摘要
针对遥操作机器人在深海等未知环境中抓取软体物体或易碎物体的工作需求,提出了一种人机交互遥操作机器人的主从软体手位置跟踪方法。首先,介绍了主从软体手控制系统的构成与工作原理,该系统主要包括主端控制回路、通信链路和从端控制回路,用于控制软体手的抓取动作。然后,阐述了软体手指的建模过程与控制器设计,采用假设模态法进行建模,并在软体手指的位置跟踪控制中引入模型预测控制算法,以解决软体从手跟踪软体主手性能差的问题。最后,设计并研制了软体主手、软体从手及其控制系统,其中软体从手以硅胶为原料并嵌入固体材料来增大刚度,同时开展了软体从手跟踪软体主手抓握目标物体的实验。仿真结果表明,所设计的模型预测控制器能有效解决软体手指因模型失配而引起的控制精度下降问题;实验结果表明,所研制的软体从手能有效跟踪软体主手并实现目标物体的抓握,整个控制系统运行良好。研究结果为人机交互遥操作机器人软体手的跟踪控制应用提供了参考。
Aiming at the requirements of teleoperation robots grasping soft or fragile objects in unknown environments such as the deep sea,a master-slave soft hand position tracking method for humancomputer interactive teleoperation robot was proposed.Firstly,the structure and working principle of the master-slave soft hand control system were introduced.The system mainly included the master control loop,acommunication link and slave control loop,which was used to control the grasping action of the soft hand.Then,the modeling process and controller design of the soft finger were described.The hypothetical modal method was used to model the soft finger,and the model predictive control algorithm was introduced into the position tracking control of the soft finger to solve the problem of poor performance of the soft slave hand tracking soft master hand.Finally,the soft master hand,soft slave hand and their control systems were designed and developed,in which the soft slave hand was made of silicone and embedded with solid materials to increase the stiffness.Meanwhile,the experiment of soft slave hand tracking soft master hand to grasp the target object was conducted.The simulation results showed that the designed model predictive controller could effectively solve the problem of control accuracy degradation caused by model mismatch for soft fingers;the experimental results showed that the developed soft slave hand could effectively track the soft master hand to grasp the target objects,and the entire control system ran well.The research results provide a reference for the tracking control application of soft hands of human-computer interactive teleoperation robot.
作者
杨淦华
曾庆军
韩春伟
黄鑫
戴晓强
YANG Ganhua;ZENG Qingjun;HAN Chunwei;HUANG Xin;DAI Xiaoqiang(College of Automation,Jiangsu University of Science and Technology,Zhenjiang 212100,China)
出处
《工程设计学报》
CSCD
北大核心
2023年第2期164-171,共8页
Chinese Journal of Engineering Design
基金
国家自然科学基金资助项目(11574120)
江苏省产业前瞻与共性技术项目(BE2018103)
江苏省研究生科研创新计划(KYCX22_3820)。
关键词
遥操作机器人
软体手
位置跟踪
模型预测控制算法
teleoperation robot
soft hand
position tracking
model predictive control algorithm
