Please wait a minute...
浙江大学学报(工学版)
Journal of ZheJiang University (Engineering Science)  2024, Vol. 58 Issue (4): 779-789    DOI: 10.3785/j.issn.1008-973X.2024.04.013
    
Dual-mechanism tangential obstacle avoidance of autonomous robots in dynamic environment
Yiming ZHANG1(),Wenguang YAO2,Haijin CHEN1,*()
1. Jiangsu Provincial Key Laboratory of Application-Specific Integrated Circuit Design, Nantong University, Nantong 226001, China
2. Atekon Technology Limited Company, Nanjing 210012, China
Download: HTML     PDF(3362KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

Aiming at the dynamic randomness of robot working environment, an improved artificial potential field method based on dual-mechanism tangential obstacle avoidance was proposed. Aiming at the local minimum trap of the traditional artificial potential field method, a static obstacle avoidance mechanism was proposed. The map was preprocessed before planning, local minimum points were predicted and obstacles were divided into connected and non-connected, and the static tangential obstacle avoidance was realized by combining the tangential obstacle avoidance. Based on the static obstacle avoidance mechanism, a dynamic obstacle avoidance mechanism was proposed for dynamic obstacles. By adjusting the collision risk coefficient in real time and selecting the obstacle with the largest coefficient for obstacle avoidance angle compensation, the dynamic tangential obstacle avoidance was realized. By state decision making, the static and dynamic tangential obstacle avoidance mechanism and the global path planning were integrated to realize the hybrid planning and design. Simulation and omnidirectional mobile platform was designed, and the proposed method was verified and tested. Results showed that the proposed method effectively resolved the local minimum trap of the traditional artificial potential field method under different complex environments, and realized fast autonomous obstacle avoidance under dynamic environments. Comparing the average obstacle avoidance time of three methods to avoid different types of obstacles, the proposed method was 55% better than the dynamic window approach (DWA) and 40% better than the time elastic band (TEB). Comparing the average navigation time of three methods for navigating maps of different complexity, the proposed method was 39% better than DWA and 22% better than TEB.

Key words： dynamic environment      artificial potential field method      local minimum trap      dual-mechanism tangential obstacle avoidance      status decision      hybrid planning     
Received: 29 June 2023      Published: 27 March 2024
CLC:  TP 242.6  
Fund:  江苏省科技成果转化专项资金资助项目（BA2022001）.
Corresponding Authors: Haijin CHEN     E-mail: 1242208320@qq.com;chen.hj@ntu.edu.cn
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Yiming ZHANG
Wenguang YAO
Haijin CHEN
Cite this article:

Yiming ZHANG,Wenguang YAO,Haijin CHEN. Dual-mechanism tangential obstacle avoidance of autonomous robots in dynamic environment. Journal of ZheJiang University (Engineering Science), 2024, 58(4): 779-789.

URL:

https://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2024.04.013     OR     https://www.zjujournals.com/eng/Y2024/V58/I4/779


动态环境下自主机器人的双机制切向避障

针对机器人工作环境的动态随机性，提出面向双机制切向避障的改进人工势场法. 针对传统人工势场法的局部极小值陷阱问题，提出静态避障机制，在规划开始前对地图进行预处理，预测局部极小值点并将障碍物分成连通与非连通障碍物，结合切向避障实现静态切向避障. 以静态避障机制为基础，针对动态障碍物，提出动态避障机制，通过实时调整碰撞风险系数并选择系数最大的障碍物进行避障角补偿，实现动态切向避障. 通过状态决策统筹静态、动态切向避障机制与全局路径规划，实现混合规划与设计. 设计仿真和全向移动平台，对所提方法进行验证测试. 结果表明，所提方法在不同环境复杂下均有效解决了传统人工势场法的局部极小值陷阱问题，实现了动态环境下快速自主避障. 对比3种方法避开不同类型障碍物的平均耗时，所提方法比动态窗口法（DWA）提升55%，比时间弹性带法（TEB）提升40%；对比3种方法导航不同复杂度地图的平均耗时，所提方法比DWA提升39%，比TEB提升22%.


