网络化遥操作系统控制器设计及性能分析

发布时间：2018-12-29 16:26

【摘要】：近几年,融合了控制理论、计算机技术和通信技术的机器人遥操作系统吸引了大量研究者的兴趣,并已在核事故救援、海底作业、空间探测和远程医疗等方面有着广泛应用。正因如此,学术界和工业界投入了巨大的人力和物力进行科学研究。为了便于操作者远距离操作特定对象(如危险品等),需要通信网络保证数据实时有效的传输,进而产生了网络化遥操作系统。然而,网络环境中的时延、量化等现象使得传统的控制器难以直接应用于网络化遥操作系统中。基于此,本文结合网络环境中的时延、量化等现象开展网络化遥操作系统的控制器设计及性能分析。本文的研究工作有如下几个方面:(1)针对带有定常时滞网络化遥操作系统稳定域问题,提出了整数阶控制器及分数阶控制器。通过构建闭环系统的误差系统,在频域内利用特征根集中处理方法,建立了整数阶比例微分控制策略及位置误差控制策略下遥操作系统稳定的充分必要条件。此外,通过定义一系列状态变量,构建闭环遥操作系统状态方程,并提出分数阶遥操作系统精确稳定域算法。(2)针对网络化遥操作系统的时变时延及执行器饱和问题,通过定义两对平行面,在这两对平行面所包围的空间内将饱和控制器表示成实际反馈和其辅助项的凸组合形式。构建基于位置误差和速度的新型Lyapunov-Krasovskii函数,并建立基于位置误差的遥操作系统Newton-Leibnitz等式放松一些强关联项,提出了区间时滞遥操作系统存在执行器饱和下的稳定性判据及吸引域估计方法,并提出优化算法最大化吸引域,进一步为遥操作系统的应用提供了理论支撑。(3)针对带有时变时滞的非线性网络化遥操作系统线性化问题,建立了自由度遥操作系统T-S模糊模型。设计了带有重力补偿的控制策略并增加力反馈项。分别针对主从端模型参数不同和相同两种情况提出稳定性分析方法,通过对非线性矩阵不等式线性化,提出了控制器增益设计方法。n(4)针对网络化遥操作系统中通信信道带宽受限问题,分析了带有量化反馈信息的遥操作系统稳定性能,并设计了基于位置误差和阻尼输入的控制器和力矩反馈控制器。建立了量化器参数、控制器增益、时延和饱和界的关系。(5)针对网络环境下遥操作系统局部信息(如关节速度信息)不可测问题,设计了基于浸入与不变流形的速度观测器。定义流形面及状态到流形面的距离并证明流形面是不变且吸引的,进一步定义扩张状态及观测误差,分别构建状态及误差的子动态系统,证明扩张状态及观测误差趋于零。此外,根据所得的速度观测信号设计控制器,能有效的保证遥操作系统的跟踪性能。
[Abstract]:In recent years, the robotic teleoperation system, which combines control theory, computer technology and communication technology, has attracted the interest of a large number of researchers, and has been widely used in nuclear accident rescue, submarine operations, space exploration and telemedicine. For this reason, academia and industry have invested huge human and material resources in scientific research. In order to facilitate the operator to operate a specific object (such as dangerous goods) from a long distance, the communication network is needed to ensure the real-time and effective transmission of data, and then a networked teleoperation system is produced. However, the delay, quantization and other phenomena in the network environment make it difficult for traditional controllers to be directly applied to networked teleoperation systems. Based on this, the controller design and performance analysis of networked remote operating system are carried out in this paper. The research work of this paper is as follows: (1) for the stability domain of networked remote operating system with time-delay, an integer order controller and a fractional order controller are proposed. By constructing the error system of the closed-loop system and using the characteristic root centralized processing method in the frequency domain, the sufficient and necessary conditions for the stability of the teleoperation system under the integer order proportional differential control strategy and the position error control strategy are established. In addition, by defining a series of state variables, the closed-loop teleoperating system state equation is constructed, and the fractional teleoperating system precise stability region algorithm is proposed. (2) aiming at the problem of time-varying delay and actuator saturation of networked teleoperating system, By defining two pairs of parallel surfaces, the saturation controller is represented as a convex combination of the actual feedback and its auxiliary terms in the space surrounded by these two pairs of parallel faces. A new Lyapunov-Krasovskii function based on position error and velocity is constructed, and the Newton-Leibnitz equation based on position error is established to relax some strong correlation terms. In this paper, the stability criterion and the estimation method of the attraction domain in the interval delay teleoperation system with actuator saturation are proposed, and the optimal algorithm is proposed to maximize the attraction domain. It provides theoretical support for the application of remote operating system. (3) aiming at the linearization problem of nonlinear networked remote operating system with time-varying time delay, a T-S fuzzy model of the degree-of-freedom teleoperation system is established. The control strategy with gravity compensation is designed and the force feedback term is added. A stability analysis method is proposed for the different parameters of the master-slave model and the same one. The nonlinear matrix inequalities are linearized. A controller gain design method (. N (4) is proposed to analyze the stability of remote operating system with quantized feedback information, aiming at the bandwidth limitation of communication channel in networked telecontrol system. The controller and torque feedback controller based on position error and damping input are designed. The relations among quantizer parameters, controller gain, delay and saturation bounds are established. (5) for the unmeasurable local information (such as joint velocity information) of remote operating system in network environment, A speed observer based on immersion and invariant manifold is designed. Define the manifold surface and the distance from the state to the manifold surface and prove that the manifold surface is invariant and attractive. Further define the extended state and the observation error, construct the sub-dynamic system of the state and error, and prove that the extended state and the observation error tend to zero. In addition, the controller is designed according to the obtained velocity observation signal, which can effectively guarantee the tracking performance of remote operating system.
【学位授予单位】：燕山大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：TP872

【相似文献】