Human-machine shared obstacle avoidance method for wheelchair robot considering individual habit

doi:10.3785/j.issn.1008-973X.2024.11.011

Journal of ZheJiang University (Engineering Science)

2024, Vol. 58

Issue (11): 2299-2308 DOI: 10.3785/j.issn.1008-973X.2024.11.011

Human-machine shared obstacle avoidance method for wheelchair robot considering individual habit

Yina WANG1(

),Chen CAO1,Jiaqi YANG1,Yanjun YU1,Guoqiang FU1,Shuoyu WANG2

1. School of Electrical Engineering, Shenyang University of Technology, Shenyang 110870, China
2. School of Systems Engineering, Kochi University of Technology, Kochi 7820003, Japan

Download:

HTML

PDF(2645KB) HTML
Export: BibTeX | EndNote (RIS)

Abstract

A dynamic reinforcement learning strategy was introduced to establish the correlation between individual operating habits and collision risk based on a triple-reward system in order to resolve the impact of individual operating habits on the shared motion control of intelligent wheelchair robots (WR). A fuzzy reinforcement learning state fusion-based shared control strategy was proposed, which could adapt to user behavior while ensuring safety. A distance fuzzy reasoning algorithm was employed to develop a direction intention recognition model based on seat pressure, which served as the foundation for establishing a human-machine shared control framework in order to achieve intelligent robot control. The current and predictive reward functions were established via the Gaussian function and deviation rate, respectively, focusing on the deviation between the user’s intended direction and the robot’s actual direction in order to estimate user operating habits. A task reward function was created according to boundary distance in order to predict human-machine safety. The correlation between user operating habits and safety was constructed by utilizing the fuzzy reinforcement learning strategy and the triple-reward system in order to dynamically adjust the user control weight within the shared control to adapt to individual habits. Then the precision and safety of human-machine shared control were enhanced. The effectiveness of the proposed algorithm was verified by experiments in a test environment.

Key words： intelligent wheelchair robot distance fuzzy reasoning algorithm fuzzy reinforcement learning individual habit human-machine shared control

Received: 13 December 2023 Published: 23 October 2024

CLC:

TP 242

Fund: 国家自然科学基金资助项目（52175105）；教育部春晖计划资助项目（HZKY20220415）；辽宁省科技厅资助项目（2024MS107）；辽宁省教育厅资助项目（JYTMS20231207）．

	Service
	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Yina WANG
	Chen CAO
	Jiaqi YANG
	Yanjun YU
	Guoqiang FU
	Shuoyu WANG

Cite this article:

Yina WANG,Chen CAO,Jiaqi YANG,Yanjun YU,Guoqiang FU,Shuoyu WANG. Human-machine shared obstacle avoidance method for wheelchair robot considering individual habit. Journal of ZheJiang University (Engineering Science), 2024, 58(11): 2299-2308.

URL:

https://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2024.11.011 OR https://www.zjujournals.com/eng/Y2024/V58/I11/2299

考虑个体习惯的轮椅机器人人机共享避障方法

为了避免个体操作习惯对智能轮椅机器人(WR)人机共享运动控制的影响，引入动态强化学习策略，基于三重奖励系统建立个体操作习惯与碰撞风险的关联特性，提出能够自适应用户行为及保证安全性的模糊强化学习状态融合式共享控制策略. 为了实现机器人的智能操控，采用距离型模糊推理算法建立基于座椅压力的方向意图识别模型和机器人人机共享控制框架. 面向用户意图方向与机器人实际方向的偏差度，分别基于高斯函数与偏差率建立当前奖励函数与预测奖励函数，以估计用户操作习惯. 基于边界距离建立任务奖励函数，以估计人机安全性. 基于模糊强化学习策略，利用三重奖励函数构建用户操作习惯与安全性的关联性，以动态调整共享控制中的用户控制权重，适应个体习惯，提高人机共享的操控精度和安全性. 在实验室搭建试验环境，验证了所提算法的有效性.

关键词： 智能轮椅机器人, 距离型模糊推理算法, 模糊强化学习, 个体习惯, 人机共享控制

Fig.1 Structure of wheelchair robot

Fig.2 System flowchart of human-machine shared obstacle avoidance method considering individual operating habit

Fig.3 Subject’s required direction and angle in experiment

Tab.1 If-then fuzzy rules

Fig.4 Definition of linguistic variables’ membership function

Fig.5 Schematic diagram of shared control system

Fig.6 Membership function of shared fuzzy controller

Tab.2 Fuzzy rule table

Fig.7 Structure of fuzzy reinforcement learning

Tab.3 Identification error of each subject

Fig.8 Actual photo of experimental environment

Fig.9 Obstacle avoidance path with desired speed of 0.2 m/s

Fig.10 Obstacle avoidance path with desired speed of 0.3 m/s

Fig.11 Correlated data changes in command direction, actual travel direction and operational weight

Tab.4 Comparison result before and after shared control

Fig.12 Actual photo of experimental environment 1

Fig.13 Actual photo of experimental environment 2

Fig.14 Comparison of robot travel directions in experimental environment 1

Fig.15 Online training process of relevant data in experimental enviroment 1

Fig.16 Comparison of robot travel directions in experimental enviroment 2

Fig.17 Online training process of relevant data in experimental enviroment 2

Fig.18 Changing trend of evaluation indicator of same subject in different experimental environment

Fig.19 Changing trend in evaluation criteria of different experimental subject in experimental scenario 1

Tab.5 Result of paired T-test for |β_h−β_s|


[1]	OHARA E, WATANABE T, OISHI T, et al. Assistance control of wheelchair operation using active cast for the upper limb [C]// IEEE International Conference on Robotics and Automation . Shanghai: IEEE, 2011: 2223-2228.

