guidance and accumulator adaptive adjustment-RRT*，APGA-RRT*)。该算法从采样、扩展和路径优化三个方面提升整体规划性能。在采样阶段，构建基于目标距离加权与虚拟目标切换的势场引导采样策略，并结合障碍抑制与安全距离门控机制动态调节引导强度，以增强目标导向性与局部脱困能力。在扩展阶段，引入ACC碰撞反馈机制，联合调节扩展步长与重连深度，实现开阔区域快速探索与障碍密集区域精细避障的动态平衡。在路径优化阶段，采用双向冗余节点剪枝与三次B样条平滑方法，提升路径的连续性与可执行性。实验结果表明，相较于AAPF-RRT*算法，所提方法在简单三维环境下的路径长度、规划时间和采样节点数分别降低12.0%、30.2%和18.5%，在复杂三维障碍环境下分别降低6.2%、13.9%和15.6%。大量重复实验的统计分析与UR5机械臂实物验证结果共同表明，APGA-RRT*算法在路径质量、规划效率、搜索稳定性及工程可执行性方面均具有显著优势。

关键词: 机械臂路径规划, RRT*算法, 势场引导采样, 自适应扩展, 路径优化

Abstract:

To address the problems of blind sampling, slow convergence, and path redundancy in robotic manipulator path planning in cluttered and unstructured three-dimensional environments, An RRT* algorithm integrating artificial potential field guidance and accumulator-based adaptive adjustment, namely APGA-RRT* (artificial potential field guidance and accumulator adaptive adjustment-RRT*). The proposed algorithm improves the overall planning performance from three aspects: sampling, expansion, and path optimization. In the sampling stage, a potential-field-guided sampling strategy based on target-distance weighting and virtual-target switching is constructed, obstacle suppression and safe-distance gating mechanisms are incorporated to dynamically adjust the guidance intensity, thereby enhancing target directivity and local escape capability. In the expansion stage, an ACC collision feedback mechanism is introduced to jointly adjust the extension step size and rewiring depth, achieving a dynamic balance between fast exploration in open areas and precise obstacle avoidance in cluttered areas. In the path optimization stage, bidirectional redundant-node pruning and cubic B-spline smoothing are adopted to improve path continuity and executability. Experimental results show that, compared with the AAPF-RRT* algorithm, the proposed method reduces path length, planning time, and the number of sampling nodes by 12.0%, 30.2%, and 18.5%, respectively, in simple three-dimensional environments, and by 6.2%, 13.9%, and 15.6%, respectively, in complex three-dimensional obstacle environments. Extensive repeatability experiments with statistical analysis, together with UR5 manipulator physical validation results, collectively demonstrate that the APGA-RRT* algorithm exhibits significant advantages in path quality, planning efficiency, search stability, and engineering executability.

Key words: robotic manipulator path planning, RRT* algorithm, potential-field-guided sampling, adaptive expansion, path optimization

中图分类号:

TP242.2

高芳征,郭旭,孙权等 . 人工势场引导与ACC自适应调节的RRT*机械臂路径规划[J]. 系统仿真学报, 2026, 38(9): 2755-2772.

Gao Fangzheng,Guo Xu,Sun Quan,et al . RRT*-based Robotic Manipulator Path Planning with Artificial Potential Field Guidance and Accumulator-based Adaptive Adjustment[J]. Journal of System Simulation, 2026, 38(9): 2755-2772.

图/表 18

图1

表1

标准D-H参数

关节i	$α i / r a d$	$a i / m$	$d i / m$
关节1	π/2	0	0.089 15
关节2	0	-0.425 00	0
关节3	0	-0.392 25	0
关节4	π/2	0	0.109 15
关节5	-π/2	0	0.094 65
关节6	0	0	0.082 30

表1

图2

图3

图4

图5

图6

图7

图8

表2

APGA-RRT*算法核心参数设置

参数	取值
最大迭代次数 $N m a x$	5 000
独立重复实验次数 $N r u n$	100
初始扩展步长 $S 0 / m m$	15
吸引势场系数 $k a$	1.0
障碍物影响距离阈值 $d 0$ /mm	80
安全距离阈值 $d s$ /mm	45
势场引导采样步长 $ρ$ /mm	15
ACC最大限幅值 $A C C m a x$	12
ACC步长调节阈值 $θ s$	8
最小扩展步长 $S m i n / m m$	5
最大扩展步长 $S m a x / m m$	30
步长增大系数 $a$ /mm	2
步长减小系数 $b$ /mm	3
重连深度第一阈值 $θ D 1$	4
重连深度第二阈值 $θ D 2$	8
最小重连深度 $D m i n$	1
中等重连深度 $D m i d$	2
最大重连深度 $D m a x$	3

