基于MPC的仿生尺蠖机器人运动控制  

Motion control of bionic inchworm robot based on MPC

作　　者：杜巧玲[1] 薛成泽 郑伟[2] DU Qiaoling;XUE Chengze;ZHENG Wei(College of Electronic Science&Engineering,Jilin University,Changchun 130015,Jilin,China;College of Computer Science and Technology,Jilin University,Changchun 130015,Jilin,China)

机构地区：[1]吉林大学电子科学与工程学院,吉林长春130015 [2]吉林大学计算机科学与技术学院,吉林长春130015

出　　处：《沈阳工业大学学报》2025年第1期124-129,共6页Journal of Shenyang University of Technology

基　　金：国家自然科学基金项目(52272080);国家重点研发计划项目(2016YFD0700101-04)。

摘　　要：【目的】随着科技的日新月异,人们的生活环境也在发生着巨大的变化。工业化的聚集以及人口密度的增加促使建筑不断向上增高或向下延伸,出现了很多人们不方便到达或是无法到达的地方,但这些地方机器人却可以到达。目前机器人最热门的研究方向之一是非常规结构机器人,通过精简自身的冗余结构以及体积,增加对应的功能结构,使机器人可以在不适宜人们工作的地方代替人们进行相应的工作。【方法】研制了一款适合攀爬的仿尺蠖机器人。针对优化尺蠖机器人在运动过程中的步态问题,提出了基于模型预测控制(MPC)的仿生吸盘式尺蠖机器人运动控制方法。根据章鱼的吸盘结构设计了仿章鱼负压式吸盘以及吸盘控制模型,利用气压传感器反馈的数据实现闭环控制;基于MPC仿生吸盘控制器提出了基于模型预测仿生吸盘控制的仿尺蠖机器人运动步态。【结果】机器人仿真实验和测试结果表明,基于MPC仿生吸盘控制器(MPC-BSC)在吸盘气压优化上相较PID控制有明显优势,MPC-BSC仿尺蠖机器人可以在水平和竖直的瓷砖平面完成“Ω”型运动步态。本文的主要研究结果如下:设计了仿章鱼负压式吸盘,并提出了吸盘控制模型。在负压控制下,可以调整吸盘的吸附压力以满足躯干结构运动对吸附力的需求。提出了基于MPC的仿生吸盘控制算法,该算法通过调节真空泵流速进而实现对吸盘吸附力的调节。设计了仿生吸盘式尺蠖机器人,其躯干结构安装了4个纵向伺服电机和2个横向伺服电机,可以实现“Ω”型运动步态和转向步态。提出了MPC-BSC的运动步态规划,实现“Ω”型运动步态和转向步态。实验测试了仿生吸盘的性能,验证边缘结构对仿生吸盘吸附力的作用。【结论】通过仿真实验测试了基于MPC的仿生吸盘控制算法的可行性,实验验证了仿尺蠖机器人可以在不同倾角[Objective]With the rapid advancement of technology,our living environment is undergoing tremendous changes.The agglomeration of industrialization and the increase in population density have prompted buildings to continuously rise or extend downward,resulting in many places that are inconvenient or impossible for people to reach,but robots are accessible to them.One of the most popular research directions for robots currently is the development of robots with unconventional structures.By streamlining their redundant structures and reducing their size,while adding corresponding functional structures,robots can replace people in performing tasks in places where it is not suitable for humans to work.[Methods]This paper developed a climbing inchworm-inspired robot.To optimize the gait of the inchworm robot during motion,the paper proposeed a motion control method for the inchworm robot with bionic suckers based on model predictive control(MPC).An octopus-inspired negative-pressure sucker was designed according to the sucker structure of octopuses,and a sucker control model was established.Closed-loop control was achieved using data fed back by a pressure sensor.Based on the MPC bionic sucker controller(MPC-BSC),the paper also proposed a motion gait for the inchworm robot under model predictive bionic sucker control.[Results]The results of robot simulation experiments and tests show that the MPC-BSC has obvious advantages over PID control in terms of sucker pressure optimization.The inchworm robot based on MPC-BSC can complete anΩ-shaped motion gait in both horizontal and vertical tile planes.The main research contents of this paper are as follows:An octopus-inspired negative-pressure sucker was designed,and a sucker control model was proposed.Under negative pressure control,the suction pressure of the sucker can be adjusted to meet the adsorption force requirements for body structure movement.A bionic sucker control algorithm based on MPC was proposed.This algorithm adjusts suction force of the sucker by regulating

关 键 词：仿生吸盘 仿生尺蠖 机器人 运动规划 机械设计 电机控制 嵌入式系统 姿态控制 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]