[2]	CHALOEM T, YOKOTA S, HASHIMOTO H, et al. Oscillation suppression control for electric wheelchair using human body motion interface [C]// IEEE International Conference on Industrial Technology . Lyon: IEEE, 2018: 1991-1996.

[3]	LU T. A motion control method of intelligent wheelchair based on hand gesture recognition [C]// IEEE 8th Conference on Industrial Electronics and Applications . Melbourne: IEEE, 2013: 957-962.

[4]	NASIF S, KHAN M A G. Wireless head gesture controlled wheel chair for disable persons [C]// IEEE Region 10 Humanitarian Technology Conference . Dhaka: IEEE, 2017: 156-161.

[5]	ÖZLÜK Y, AKMAN-AYDIN E Fuzzy logic control of a head-movement based semi-autonomous human–machine interface[J]. Journal of Bionic Engineering, 2023, 20 (2): 645- 655 doi: 10.1007/s42235-022-00272-3

[6]	KAUR A Wheelchair control for disabled patients using EMG/EOG based human machine interface: a review[J]. Journal of Medical Engineering and Technology, 2021, 45 (1): 61- 74 doi: 10.1080/03091902.2020.1853838

[7]	张亚徽, 王斐, 李景宏, 等基于稳态视觉诱发电位的智能轮椅半自主导航控制[J]. 机器人, 2019, 41 (5): 620- 627 ZHANG Yahui, WANG Fei, LI Jinghong, et al Semi-autonomous navigation control of intelligent wheelchair based on steady state visual evoked potential[J]. Robot, 2019, 41 (5): 620- 627

[8]	RECHY-RAMIREZ E J, HU H A flexible bio-signal based HMI for hands-free control of an electric powered wheelchair[J]. International Journal of Artificial Life Research, 2014, 4 (1): 59- 76 doi: 10.4018/ijalr.2014010105

[9]	JAMEEL H F, MOHAMMED S L, GHARGHAN S K. Wheelchair control system based on gyroscope of wearable tool for the disabled [C]// IOP Conference Series: Materials Science and Engineering . Baghdad: IOP Publishing, 2020, 745(1): 012091.

[10]	PAING M P, JUHONG A, PINTAVIROOJ C Design and development of an assistive system based on eye tracking[J]. Electronics, 2022, 11 (4): 535 doi: 10.3390/electronics11040535

[11]	HORI J, OHARA H, INAYOSHI S Control of speed and direction of electric wheelchair using seat pressure mapping[J]. Biocybernetics and Biomedical Engineering, 2018, 38 (3): 624- 633 doi: 10.1016/j.bbe.2018.04.007

[12]	FAN J, JIA S, LI X, et al. Motion control of intelligent wheelchair based on sitting postures [C]// IEEE International Conference on Mechatronics and Automation . Beijing: IEEE, 2011: 301-306.

[13]	KHALILULLAH K M I, OTA S, YASUDA T, et al Road area detection method based on DBNN for robot navigation using single camera in outdoor environments[J]. Industrial Robot: An International Journal, 2018, 45 (2): 275- 286 doi: 10.1108/IR-08-2017-0139

[14]	SEZER V An optimized path tracking approach considering obstacle avoidance and comfort[J]. Journal of Intelligent and Robotic Systems, 2022, 105 (1): 21 doi: 10.1007/s10846-022-01636-x

[15]	陈英龙, 宋甫俊, 张军豪, 等基于临场感的遥操作机器人共享控制研究综述[J]. 浙江大学学报: 工学版, 2021, 55 (5): 831- 842 CHEN Yinglong, SONG Fujun, ZHANG Junhao, et al Telerobotic shared control strategy based on telepresence: a review[J]. Journal of Zhejiang University: Engineering Science, 2021, 55 (5): 831- 842

[16]	JAIN S, ARGALL B. Recursive Bayesian human intent recognition in shared-control robotics [C]// IEEE/RSJ International Conference on Intelligent Robots and Systems . Madrid: IEEE, 2018: 3905-3912.

[17]	NARAYANAN V K, SPALANZANI A, BABEL M. A semi-autonomous framework for human-aware and user intention driven wheelchair mobility assistance [C]// IEEE/RSJ International Conference on Intelligent Robots and Systems . Daejeon: IEEE, 2016: 4700-4707.

