Lyapunov Function-guided Reinforcement Learning for Flight Control
Preprint
Abstract
A cascaded online learning flight control system has been developed and enhanced with respect to action smoothness. In this paper, we investigate the convergence performance of the control system, characterized by the increment of a Lyapunov function candidate. The derivation of this metric accounts for discretization errors and state prediction errors introduced by the incremental model. Comparative results are presented through flight control simulations.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Over one Century of Spectrum Analysis in Delay Systems: An Overview and New Trends in Pole Placement MethodsIFAC-PapersOnLine, 2022
- Multiplicity–Induced–Dominancy property for second–order neutral differential equations with application in oscillation dampingEuropean Journal of Control, 2022
- Robust modification of nonlinear L1 adaptive flight control system via noise attenuationAerospace Science and Technology, 2021
- Stability Analysis of Optimal Adaptive Control Using Value Iteration With Approximation ErrorsIEEE Transactions on Automatic Control, 2018
- Revisiting Approximate Dynamic Programming and its ConvergenceIEEE Transactions on Cybernetics, 2014
- Flight Dynamics and Hybrid Adaptive Control of Damaged AircraftJournal of Guidance, Control, and Dynamics, 2008
- Design and evaluation of robust nonlinear missile autopilots from a performance perspective2005
- Continuous pole placement for delay equationsAutomatica, 2002
- An effective algorithm for finite spectrum assignment of single-input systems with delaysMathematics and Computers in Simulation, 1998
- Stabilizability, detectability, and spectrum assignment for linear autonomous systems with general time delaysIEEE Transactions on Automatic Control, 1978