表2

图9

表3

表4

图10

表5

图11

图12

表6

参考文献 24

[1]	Waseem Saba, Adnan Muhammad, Muhammad Sajid Iqbal, et al. From Classical to Intelligent Control: Evolving Trends in Robotic Manipulator Technology[J]. Computers and Electrical Engineering, 2025, 127: 110559.
[2]	万宇航, 朱子璐, 钟春富, 等. 基于改进PPO算法的机械臂动态路径规划[J]. 系统仿真学报, 2025, 37(6): 1462-1473.
	Wan Yuhang, Zhu Zilu, Zhong Chunfu, et al. Dynamic Path Planning for Robotic Arms Based on an Improved PPO Algorithm[J]. Journal of System Simulation, 2025, 37(6): 1462-1473.
[3]	霍韩淋, 邹湘军, 陈燕, 等. 基于视觉机器人障碍点云映射避障规划及仿真[J]. 系统仿真学报, 2024, 36(9): 2149-2158.
	Huo Hanlin, Zou Xiangjun, Chen Yan, et al. Visual Robot Obstacle Avoidance Planning and Simulation Using Mapped Point Clouds[J]. Journal of System Simulation, 2024, 36(9): 2149-2158.
[4]	Lozano-Pérez Tomás, Wesley M A. An Algorithm for Planning Collision-free Paths Among Polyhedral Obstacles[J]. Communications of the ACM, 1979, 22(10): 560-570.
[5]	Ning Chunlin, Yao Zihao, Li Chao, et al. Research on a Dynamic Heuristic Algorithm for Time-optimal Path Planning of Autonomous Sailboats[J]. Ocean Engineering, 2026, 347: 123961.
[6]	Qin Xuebin, Chen Peng. Three-dimensional Path Planning for Mobile Robots Based on Hybrid Strategy and Artificial Potential Field Methods[J]. Signal Image and Video Processing, 2025, 19(13): 1076.
[7]	Lin Shiwei, Fan Angngxi, Xie Zhixuan, et al. Self-adaptive Search Algorithm for Path Planning Based on the A* Algorithm[J]. Scientific Reports, 2025, 15(1): 44901.
[8]	Lu Dandan, Wei Lo, Yang Chunming. PerceptPath: Travel Path Planning and Display Algorithm Optimized by Dijkstra's Algorithm and a Perceptual Visual Attention Model[J]. Signal Image and Video Processing, 2025, 19(17): 1425.
[9]	Jiang Jiefeng, You Yong, Shao Youtao, et al. Robot Path Planning Based on Improved PRM for Wing-box Internal Assembly[J]. Machines, 2025, 13(10): 952.
[10]	Kavraki L E, Svestka P, Latombe J C, et al. Probabilistic Roadmaps for Path Planning in High-dimensional Configuration Spaces[J]. IEEE Transactions on Robotics and Automation, 1996, 12(4): 566-580.
[11]	Lavalle S. Rapidly-exploring Random Trees: A New tool for Path Planning[J]. Research Report, 1998.
[12]	Wang Lihui, Wang Zhuoxuan, Ying Zehua, et al. A Path Planning Framework Based on an Improved Weighted Heuristic RRT and Optimization Strategy[J]. IEEE Transactions on Intelligent Vehicles, 2024, 9(1): 1941-1952.
[13]	Wang Yan, Jiang Wensong, Luo Zai, et al. Path Planning of a 6-DOF Measuring Robot with a Direction Guidance RRT Method[J]. Expert Systems with Applications, 2024, 238: 122057.
[14]	齐本胜, 李岩, 苗红霞, 等. 基于改进启发式RRT的AUV路径规划[J]. 系统仿真学报, 2025, 37(1): 245-256.
	Qi Bensheng, Li Yan, Miao Hongxia, et al. Research on Path Planning Method for Autonomous Underwater Vehicles Based on Improved Informed RRT[J]. Journal of System Simulation, 2025, 37(1): 245-256.
