Bernard Hebert

An adaptive controller for a direct-drive SCARA robot

A direct adaptive controller for trajectory tracking of high-speed robots such as a direct-drive SCARA robot is presented. In this robot, nonlinear effects due to centrifugal, Coriolis, and inertial forces are more important than friction and gravity forces, unlike most industrial robots. The control law of the adaptive scheme consists of a PD regulator plus feedforward compensation of full dynamics. The feedforward terms are adjusted by an adaptation law so that the steady-state position errors are zero. With this adaptive controller, the joint acceleration measurement is not required and no inversion of the estimated mass matrix is involved. The tracking performances of the controller applied to a two-degree-of-freedom SCARA is illustrated by a real-time implementation based on a single-chip digital signal processor (DSP). >

A DSP-based adaptive controller for a smooth positioning system

The implementation of a self-tuning regulator for the positioning of a direct-drive servomotor is described. The servo motor is a permanent magnet DC motor in which no speed reducer is used. The auto-tuning regulator consists of two major loops. The inner loop contains a feedback (PD or PID) regulator with additional feedforward terms. The parameters of the feedforward compensation are adjusted by the outer loop, which contains an online parameter estimator. The estimator is based on a recursive least-squares equation, and the estimated parameters are the load inertia and viscous friction. This self-tuning regulator has been simulated with PC.MATLAB, and the results demonstrate the high performance of the scheme. Experimental results obtained with a small DC motor (Electrocraft E-576) are presented, and these results show good agreement with the digital simulation results. There are two innovative aspects to this work. First, parameter estimation is used to adapt the feedforward compensation terms instead of the gains of the feedback controller, as usually is the case in conventional indirect self-tuning regulators. Secondly, the complete adaptive controller has been implemented using a single-chip digital signal processor (DSP), which results in the reduction of system hardware and cost. >

A high efficiency interface for a biphase incremental encoder with error detection (servomotor control)

The use of incremental instead of absolute encoders for servomotor control involves the design of an interface to feed bidirectional counters. A robust method that the authors have implemented with a minimum number of components is presented. The method is also suitable for an all-software version that could be used in microcontroller applications with low CPU overhead. >

An adaptive controller for a direct-drive SCARA robot: Analysis and simulation

A direct adaptive controller for trajectory tracking of high-speed robots such as a direct-drive SCARA robot is presented. In this robot, nonlinear effects due to centrifugal, Coriolis, and gravity forces are more important than friction forces. The control law of the adaptive scheme consists in a PD regulator and feedforward compensation of full dynamics. The feedforward terms are adjusted by an adaptation law so that the steady-state position errors are zero. The regulator gain matrix can be fixed or time-varying; in the latter case, the adaptive scheme becomes similar to the computed torque scheme. However, with the adaptive controller, joint accelerations measurement is not required and no inversion of the estimated mass matrix is involved. The tracking performances of the controller applied to a 4 d.o.f. SCARA robot are illustrated by a digital simulation.<<ETX>>

An hybrid controller for a SCARA robot: Analysis and simulation

The authors present a hybrid controller that allows simultaneous force tracking in constrained directions and position tracking in unconstrained directions. The controller is an extension of a well-known adaptive position control scheme that exhibits remarkable performance in joint space. The hybrid controller is obtained by adding a commonly used force controller to the Cartesian space version of the adaptive position tracking controller. Analysis and simulation are used to evaluate the hybrid controllers ability to execute trajectories using force and position feedback. The application robot is a four-degree-of-freedom SCARA.<<ETX>>

Digital simulation of a smooth positioning system with parameter estimation (permanent magnet synchronous motors)

The authors present the results of a digital simulation of a smooth positioning system for a self-controlled permanent-magnet synchronous motor. The parameter estimation that is included in the controls is based only on the position feedback. The use of predictive control and parameter estimation enables fast, accurate, and nonoscillatory positioning of a directly driven load varying over a wide range.<<ETX>>