关键词： 动态环境,  人工势场法,  局部极小值陷阱,  双机制切向避障,  状态决策,  混合规划 
Fig.1 Local minimum trap of traditional artificial potential field method
Fig.2 Grid map pyramid scaling
Fig.3 Search areas with dense obstacles
Fig.4 Direction determination of repulsive force for connected obstacles
Fig.5 Importance evaluation of non-connected obstacles
Fig.6 Influence range division of repulsive force for non-connected obstacles
Fig.7 Static tangential obstacle avoidance for non-connected obstacles
Fig.8 Analysis of resultant force direction after repulsive force adjustment
Fig.9 Flowchart of static tangential obstacle avoidance
Fig.10 Compensation principle of compensation angle for dynamic tangential obstacle avoidance
Fig.11 Block diagram of mixed path planning module system
Fig.12 Decision process of mixed path planning state
Fig.13 Flowchart of mixed path planning algorithms
输入参数语义输入参数语义输出参数语义
$d \in \left[ {{r_{{\text{obs}}}},{{\overline r }_{\min }}} \right)$DN（近）$ \omega \in \left[-{20}^{\circ },{20}^{\circ }\right) $ZJ（零）${\text{output}} \in \left[ {0,2} \right)$ZE（紧急制动）
$d \in \left[ {{{\overline r }_{\min }},{{\overline r }_{\max }}} \right)$DM（适中）$ \omega \in \left[{20}^{\circ },{65}^{\circ }\right) $PS（正小）${\text{output}} \in \left[ {2,4} \right)$APF（人工势场法）
$d \in \left[ {{{\overline r }_{\max }},{\rho _{\text{o}}}} \right)$DF（远）$ \omega \in \left[{65}^{\circ },{90}^{\circ }\right] $PB（正大）${\text{output}} \in \left[ {4,6} \right)$LQM（向左切向避障）
$ \omega \in \left[-{90}^{\circ }, -{65}^{\circ }\right) $NB（负大）$\delta \in \left[ {{{30}^ \circ },{{150}^ \circ }} \right]$QB（大）${\text{output}} \in \left[ {6,8} \right]$RQM（向右切向避障）
$\omega \in \left[ { - {{65}^ \circ }, - {{20}^ \circ }} \right)$NS（负小）$\delta \in \left[ {0,{{30}^ \circ }} \right)$QS（小）——
Tab.1 Semantic comparison table of tangential obstacle avoidance
输入参数语义输入参数语义输出参数语义
$d \in \left[ {{\text{0}}{\text{.00}},{\text{0}}{\text{.04}}} \right) \;{\mathrm{m}}$HJ（很近）$ |\gamma |\in \left[{54}^{\circ },{72}^{\circ }\right) $JD（较大）$\varepsilon \in \left[ {0,2} \right)$ZE（零）
$d \in \left[ {{\text{0}}{\text{.40,2}}{\text{.00}}} \right)\;{\mathrm{m}} $JJ（较近）$ |\gamma |\in \left[{72}^{\circ },{90}^{\circ }\right] $HD（很大）$\varepsilon \in \left[ {2,4} \right)$PS（小）
$ d\in \left[\text{2}\text{.00},\text{4}\text{.00}\right)\;{\mathrm{m}} $SZ（适中）${V_{{\text{rel}}}} \in \left[ {{\text{0}}{\text{.00,0}}{\text{.08}}} \right){\mathrm{m}}\cdot {\mathrm{s}}^{-1}$HX（很小）$\varepsilon \in \left[ {4,6} \right)$MD（中）
$ d\in \left[\text{4}\text{.00},\text{6}\text{.00}\right)\;{\mathrm{m}} $JY（较远）${V_{{\text{rel}}}} \in \left[ {{\text{0}}{\text{.08,0}}{\text{.16}}} \right){\mathrm{m}}\cdot {\mathrm{s}}^{-1} $JX（较小）$\varepsilon \in \left[ {6,8} \right]$PB（大）
$ d\in \left[\text{6}\text{.00},\text{10}\text{.00}\right]\;{\mathrm{m}} $HY（很远）${V_{{\text{rel}}}} \in \left[ {{\text{0}}{\text{.16,0}}{\text{.24}}} \right){\mathrm{m}}\cdot {\mathrm{s}}^{-1} $SZ（适中）——
$ |\gamma |\in \left[{0}^{\circ },{18}^{\circ }\right) $HX（较小）${V_{{\text{rel}}}} \in \left[ {{\text{0}}{\text{.24,0}}{\text{.32}}} \right){\mathrm{m}}\cdot {\mathrm{s}}^{-1} $JD（较大）——
$ |\gamma |\in \left[{18}^{\circ },{36}^{\circ }\right) $JX（较小）${V_{{\text{rel}}}} \in \left[ {{\text{0}}{\text{.32,0}}{\text{.40}}} \right){\mathrm{m}}\cdot {\mathrm{s}}^{-1} $HD（很大）——
$ |\gamma |\in \left[{36}^{\circ },{54}^{\circ }\right) $SZ（适中）————
Tab.2 Semantic comparison table of obstacle avoidance angle compensation
Fig.14 Performance comparison of improved artificial potential field method under different simulation environments
算法环境复杂度CT/sL/mS
沿边走简单2012.748.26
较复杂2021.580.68
复杂1952.7197.417
虚拟目标点简单2014.051.85
较复杂1426.096.47
复杂945.7169.77
虚拟障碍物简单2014.753.24
较复杂1825.195.56
复杂1344.5166.17
本研究简单2012.242.22
较复杂2020.978.85
复杂1835.4130.19
Tab.3 Statistics of different algorithm planning results for different environmental complexity
Fig.15 Obstacle avoidance verification of Gazebo simulation environment
参数数值参数数值
