Based on the incremental model, this paper investigates an adaptive dynamic programming (ADP) control scheme for reentry vehicles. First, the attitude dynamics of the reentry vehicle and its incremental form are established, then the ADP-based optimal controller is devised with reference to the dual heuristic dynamic programming (DHP) framework. Expediently, some local incremental parameters that the controller requires can be estimated through recursive least squares (RLS) technology online, replacing the model network that needs to be globally trained in advance in DHP. In addition, a target critic network is introduced to reduce the impact of state changes on the network parameters during the internal iterations of a time step, so it improves the learning stability. Finally, this control strategy is implemented to an attitude control system of the reentry vehicle, and online simulations verify its effectiveness.
Attitude Control of Reentry Vehicle Based on Adaptive Dynamic Programming with Incremental Model
Lect. Notes Electrical Eng.
30.10.2021
14 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Attitude Control of Reentry Vehicle Based on Adaptive Dynamic Programming with Incremental Model
British Library Conference Proceedings | 2022
|Reentry Vehicle Formation Control Method Based on Adaptive Dynamic Programming
Springer Verlag | 2024
|Adaptive fault-tolerant attitude control for reentry vehicle involving actuator saturation
SAGE Publications | 2019
|