Learning self-triggered controllers with Gaussian processes
arXiv:1909.00178
Abstract
This paper investigates the design of self-triggered controllers for networked control systems (NCSs), where the dynamics of the plant is \textit{unknown} apriori. To deal with the unknown transition dynamics, we employ the Gaussian process (GP) regression in order to learn the dynamics of the plant. To design the self-triggered controller, we formulate an optimal control problem, such that the optimal control and communication policies can be jointly designed based on the GP model of the plant. Moreover, we provide an overall implementation algorithm that jointly learns the dynamics of the plant and the self-triggered controller based on a reinforcement learning framework. Finally, a numerical simulation illustrates the effectiveness of the proposed approach.
appear in IEEE Transactions on Cybernetics
References in corpus (6)
- Gaussian Processes for Data-Efficient Learning in Robotics and Control
- An ISS self-triggered implementation of linear controller
- Stable Gaussian Process based Tracking Control of Euler-Lagrange Systems
- Stochastic data-driven model predictive control using Gaussian processes
- Self-triggered Model Predictive Control for Nonlinear Input-Affine Dynamical Systems via Adaptive Control Samples Selection
- Cautious NMPC with Gaussian Process Dynamics for Autonomous Miniature Race Cars