Data-Driven Bias-Lyapunov Iteration for Optimal Control of Unknown Markovian Jump Linear Systems
- 29 December 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 73 (5) , 3667-3679
Abstract
In this paper, a bias-Lyapunov iteration method is proposed to solve the optimal control problem of unknown Markovian jump linear systems. By incorporating a bias parameter into the conventional Lyapunov iteration method, the proposed method eliminates initial admissible control requirements. A model-based theoretical framework is subsequently established, accompanied by a rigorous convergence proof for the modified iteration process. Subsequently, a data-driven version of bias-Lyapunov iteration is developed to learn an optimal control for Markovian jump linear systems with completely unknown dynamics. Simulation examples validate the efficacy and advantage of the proposed bias-Lyapunov iteration method.Keywords
Funding Information
- National Science Fund for Distinguished Young Scholars (62125303)
- National Natural Science Foundation of China-“Qisun Ye” Science Foundation (U2441243)
- National Natural Science Foundation of China (62303132)
- National Natural Science Foundation of China (62573165)
- National Natural Science Foundation of China (62188101)
- National Natural Science Foundation of China (62173111)
- Heilongjiang Post-Doctoral Fund (LBH-Z23148)
- China Post-Doctoral Science Foundation (2024M754186)
- Heilongjiang Touyan Team Program
This publication has 43 references indexed in Scilit:
- Reinforcement learning-based optimal control for Markov jump systems with completely unknown dynamicsAutomatica, 2024
- Newton’s method for coupled continuous-time algebraic Riccati equationsJournal of Applied Mathematics and Computing, 2024
- Modified Riccati iterative algorithms for stochastic coupled algebraic Riccati equations of linear stochastic Markovian jump systemsAsian Journal of Control, 2024
- Iterative algorithms with the latest update for Riccati matrix equations in Itô Markov jump systemsScience China Technological Sciences, 2020
- A novel iterative algorithm for solving coupled Riccati equationsApplied Mathematics and Computation, 2020
- Value iteration and adaptive dynamic programming for data-driven adaptive optimal control designAutomatica, 2016
- A new iterative algorithm for solvingH∞control problem of continuous-time Markovian jumping linear systems based on online implementationInternational Journal of Robust and Nonlinear Control, 2016
- Computational adaptive optimal control for continuous-time linear systems with completely unknown dynamicsAutomatica, 2012
- Stochastic optimal control of unknown linear networked control system in the presence of random delays and packet lossesAutomatica, 2012
- Positive operator based iterative algorithms for solving Lyapunov equations for Itô stochastic systems with Markovian jumpsApplied Mathematics and Computation, 2011