Foam-Embedded Soft Robotic Joint with Inverse Kinematic Modeling by Iterative Self-Improving Learning

Soft robotic arms have gained significant attention owing to their flexibility and adaptability. Nonetheless, the insta- bility due to their high-elasticity structure further leads to the difficulty of precise kinematic modeling and control. This letter introduces a novel solution employing foam-embedded joint design (Fe-Joint), effectively mitigating oscillations and enhancing motion stability. This innovation is integrated into the new continuum soft robotic arm (Fe-Arm). Through iterative design optimization, the Fe-Arm attains superior mechanical performance and control capabilities, enabling a settling state in 0.4 seconds post external force. Enabled by the quasi-static behavior of Fe-Arm, we pro- pose a long short-term memory network (LSTM) based iterative self-improving learning strategy (ISL) for end-to-end inverse kine- matics modeling, tailored to Fe-Arm’s mechanical traits, enhancing modeling performance with limited data. Investigating key control parameters, we achieve target trajectory modeling errors within 9% of the workspace radius. The generalization potential of the ISL method is demonstrated using the pentagonal trajectory and on a different Fe-Arm configuration. Manuscript received 21 August 2023; accepted 19 December 2023. Date of publication 5 January 2024; date of current version 15 January 2024. This letter was recommended for publication by Associate Editor J. Hughes and Editor Y.-L. Park upon evaluation of the reviewers’ comments. This work was supported in part by the National Key R&D Program of China under Grant 2022YFB4701200, in part by the Shenzhen Science and Technology Program under Grants JCYJ20220530114615034 and JCYJ20220818100417038, in part by the Guangdong Basic and Applied Basic Research Foundation under Grant 2021A1515110658, in part by NSFC under Grant U1913603, and in part by the Guangdong Provincial Key Laboratory of Human-Augmentation and Rehabilitation Robotics in Universities. (Anlun Huang, Yongxi Cao, and Jiajie Guo contributed equally to this work.) (Corresponding authors: Sicong Liu; Zheng Wang.) Anlun Huang, Yongxi Cao, Jiajie Guo, Zhonggui Fang, Sicong Liu, Juan Yi, and Zheng Wang are with the Guangdong Provincial Key Laboratory of Human Augmentation and Rehabilitation Robotics in Universities, Department of Mechanical and Energy Engineering, Southern University of Science and Technology, Shenzhen 518000, China (e-mail: 11912702@mail.sustech.edu.cn; 11811401@mail.sustech.edu.cn; 11811503@mail.sustech.edu.cn; 12231129@ mail.sustech.edu.cn; liusc@sustech.edu.cn; yij3@sustech.edu.cn; wangz@ sustech.edu.cn). Yinyin Su is with the Guangdong Provincial Key Laboratory of Human Augmentation and Rehabilitation Robotics in Universities, Department of Me- chanical and Energy Engineering, Southern University of Science and Tech- nology, Shenzhen 518000, China, and also with the Department of Mechani- cal Engineering, University of Hong Kong, Hong Kong SAR, China (e-mail: yinyinsu1991@gmail.com). Hongqiang Wang and Jian S. Dai are with the Department of Mechanical and Energy Engineering, Southern University of Science and Technology, Shenzhen 518000, China (e-mail: wanghq6@sustech.edu.cn; daijs@sustech.edu.cn). This letter has supplementary downloadable material available at https://doi.org/10.1109/LRA.2024.3349831, provided by the authors. Digital Object Identifier 10.1109/LRA.2024.3349831