Neural Network Model-Based Control for Manipulator: An Autoencoder Perspective
- 14 September 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 34 (6), 2854-2868
- https://doi.org/10.1109/tnnls.2021.3109953
Abstract
Recently, neural network model-based control has received wide interests in kinematics control of manipulators. To enhance learning ability of neural network models, the autoencoder method is used as a powerful tool to achieve deep learning and has gained success in recent years. However, the performance of existing autoencoder approaches for manipulator control may be still largely dependent on the quality of data, and for extreme cases with noisy data it may even fail. How to incorporate the model knowledge into the autoencoder controller design with an aim to increase the robustness and reliability remains a challenging problem. In this work, a sparse autoencoder controller for kinematic control of manipulators with weights obtained directly from the robot model rather than training data is proposed for the first time. By encoding and decoding the control target though a new dynamic recurrent neural network architecture, the control input can be solved through a new sparse optimization formulation. In this work, input saturation, which holds for almost all practical systems but usually is ignored for analysis simplicity, is also considered in the controller construction. Theoretical analysis and extensive simulations demonstrate that the proposed sparse autoencoder controller with input saturation can make the end-effector of the manipulator system track the desired path efficiently. Further performance comparison and evaluation against the additive noise and parameter uncertainty substantiate robustness of the proposed sparse autoencoder manipulator controller.Keywords
This publication has 46 references indexed in Scilit:
- Fuzzy Adaptive Inverse Compensation Method to Tracking Control of Uncertain Nonlinear Systems With Generalized Actuator Dead ZoneIEEE Transactions on Fuzzy Systems, 2017
- Distributed Recurrent Neural Networks for Cooperative Control of Manipulators: A Game-Theoretic PerspectiveIEEE Transactions on Neural Networks and Learning Systems, 2016
- Deep Learning of Part-Based Representation of Data Using Sparse Autoencoders With Nonnegativity ConstraintsIEEE Transactions on Neural Networks and Learning Systems, 2015
- Taylor $O(h^{3})$ Discretization of ZNN Models for Dynamic Equality-Constrained Quadratic Programming With Application to ManipulatorsIEEE Transactions on Neural Networks and Learning Systems, 2015
- Neural Network-Based Control of Networked Trilateral Teleoperation With Geometrically Unknown ConstraintsIEEE Transactions on Cybernetics, 2015
- Missile Guidance Law Based on Robust Model Predictive Control Using Neural-Network OptimizationIEEE Transactions on Neural Networks and Learning Systems, 2014
- Exponential stability of globally projected dynamic systemsIEEE Transactions on Neural Networks, 2003
- A neural network for monotone variational inequalities with linear constraintsPhysics Letters A, 2003
- A new formulation technique for local torque optimization of redundant manipulatorsIEEE Transactions on Industrial Electronics, 1996
- Kinetic limitations on the use of redundancy in robotic manipulatorsIEEE Transactions on Robotics and Automation, 1991