[15]	陈志梅, 李敏, 邵雪卷, 等. 基于改进RRT算法的桥式起重机避障路径规划[J]. 系统仿真学报, 2021, 33(8): 1832-1838.
	Chen Zhimei, Li Min, Shao Xuejuan, et al. Obstacle Avoidance Path Planning of Bridge Crane Based on Improved RRT Algorithm[J]. Journal of System Simulation, 2021, 33(8): 1832-1838.
[16]	蔡浩, 诸云, 于明爽, 等. 基于改进RRT算法的智慧小区物料小车路径规划[J]. 无人系统技术, 2024, 7(2): 92-100.
	Cai Hao, Zhu Yun, Yu Mingshuang, et al. Path Planning of Material Trolleys in Smart Communities Based on Improved RRT Algorithm[J]. Unmanned Systems Technology, 2024, 7(2): 92-100.
[17]	Cao Xiaoman, Zou Xiangjun, Jia Chunyang, et al. RRT-based Path Planning for an Intelligent Litchi-picking Manipulator[J]. Computers and Electronics in Agriculture, 2019, 156: 105-118.
[18]	Huang Anxiang, Zhou Meng, Liu Mengfei, et al. Adaptive Path Planning for Robotic Winter Jujube Harvesting Using an Improved RRT-connect Algorithm[J]. Agriculture, 2026, 16(1): 47.
[19]	Karaman S, Frazzoli E. Incremental Sampling-based Algorithms for Optimal Motion Planning[M]//Matsuoka Y, Hugh Durrant-Whyte, José Neira. Robotics: Science and Systems VI. Cambridge: The MIT Press, 2011.
[20]	Wang Xuewu, Wei Jianbin, Zhou Xin, et al. AEB-RRT: An Adaptive Extension Bidirectional RRT Algorithm[J]. Autonomous Robots, 2022, 46(6): 685-704.
[21]	Zhang Zhiwei, Jia Yunwei, Su Qiqi, et al. ATS-RRT: An Improved RRT Algorithm Based on Alternative Paths and Triangular Area Sampling[J]. Advanced Robotics, 2023, 37(10): 605-620.
[22]	陈治润, 袁杰, 加尔肯别克, 等. 融合RRT*和APF的机械臂自适应路径规划[J]. 系统仿真学报, 2026, 38(6): 1669-1683.
	Chen Zhirun, Yuan Jie, Jia Erkenbieke, et al. Adaptive Path Planning for Robotic Arms Integrating RRT* and APF[J]. Journal of System Simulation, 2026, 38(6): 1669-1683.
[23]	Jeong I B, Lee Seung-Jae, Kim Jong-Hwan. Quick-RRT: Triangular Inequality-based Implementation of RRT with Improved Initial Solution and Convergence Rate[J]. Expert Systems with Applications, 2019, 123: 82-90.
[24]	Zhang Liding, Cai Kuanqi, Sun Zewei, et al. Motion Planning for Robotics: A Review for Sampling-based Planners[J]. Biomimetic Intelligence and Robotics, 2025, 5(1): 100207.

算法	节点数	路径代价	耗时/s
RRT	68.42±10.36	162.15±18.42	4.82±0.63
RRT*	61.28±8.74	148.73±14.55	5.14±0.58
AAPF-RRT*	43.13±5.21	127.25±8.36	2.23±0.31
本文	28.15±3.68	123.08±5.92	1.12±0.18

算法	路径长度/mm	时间/s	节点个数
RRT	1 275.13±74.26	0.52±0.07	587±58.40
RRT*	1 223.62±63.80	0.48±0.06	502±48.70
AAPF-RRT*	1 160.51±45.35	0.43±0.05	470±38.20
本文	1 020.81±31.68	0.30±0.03	383±25.60

算法	路径长度/mm	时间/s	节点个数
RRT	1 452.10±98.36	1.85±0.22	620±70.60
RRT*	1 380.45±86.72	2.12±0.25	580±62.30
AAPF-RRT*	1 290.33±64.18	1.65±0.18	450±45.70
本文	1 210.56±42.55	1.42±0.13	380±32.90

算法	重复次数	T_P/s	e_a/mm	S_r/%	d_min/mm
RRT*	50	2.23	6.28	86	31.5
APGA-RRT*	50	1.38	4.72	98	42.6