基于高斯过程的软体机器人微分运动学控制  

Gaussian Process Based Differential Kinematics Control of the Soft Robot

作　　者：沈逸 樊耕霖 唐秀川 董云龙 邓礼楠 SHEN Yi;FAN Geng-lin;TANG Xiu-chuan;DONG Yun-long;DENG Li-nan(School of Artificial Intelligence and Automation,Huazhong University of Science and Technology,Wuhan 430074,China;School of Mechanical Science and Engineering,Huazhong University of Science and Technology,Wuhan 430074,China)

机构地区：[1]华中科技大学人工智能与自动化学院,湖北武汉430074 [2]华中科技大学机械科学与工程学院,湖北武汉430074

出　　处：《控制工程》2021年第7期1360-1365,共6页Control Engineering of China

基　　金：国家自然科学基金资助项目(91748112)。

摘　　要：近年来,软体机器人因其在人体等受限环境中表现出了相对刚体机器人更好的运动连续性和柔顺性而广受关注。然而,软体机器人没有刚体关节,无法采用刚体机器人领域中成熟的刚体关节建模和控制方法。首先,采用高斯过程对真实驱动空间到虚拟驱动空间的非线性映射进行辨识;然后,通过链式法则建立了从虚拟关节空间到任务空间的映射,构建了软体机器人真实驱动空间到任务空间的正逆运动学模型;最后,提出了一种基于逆运动学模型的闭环控制策略。并且通过实验验证了所提出的软体机器人建模和控制算法的有效性。In recent years,soft robots have attracted more and more attentions because they show better continuity and flexibility than rigid robots in restricted environments such as human body.However,the methods regarding the rigid joint modeling and control in previous works cannot be adopted since soft robots do not have rigid joints.To deal with the above-mentioned issue,this paper introduces Gaussian process to identify the mapping from the real driven space to the virtual driven space for soft robot kinematics.Then,the mapping from the virtual joint space to the task space is established by the chain rule.Finally,a closed-loop control strategy based on the inverse kinematics model is proposed.The effectiveness of the soft robot model and control algorithm proposed in this work is verified byexperiments.

关 键 词：高斯过程 软体机器人 微分逆运动学 运动控制 常曲率模型 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]