Setpoint Tracking for the Suspension System of Medium-Speed Maglev Trains via Reinforcement Learning
- 1 July 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1620-1625
Abstract
In this paper, we consider setpoint tracking for the suspension system of medium-speed maglev trains for keeping constant distance above the track. Due to complicated environment, unknown disturbances and strong nonlinear coupling of model, the problem can not be effectively solved by most of the model-based controllers. To this purpose, we formulate the setpoint tracking problem as continuous-state, continuous-action Markov decision processes under unknown transition probabilities. Based on the deterministic policy gradient and neural network approximation, we propose a model-free reinforcement learning (RL) algorithm that learns a state-feedback controller from sampled data of the suspension system. We illustrate with simulations that our model-free method has high performance.Keywords
This publication has 18 references indexed in Scilit:
- Robust control of nonlinear MAGLEV suspension system with mismatched uncertainties via DOBC approachISA Transactions, 2011
- Reinforcement learning for robot soccerAutonomous Robots, 2009
- Nonlinear robust control design for levitation and propulsion of a maglev systemIEE Proceedings - Control Theory and Applications, 2004
- Dynamic balance of a biped robot using fuzzy reinforcement learning agentsFuzzy Sets and Systems, 2003
- Feedback linearization of an active magnetic bearing with voltage controlIEEE Transactions on Control Systems Technology, 2002
- Adaptive nonlinear control of repulsive maglev suspension systemsControl Engineering Practice, 2000
- H/sub 2/ and H/sub /spl infin// control for MagLev vehiclesIEEE Control Systems, 1998
- Linearizing control of magnetic suspension systemsIEEE Transactions on Control Systems Technology, 1997
- Design and analysis of the nonlinear feedback linearizing control for an electromagnetic suspension systemIEEE Transactions on Control Systems Technology, 1997
- Self-improving reactive agents based on reinforcement learning, planning and teachingMachine Learning, 1992