Название: Robot Control and Calibration: Innovative Control Schemes and Calibration Algorithms
Автор: Xin Luo, Zhibin Li, Long Jin, Shuai Li
Издательство: Springer
Серия: SpringerBriefs in Computer Science
Год: 2023
Страниц: 132
Язык: английский
Формат: pdf (true), epub
Размер: 29.1 MB

This book mainly shows readers how to calibrate and control robots. In this regard, it proposes three control schemes: an error-summation enhanced Newton algorithm for model predictive control; recurrent neural network (RNN) for solving perturbed time-varying underdetermined linear systems; and a new joint-drift-free scheme aided with projected ZNN, which can effectively improve robot control accuracy. Moreover, the book develops four advanced algorithms for robot calibration – Levenberg-Marquarelt with diversified regularizations; improved covariance matrix adaptive evolution strategy; quadratic interpolated beetle antennae search algorithm; and a novel variable step-size Levenberg-Marquardt algorithm – which can effectively enhance robot positioning accuracy.

We develop some control and calibration schemes for enhancing the tracking and positioning accuracy of the robot. The first model is a Model Predictive Control (MPC) scheme for robot trajectory tracking, which can optimize tracking error, velocity norm, and acceleration norm. The second model is a new recurrent neural network (RNN), which has better performance in dealing with time-varying underdetermined linear systems with double limits. The third model is a novel Joint Drift Free (JDF) scheme synthesized by the Projection Zeroing Neural Network (PZNN) model, which can effectively solve the motion generation and control of redundant manipulators under disturbance. The fourth model is an ensemble based on six different regularizations, which can obtain better calibration results and effectively solve the overfitting problem. The fifth model is a new algorithm based on an improved covariance matrix adaptive evolution strategy and an extended Kalman filter, which addresses the local optimum of the covariance matrix adaptive evolution strategy. The sixth model is a novel algorithm combining a quadratic interpolated beetle antennae search algorithm and an extended Kalman filter, which successfully suppresses noise during the robot calibration process and addresses the local optimum of the beetle antennae search algorithm. The last model is a new calibrator based on an unscented Kalman filter and a novel variable step-size Levenberg-Marquardt algorithm, which addresses the local optimum of the conventional Levenberg-Marquardt algorithm and suppresses the measurement noises.

This book first introduces how to calibrate and control a robot, which can help researchers and engineers fully understand robot control and robot calibration algorithm.