Dual-Evaluation Network-Based Adaptive Dynamic Programming Control for Flexible Flapping Wings

Le résumé fourni par la source

This paper focuses on addressing tracking control and vibration suppression issues in bio-inspired flapping-wing Unmanned Aerial Vehicles (UAVs). The solution is based on an Adaptive Dynamic Programming (ADP) control strategy. Flapping-wing UAVs are recognized for their efficient flight performance and superior agility. Nevertheless, they are vulnerable to uncertainties and actuator failures during operation. To address this issue, a novel ADP control strategy is proposed. This strategy incorporates a reference network that generates internal reinforcement signals. Additionally, a visual model of the rigidflexible coupled bio-inspired flapping wing is developed using the advanced system modeling tool MapleSim. Then, a dual-critic ADP controller is introduced to evaluate the effectiveness of the proposed strategy. The design of the reference network enables a better approximation of the desired cost function. This approach effectively suppresses wing vibrations and accurately tracks the desired trajectory. Finally, The closed-loop system's stability is analyzed using the Lyapunov direct method.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.