【作者】 郭崇滨；

【导师】 郝矿荣；

【作者基本信息】 东华大学 ， 控制理论与控制工程， 2013， 博士

【摘要】 随着机器人技术的迅速发展,并联机器人的优势日益突现,正逐步广泛应用于航空、航天、深海、危险化工、核工业、医疗手术、精密制造业等高精、尖端领域。因此,人们对并联机器人的定位精度、运行平稳性、设备稳定性、容错能力、自适应性、多机器人协调性等工作性能也提出了更高的要求,其控制问题逐步成为机器人领域中最重要的研究方向之一。同时在现代复杂的信息环境下,以往单目标、少目标的传统控制方法已经无法满足并联机器人多性能指标的控制要求,越来越多的机器人控制系统需要智能化程度更高、实用性更强的多目标智能控制算法。本文从并联机器人工程实践中一些有待解决的实际问题出发,借鉴具有多目标协同调节特性的生物网络机制,对并联机器人的一些多目标协同智能控制问题进行研究。主要研究成果归纳如下：(1)研究了并联机器人运动支链的自适应控制器设计。首先基于内分泌甲状腺激素调节机制设计了一种多级协同调节自适应控制器,能够实现控制器各参数之间的协同自适应调节及各级控制器之间的协同补偿,从而提高机器人控制器的响应速度、控制精度和稳定性。其次基于内分泌调控网络结构提出了一种新颖的速、位协同智能控制器,进一步改进参数自适应调节方法及速、位协同控制能力。(2)研究了并联机器人的位姿、速度、加速度协同智能控制。基于神经内分泌系统的多环反馈机制和多目标协同调节机制,设计了一种多环反馈的多目标协同智能控制系统,并应用于真实的多通道并联机器人设备上。实验结果表明,该系统能够让并联机器人的子通道控制性能和全局控制性能都能得到较大的提高,能够较好地实现位姿、速度、加速度之间的多目标协同智能控制。(3)研究了并联机器人的精准协同容错控制。基于人体生理止血机制的多目标协同调节原理,设计了一种应用于冗余并联机器人的精准容错控制系统,确保冗余并联机器人发生局部故障时,能够继续完成精准的容错任务,实现正常和故障工作状态之间的协同。对比实验结果表明,相对于传统的PID控制器,所提出的容错控制系统对子通道控制和并联机器人整体控制均有更高的控制精度和容错能力。(4)研究了并联机器人的力、位解耦协同智能控制。基于人体体内双边解耦协同调节机制,提出了一种应用于微创手术主-从机器人系统的力、位解耦协同智能控制系统,解决了达芬奇等微创手术机器人无力反馈信号等问题。实验结果表明：提出的控制系统能够在无力传感器情况下,利用运动信号获得较精准的反馈力；基于生物调节启发的解耦、协同机制具有较好的力、位协同控制性能；让操作者准确获得额外的力反馈信号,能够确保机器人平稳地完成力位协同任务。最后,总结了全文的工作内容,指出了目前研究中存在的缺陷与不足,并对今后的研究展望和研究重点进行了讨论。更多还原

【Abstract】 With the rapid development of robot technology, the advantage of parallel robot becomes more observably. Parallel robots are now widely applied to high precision and cutting-edge fields, such as aeronautics, aerospace, deep sea, hazardous chemicals, nuclear, medical treatment, and precision manufacturing, etc. That people requires higher standards for parallel robots’ working performance such as positioning accuracy, running smoothness, equipment stability, fault tolerance, adaptability, and coordination of multi-robot. The control problem gradually becomes one of the most important research directions in the robotic field. Meanwhile, under the modern complex information environment, the single objective and few objective traditional control methods are hard to satisfy parallel robot multi-objective control performance requirements. More and more robot control system needs better intelligent and practical multi-objective control algorithm. To resolve some parallel robot problems in engineering practice, the author does a lot of researches on parallel robot multi-objective cooperative intelligent control according to some biological network regulation mechanisms which owing multi-objective cooperative regulation characteristics. Mainly summarized as follows:(1) Study on adaptive controller design of the parallel robot motion branched chain. First, based on the endocrine thyroid hormone regulation mechanism, a multi-level cooperative regulation adaptive controller is proposed. The controller can regulate control parameters and make cooperative compensates automatically to improve response, accuracy and stability of the robot controller. What’s more, based on endocrine regulation network structure, a novel position-velocity cooperative intelligent controller is proposed, which improve the parameters regulation method and cooperative ability of the position and velocity.(2) Study on position, velocity and acceleration cooperative intelligent control of the parallel robot. Based on multi-loop feedback and multi-objective cooperative regulation mechanism of neuroendocrine system, a multi-loop and multi-objective cooperative intelligent control system is presented and applied to an actual multi-channel parallel robot. The experimental result shows that the proposed system greatly improves the control performance in sub-channels and global robot that can easy to achieve multi-objective cooperative intelligent control of the position, velocity and acceleration.(3) Study on precise cooperative fault-tolerant control of the parallel robot. Based on multi-objective cooperative regulation principle in human hemostasis system, a precise fault-tolerant system is proposed for redundant parallel manipulator to continue their fault-tolerant task accurately with the presence of failure. That robot can achieve cooperative control between the normal state and the fault state. The comparative experiment result shows that compared with traditional PID controller, the proposed fault-tolerant system has better control precision and fault-tolerant capability both in robot sub-channel and global.(4) Study on force-position decoupling and cooperation intelligent control of the parallel robot. Based on bilateral decoupling and cooperation bio-regulation mechanism in human body, a force and position decoupling and cooperation intelligent control system is proposed for minimally invasive surgical master-slave robot system. The new system solves the lacking force feedback problem in minimally invasive surgical robots such as da Vinci system. The experimental results indicate the proposed control system can acquire precise force feedback by motion signal without force sensor; Based on bio-regulation inspired decoupling and cooperation mechanism has a good force-position cooperative control performance; Operator gets accurate force feedback to make sure robot performing task steady.At last, a summary of the thesis is made, and the deficiency in the project and the further development are narrated respectively.更多还原