Adaptive Dynamic Programming for Optimal Control of Discrete-Time Nonlinear Systems With Trajectory-Based Initial Control Policy
- 10 November 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Systems, Man, and Cybernetics: Systems
- Vol. 54 (3) , 1489-1501
Abstract
The policy gradient adaptive dynamic programming (PGADP) technique has gained recognition as an effective approach for optimizing the performance of nonlinear systems. Nonetheless, existing PGADP algorithms often demand a substantial volume of expensive or potentially risky interaction data with the system. Moreover, the utilization of neural networks in these algorithms can result in suboptimal learning efficiency and unstable training procedures. To address these challenges, a novel algorithm, referred to as OptNet-PGADP, has been introduced. This algorithm integrates an initially tailored control policy based on OptNet to tackle the optimization of control problems in discrete-time nonlinear systems. The OptNet-PGADP algorithm operates through a two-step process. Initially, the input–output trajectory of the system is computed using the nonlinear model predictive control (NMPC) method. Subsequently, an initial admissible control policy is acquired through OptNet. This policy is iteratively enhanced using the PGADP algorithm to attain the optimal controller. The resulting closed-loop control policy can be readily deployed in real-time applications. The implementation of the algorithm employs OptNet for the actor network and integrates an experience replay mechanism to bolster the controller’s learning efficiency. Furthermore, a convergence and optimality analysis of the algorithm is included. Simulation and experimental results conducted on two nonlinear systems conclusively demonstrate that the approach outperforms traditional PGADP and NMPC algorithms. These findings underscore the efficacy of OptNet-PGADP in mitigating the constraints of current methods and achieving superior control performance for nonlinear systems.Keywords
Funding Information
- National Natural Science Foundation of China (62273234)
This publication has 25 references indexed in Scilit:
- Identification and optimal control of nonlinear systems using recurrent neural networks and reinforcement learning: An overviewNeurocomputing, 2021
- Generalized value iteration for discounted optimal control with stability analysisSystems & Control Letters, 2021
- Generalized Risk-Sensitive Optimal Control and Hamilton–Jacobi–Bellman EquationIEEE Transactions on Automatic Control, 2020
- Data-Driven Iterative Adaptive Critic Control Toward an Urban Wastewater Treatment PlantIEEE Transactions on Industrial Electronics, 2020
- Optimal Finite-Horizon Control for Networked Control Systems in the Presence of Random Delays and Packet LossesIEEE Control Systems Letters, 2020
- Deterministic policy gradient adaptive dynamic programming for model-free optimal controlNeurocomputing, 2020
- An Approximate Neuro-Optimal Solution of Discounted Guaranteed Cost Control DesignIEEE Transactions on Cybernetics, 2020
- Robust Guaranteed Cost Control Under Digital Communication ChannelsIEEE Transactions on Industrial Informatics, 2019
- Adaptive $Q$ -Learning for Data-Based Optimal Output Regulation With Experience ReplayIEEE Transactions on Cybernetics, 2018
- Adaptive Neural Network Control of AUVs With Control Input Nonlinearities Using Reinforcement LearningIEEE Transactions on Systems, Man, and Cybernetics: Systems, 2017