A multilayer neural-net (NN) controller for a general serial-link rigid robot arm is developed. The structure of the NN controller is derived using a filtered error/passivity approach. No off-line learning phase is needed for the proposed NN controller and the weights are easily initialized. The nonlinear nature of the NN, plus NN functional reconstruction inaccuracies and robot disturbances, mean that the standard delta rule using backpropagation tuning does not suffice for closed-loop dynamic control. Novel online weight tuning algorithms, including correction terms to the delta rule plus an added robust signal, guarantee bounded tracking errors as well as bounded NN weights. Specific bounds are determined, and the tracking error bound can be made arbitrarily small by increasing a certain feedback gain. The correction terms involve a second-order forward-propagated wave in the backpropagation network. New NN properties including the notions of a passive NN, a dissipative NN, and a robust NN are introduced.

Multilayer neural-net robot controller with guaranteed tracking performance

A stabilizing adaptive controller for a nonlinear system depending affinely on some unknown parameters is presented. It is assumed that this system is feedback stabilizable. A key feature of the method is the use of the Lyapunov equation to design the adaptive law. A result on local stability, two different conditions for global stability, and a local result where the initial conditions of the state of the system only are restricted are given. >

Adaptive nonlinear regulation: estimation from the Lyapunov equation

The exploitation of kinematic redundancies in robotic systems may provide more dexterity and versatility in the execution of complex tasks. When functional constraint tasks are imposed in addition to the end-effector tasks, a task priority strategy is advisable. The authors propose a general framework for managing multiple tasks in highly redundant systems. In particular, they derive joint velocity and acceleration solutions which can be used as reference input trajectories to suitable model-based controllers. They also develop a recursive implementation, and discuss the occurrence of singularities in the Jacobian associated with the generic task. Two case studies illustrate the effectiveness of the algorithm on a snake-like robot. >

A general framework for managing multiple tasks in highly redundant robotic systems

A neural net (NN) controller for a general serial-link robot arm is developed. The NN has two layers so that linearity in the parameters holds, but the "net functional reconstruction error" and robot disturbance input are taken as nonzero. The structure of the NN controller is derived using a filtered error/passivity approach, leading to new NN passivity properties. Online weight tuning algorithms including a correction term to backpropagation, plus an added robustifying signal, guarantee tracking as well as bounded NN weights. The NN controller structure has an outer tracking loop so that the NN weights are conveniently initialized at zero, with learning occurring online in real-time. It is shown that standard backpropagation, when used for real-time closed-loop control, can yield unbounded NN weights if (1) the net cannot exactly reconstruct a certain required control function or (2) there are bounded unknown disturbances in the robot dynamics. The role of persistency of excitation is explored. >

Neural net robot controller with guaranteed tracking performance

Composite adaptive control of robot manipulators

The author's previous work (1986, 1987) utilized the particular structure of manipulator dynamics to develop a simple, globally convergent adaptive controller for manipulator trajectory control problems. After summarizing the basic algorithm, they demonstrate the approach on a high-speed two-degree-of-freedom semi-direct-drive robot. They show that the dynamic parameters of the manipulator, assumed to be initially unknown, can be estimated within the first half second of a typical run, and that accordingly, the manipulator trajectory can be precisely controlled. These experimental results demonstrate that the adaptive controller enjoys essentially the same level of robustness to unmodeled dynamics as a PD (proportional and differential) controller, yet achieves much better tracking accuracy than either PD or computed-torque schemes. Its superior performance for high-speed operations, in the presence of parametric and nonparametric uncertainties, and its relative computational simplicity, make it an attractive option both for addressing complex industrial tasks, and for simplifying high-level programming of more standard operations. >

Li Weiping

Papers

Adaptive manipulator control: A case study