Inversion techniques for trajectory control of flexible robot arms

A general framework is given for computing the torques that are needed for moving a flexible arm exactly along a given trajectory. This torque computation requires a dynamic generator system, as opposed to the rigid case, and can be accomplished both in an open- or in a closed-loop fashion. In the open-loop case, the dynamic generator is the full or reduced order inverse system associated to the arm dynamics and outputs. In order to successfully invert the arm dynamics, the torque generator should be a stable system. The stability properties depend on the chosen system output, that is on the robot variables (e.g., joint or end-effector) to be controlled. The same inversion technique can be applied for closed-loop trajectory control of flexible robots. A simple but meaningful nonlinear dynamic model of a one-link flexible arm is used to illustrate different feasible control strategies. Simulation results are reported that display the effects of the system output choice on the closed-loop stability and on the overall tracking performance.     

Neuro-fuzzy self-tuning of PID control for semiglobal exponential tracking of robot arms

The PID controller with constant feedback gains has withstood as the preferred choice for control of linear plants or linearized plants, and under certain conditions for non-linear ones, where the control of robotic arms excels. In this paper a model-free self-tuning PID controller is proposed for tracking tasks. The key idea is to exploit the passivity-based formulation for robotic arms in order to shape the damping injection to enforce dissipativity and to guarantee semiglobal exponential convergence in the sense of Lyapunov. It is shown that a neuro-fuzzy network can be used to tune dissipation rate gain through a self-tuning policy of a single gain. Experimental studies are presented to confirm the viability of the proposed approach.     

Direct strain feedback control of flexible robot arms: newtheoretical and experimental results

This paper addresses control problems for flexible robot arms by using direct strain feedback. The purpose is to make clear why direct strain feedback can damp out vibration of flexible arms satisfactorily. We concentrate on one-link flexible robot arms whose dynamic models can be represented by linear partial differential equations with appropriate boundary conditions which have been well examined in a number of papers. A key contribution of this paper is the introduction of the concept of (strict) A-dependent operators, which allows us to prove rigorously the closed loop stability of direct strain feedback and the existence and uniqueness of nonstandard second order abstract differential equations in Hilbert spaces. Several control experiments are performed, verifying the main theoretical points of this paper, demonstrating satisfactory control results of direct strain feedback, and leading to potential application of this simple control method for flexible robot control     

Lyapunov function-based control laws for revolute robot arms:tracking control, robustness, and adaptive control

A new class of joint level control laws for all-revolute robot arms is introduced. The analysis is similar to an energy-like Lyapunov function approach, except that the closed-loop potential function is shaped in accordance with the underlying joint space topology. This approach gives way to a much simpler analysis and leads to a new class of control designs which guarantee both global asymptotic stability and local exponential stability. When Coulomb and viscous friction and parameter uncertainty are present as model perturbations, a sliding mode-like modification of the control law results in a robustness-enhancing outer loop. Adaptive control is formulated within the same framework. A linear-in-the-parameters formulation is adopted and globally asymptotically stable adaptive control laws are derived by simply replacing unknown model parameters by their estimates     

On the PID tracking control of robot manipulators

We consider the problem of PID tracking control of robotics manipulators. Our objective is to prove that under classical PID control, semiglobal stability can be assured with arbitrary small output tracking error. This means that, for any given set of initial conditions W_x, there exist PID control gains such that all trajectories starting in W_x converge to a residual set of arbitrary size. A novel PID control configuration is developed in terms of a parameter that is directly related with the size of the region of attraction and the size of the residual set. Tuning guidelines are extracted from the stability analysis.     

Adaptive tracking control for robot manipulator

In this study, an adaptive control system is proposed for the tracking control of an n-link robot manipulator to achieve high-precision position control. The presentation of the adaptive control system is divided into three parts: a feedforward controller, a state feedback controller and an uncertainty alleviator. All on-line tuning algorithms in the adaptive control system are derived in the sense of Lyapunov stability analysis, so that system-tracking stability can be guaranteed in the closed-loop system whether the uncertainties occur. It has learning ability similar to intelligent control, but with a simpler control framework. Computer simulations of a three-link SCARA robot manipulator verify the validity of the proposed control strategy in the possible presence of uncertainties. The merits of the proposed control scheme are that not only can the stability of the controlled system be guaranteed, but also no constrained conditions and prior knowledge of the controlled plant are required in the design process.     