[18]	DENG X, YU Z L, LIN C, et al Self-adaptive shared control with brain state evaluation network for human-wheelchair cooperation[J]. Journal of Neural Engineering, 2020, 17 (4): 045005 doi: 10.1088/1741-2552/ab937e

[19]	EZEH C, TRAUTMAN P, DEVIGNE L, et al. Probabilistic vs linear blending approaches to shared control for wheelchair driving [C]// International Conference on Rehabilitation Robotics . London: IEEE, 2017: 835-840.

[20]	JOHANNINK T, BAHL S, NAIR A, et al. Residual reinforcement learning for robot control [C]// International Conference on Robotics and Automation . Montreal: IEEE, 2019: 6023-6029.

[21]	XI L, SHINO M Shared control design methodologies of an electric wheelchair for individuals with severe disabilities using reinforcement learning[J]. Journal of Advanced Simulation in Science and Engineering, 2020, 7 (2): 300- 319 doi: 10.15748/jasse.7.300

[22]	WANG Y, WANG S A new directional-intent recognition method for walking training using an omnidirectional robot[J]. Journal of Intelligent and Robotic Systems, 2017, 87 (2): 231- 246 doi: 10.1007/s10846-017-0503-z

[23]	LINDBLAD J, SLADOJE N Linear time distances between fuzzy sets with applications to pattern matching and classification[J]. IEEE Transactions on Image Processing, 2013, 23 (1): 126- 136

[24]	ZHANG D, WANG Y, LIU Z, et al Intelligent obstacle avoidance wheelchair based on fuzzy reasoning[J]. ICIC Express Letters, Part B: Applications, 2021, 12 (9): 831- 838

[25]	刘智敏, 叶宝林, 朱耀东, 等基于深度强化学习的交通信号控制方法[J]. 浙江大学学报: 工学版, 2022, 56 (6): 1249- 1256 LIU Zhimin, YE Baolin, ZHU Yaodong, et al Traffic signal control method based on deep reinforcement learning[J]. Journal of Zhejiang University: Engineering Science, 2022, 56 (6): 1249- 1256

[1]	Dongxing CAO,Yanchao JIA,Xiangying GUO,Jiajia MAO. Structure design and motion analysis of bionic hexapod origami robot[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(8): 1543-1555.

[2]	Zhiwei FAN,Kai JIA,Lei ZHANG,Fengshan ZOU,Zhenjun DU,Mingmin LIU. Optimal velocity planning for mobile robot based on simultaneous dynamic optimization[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(8): 1556-1564.

[3]	Haipeng QIN,Rui QIN,Xiaofen SHI,Xiaoming ZHU. Motion control of quadruped robot based on model prediction[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(8): 1565-1576.

[4]	Songlin HUANG,Xiujuan ZHENG,Xiaoyue TAN,Xing HU,Haiyan TU,Kang LI. Design of resistance neck rehabilitation robot system[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(7): 1479-1487.

[5]	Dong CHEN,Weida LI,Hongmiao ZHANG,Juan LI. Ankle flexible exoskeleton based on force feedback admittance control[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(4): 772-778.

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

[7]	Mingjun SONG,Wen YAN,Yizhao DENG,Junran ZHANG,Haiyan TU. Light-weight algorithm for real-time robotic grasp detection[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(3): 599-610.

[8]	Yuting LIU,Shijie GUO,Shufeng TANG,Xuewei ZHANG,Tiantian LI. *Path planning based on fusion of improved A and ROA-DWA for robot**[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(2): 360-369.

[9]	Wan-jin GUO,Wu-duan ZHAO,Qian-hui LI,Li-jun ZHAO,Chu-qing CAO. Ensemble probabilistic model based variable impedance for robotic grinding force control[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(12): 2356-2366.

[10]	Yu-feng JIANG,Dong-sheng CHEN. Assembly strategy for large-diameter peg-in-hole based on deep reinforcement learning[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(11): 2210-2216.

[11]	Ya-li XUE,Jin-ze YE,Han-yan LI. Multi-agent pursuit and evasion games based on improved reinforcement learning[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(8): 1479-1486.

[12]	Wan-jin GUO,Wu-duan ZHAO,Su-yang YU,Li-jun ZHAO,Chu-qing CAO. Active adaptive online trajectory prediction for robotic grinding on surface without prior model[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(8): 1655-1666.

[13]	Chong-liang ZHONG,Yun-feng LIU,Wei-dong ZHU,Fu-dong ZHU. Multi-orientation trajectory smoothing planning of robot for dental implant[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(5): 1030-1037.

[14]	Chao-fan ZHANG,Yi-ming QIAO,Lu CAO,Zhi-gang WANG,Shao-wei CUI,Shuo WANG. Tactile slip detection method based on neuromorphic modeling[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(4): 683-692.

[15]	Tong-li CHANG,Wan-bin FU. Human-robot matching design of self-aligning artificial knee joint[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(4): 753-759.

Viewed

Full text

Abstract

Cited

Shared

Discussed