最大线速度${v_{\mathrm{l}}}$/(m·s?1)0.4全局代价地图可视化
话题发布频率${f_{{\mathrm{vgc}}}}$/Hz		1.0
最大角速度${\omega _{\mathrm{l}}}$/(rad·s?1)1.5局部代价地图可视化
话题发布频率${f_{{\mathrm{vlc}}}}$/Hz		3.0
全局代价地图
刷新频率${f_{{\mathrm{gc}}}}$/Hz		1.5速度控制指令话题
发布频率${{{f}}_{{\text{cmd}}}}$/Hz		10.0
局部代价地图
刷新频率${f_{{\mathrm{lc}}}}$/Hz		5.0——
Tab.4 Experimental parameters of mixed path planning module in simulation environment
Fig.16 Static obstacle avoidance results in indoor environment
Fig.17 Test results comparison of omnidirectional mobile platforms
Fig.18 Dynamic obstacle avoidance results in indoor environment
[12]   MIN H, LIN Y, WANG S, et al Path planning of mobile robot by mixing experience with modified artificial potential field method[J]. Advances in Mechanical Engineering, 2015, 7 (12): 1- 17
[13]   KHATIB O Real time obstacle avoidance for manipulators and mobile robots[J]. International Journal of Robotics Research, 1986, 5 (1): 90- 98
doi: 10.1177/027836498600500106
[14]   HAO G, LV Q, HUANG Z, et al UAV path planning based on improved artificial potential field method[J]. Aerospace, 2023, 10 (6): 562
[15]   林洁, 张志安 改进人工势场法的路径规划研究[J]. 机械与电子, 2022, 40 (3): 65- 70
LIN Jie, ZHANG Zhi’an Research on path planning of improved artificial potential field[J]. Machinery and Electronics, 2022, 40 (3): 65- 70
[16]   ZHENG Y, SHAO X, CHEN Z, et al Improvements on the virtual obstacle method[J]. International Journal of Advanced Robotic Systems, 2020, 17 (2): 1- 9
[1]   TAN C S, MOHD-MOKHTAR R, ARSHAD M R A comprehensive review of coverage path planning in robotics using classical and heuristic algorithms[J]. IEEE Access, 2021, 9: 119310- 119342
doi: 10.1109/ACCESS.2021.3108177
[2]   SÁNCHEZ-LBÁÑEZ J R, PÉREZ-DEL-PULGAR C J, GARCÍA-CEREZO A Path planning for autonomous mobile robots: a review[J]. Sensors, 2021, 21 (23): 7898
doi: 10.3390/s21237898
[3]   MOLINOS E J, LLAMAZARES Á, OCAÑA M Dynamic window based approaches for avoiding obstacles in moving[J]. Robotics and Autonomous Systems, 2019, 118: 112- 130
doi: 10.1016/j.robot.2019.05.003
[4]   代婉玉, 张丽娟, 吴佳峰, 等 改进TEB算法的局部路径规划算法研究[J]. 计算机工程与应用, 2022, 58 (8): 283- 288
DAI Wanyu, ZHANG Lijuan, WU Jiafeng, et al Research on local path planning algorithm based on improved TEB algorithm[J]. Computer Engineering and Applications, 2022, 58 (8): 283- 288
[5]   SZCZEPANSKI R Safe artificial potential field: novel local path planning algorithm maintaining safe distance from obstacles[J]. IEEE Robotics and Automation Letters, 2023, 8 (8): 4823- 4830
doi: 10.1109/LRA.2023.3290819
[6]   MA H, PEI W, ZHANG Q Research on path planning algorithm for driverless vehicles[J]. Mathematics, 2022, 10 (15): 2555
[7]   AZZABI A, NOURI K An advanced potential field method proposed for mobile robot path planning[J]. Transactions of the Institute of Measurement and Control, 2019, 41 (11): 3132- 3144
doi: 10.1177/0142331218824393
[8]   YANG W, WU P, ZHOU X, et al Improved artificial potential field and dynamic window method for amphibious robot fish path planning[J]. Applied Sciences, 2021, 11 (5): 2114
doi: 10.3390/app11052114
[9]   TONG X, YU S, LIU G, et al A hybrid formation path planning based on A* and multi-target improved artificial potential field algorithm in the 2D random[J]. Advanced Engineering Informatics, 2022, 54: 101755
doi: 10.1016/j.aei.2022.101755
[10]   LUO J, WANG Z X, PAN K L Reliable path planning algorithm based on improved artificial potential field method[J]. IEEE Access, 2022, 10: 108276- 108284
doi: 10.1109/ACCESS.2022.3212741
[1] Kai LIN,Xin-wu LIANG,Ji-yuan CAI. Dynamic RGB-D SLAM algorithm based on reprojection depth difference cumulative map and static probability[J]. Journal of ZheJiang University (Engineering Science), 2022, 56(6): 1062-1070.
[2] CHEN Nan, SHOU Li-dan, CHEN Gang, CHEN Ke, HU Tian-lei. elf-tuning indexing of moving objects for dynamic environment[J]. Journal of ZheJiang University (Engineering Science), 2013, 47(3): 442-448.