轨迹跟踪级联机器人编队控制方法

提出一种轨迹跟踪级联机器人编队控制方法. 该方法有效结合距离-角度(??-??) 控制和距离-距离(??-??) 控制方案, 并利用无迹卡尔曼滤波算法对Leader-Follower 级联机器人系统的状态进行估计; 根据状态估计结果设计输入-输出动态反馈控制规律, 使得跟随机器人(Follower) 准确跟踪领航机器人(Leader), 确保编队的稳定性和较快的收敛性, 并达到理想的编队控制效果. 仿真实验验证了所提出方法的可行性.

     

Tracking control of a flexible robot link

The formulation of an appropriate hybrid lumped/distributed model of a flexible robot link makes it possible to formulate and study tracking problems without the necessity for a priori approximations in the flexibility model for the link. The use of secondary support beam and active tip control provides perfect tracking with hypothetical exact measurements, and tracking to arbitrary accuracy with use of a variant of the acceleration feedback technique     

Decentralized adaptive tracking control of robot manipulators

An asymptotically stable decentralized adaptive control scheme is presented to enable accurate trajectory tracking without requiring specific knowledge about the robot dynamics. The scheme is based on expressing the robot dynamics as the product of individual joint quantities, and bounds on certain robot parameters. Parameter adaptation laws are derived using the Lyapunov theory, and asymptotic stability of tracking errors, and boundedness of parameter estimates are established. The control system is shown to be robust to torque disturbances affecting the system and to a class of unmodeled dynamics. The structure of the controller and the performance of the closed-loop system are analyzed. Simulations results using the complete dynamic model of a six degree of freedom industrial robot are presented to demonstrate the excellent tracking performance of the proposed adaptive control scheme. © 1996 John Wiley & Sons, Inc.     

三轮驱动移动机器人轨迹跟踪控制

针对三轮驱动移动机器人在轨迹跟踪控制过程中运动不平滑的问题,建立了移动机器人在一定运动约束条件下的运动学模型。根据移动机器人位姿误差微分方程的描述,设计了基于后退时变状态反馈方法的移动机器人轨迹跟踪控制器。基于李雅普诺夫方法,对轨迹跟踪控制器的稳定性进行了分析,证明了该控制器能够保证闭环系统全局一致渐进稳定。仿真结果验证了运动学模型的正确性,以及轨迹跟踪控制器的有效性。     

Stability analysis of position and force control for robot arms

Stability issues involving the control of a robot arm under the influence of external forces are discussed. Several different scenarios are considered: position control with the external force as an unmodeled disturbance, compliant control for a bounded external force in some subspace, and compliant control for a force due to the interaction with an environment whose dynamical behavior can be modeled. In each of these cases, a stability analysis using the Lyapunov method is presented. An explanation of instability is suggested in the case that the environment has flexibility and the gains are inappropriately chosen. When the environment is stiff in the force control subspace, robust (in time delay) stability can be achieved via the integral force feedback. However, the integral feedback gain should be chosen sufficiently small to account for possible flexibility in the system     

Perfomance-based adaptive tracking control of robot manipulators

This article presents two new adaptive schemes for the motion control of robot manipulators. The proposed controllers are very general and computationally efficient because they do not require knowledge of either the mathematical model or the parameter values of the manipulator dynamics, and are implemented without calculation of the robot inverse dynamics or inverse kinematic transformation. It is shown that the control strategies are globally stable in the presence of bounded disturbances, and that in the absence of disturbances the ultimate bound on the size of the tracking errors can be made arbitrarily small. Computer simulation results are given for a PUMA 560 manipulator, and demonstrate that accurate and robust trajectory tracking can be achieved by using the proposed controllers. Experimental results are presented for an IMI Zebra Zero manipulator and confirm that the control schemes provide a simple and effective means of obtaining high-performance trajectory tracking. © 1995 John Wiley & Sons, Inc.     

Modeling and robust force control of constrained one-link flexible arms

In this article modeling and robust force control of constrained flexible one-link arms on the basis of a distributed parameter model are discussed. Since the tip of the flexible arm contacts a given constraint surface, a constraint condition should be satisfied. By using the Lagrange multiplier method and the Hamilton's principle, we derive dynamic equations of the joint angle, the vibration of the flexible arm, and the constraint force. The boundary condition of the derived distributed parameter system is related to the contact force and is nonhomogeneous. We introduce a change of variables to derive a homogeneous boundary condition. On the basis of a finite-dimensional modal model of the distributed parameter system, we analyze the stability of the force feedback by using the root locus technique and the compliance control. To compensate the spillover instability an optimal controller with low-pass property and a robust H_∞ controller are constructed. Experiments have been carried out and results confirm that the controllers perform remarkably well. © 1998 John Wiley & Sons, Inc.     

Indirect robot model learning for tracking control

Learning task-space tracking control on redundant robot manipulators is an important but difficult problem. A main difficulty is the non-uniqueness of the solution: a task-space trajectory has multiple joint-space trajectories associated, therefore averaging over non-convex solution space needs to be done if treated as a regression problem. A second class of difficulties arise for those robots when the physical model is either too complex or even not available. In this situation machine learning methods may be a suitable alternative to classical approaches. We propose a learning framework for tracking control that is applicable for underactuated or non-rigid robots where an analytical physical model of the robot is unavailable. The proposed framework builds on the insight that tracking problems are well defined in the joint task- and joint-space coordinates and consequently predictions can be obtained via local optimization. Physical experiments show that state-of-the art accuracy can be achieved in both online and offline tracking control learning. Furthermore, we show that the presented method is capable of controlling underactuated robot architectures as well.     

Mouth tracking for hands-free robot control systems

In this paper, we propose a mouth tracking method for remote robot control systems. The main idea behind the work is to help disabled people, who cannot operate any keyboard or joystick, to control a robot without use of their hands. The mouth tracking method is mainly based on the AdaBoost feature detection approach. By adding new Haar-like features for detecting the corner of the mouth, the speed and accuracy of detection are improved. The AdaBoost feature detection combined with the Kalman filter accomplished continuous and accurate mouth tracking. Meanwhile, the gripping commands for the robot manipulator were obtained through recognition of mouth shape, such as for a pouting mouth or a grinning mouth. To assess the validity of the method, mouth detection experiments and robot cargo transport experiments were conducted. The results indicate that the proposed method can realize mouth tracking and robot operations that are quick and accurate in retrieving items successfully.     

机器人鲁棒轨迹跟踪控制系统

根据鲁棒控制理论和机器人的动态特性,针对机器人系统中存在的不确定性因素,利用不确定性的上界设计了一种鲁棒控制器,并将之用于机器人的跟踪轨迹控制,给出了仿真实验结果并与PID控制的结果进行了比较,仿真实验结果表明所设计的鲁棒控制器与PID控制器相比,具有很好的动态特性和很强的鲁棒性。     

Forcefree control with independent compensation for industrial articulated robot arms

Forcefree control of robot manipulators so far has been the passive motion of the arm due to the influence of external force under ideal conditions of zero gravity and zero friction, whereas this paper demonstrates forcefree control under assigned friction, gravity and inertia. The effectiveness of the proposed forcefree control with independent compensation is confirmed by comparing the experimental results with simulation results. Comparisons of the forcefree control with independent compensation to the other force control methods are also presented.     

Vibration control of a two-link flexible robot arm

Analysis and experimentation is described for a two-link apparatus in which both members are very flexible. Attention is focused on endpoint position control for point-to-point movements, assuming a fixed reference frame for the base, with two rotary joints. Each link is instrumented with acceleration sensing and is driven by a separate motor equipped with velocity and position sensing. The control perspective adopted is to implement a two-stage control strategy in which the vibration control problem for fine-motion endpoint positioning is considered separate from the gross-motion, large-angle slew problem. In the first stage the control law shapes the actuator inputsfor the large-angle movement in such a way that minimal energy is injected into the flexible modes, while in the second phase an endpoint acceleration feedback scheme is employed in independent joint controlfor vibration suppression at the link endpoints.     

Adaptive control of robot manipulators with flexible joints

Presents an adaptive control scheme for flexible joint robot manipulators. Asymptotic stability is insured regardless of the joint flexibility value, i.e., the results are not restricted to weak joint elasticity. Moreover, the joint flexibility is not assumed to be known. Joint position and velocity tracking errors are shown to converge to zero with all the signals in the system remaining bounded     

Modelling and control of non-slender flexible robot links

Based on Timoshenko's beam theory and Hamilton's principle, the dynamic model of a single non-slender flexible link rotating in a horizontal plane is derived. Spectral analysis is used to study the well-posedness of the dynamic control system. The tracking control problem is studied and a feedback control scheme, that controls the rigid-body motion and elastic behaviours simultaneously, is derived based on an n-modal model. The stabilization of the closed-loop system is studied analytically. Simulation results are presented