Predictive Control for a Type of UASs with Unmatched Disturbances Based on FAS Approaches

This study proposes a predictive control strategy for a type of under-actuated systems subject to unmatched disturbances based on the fully actuated system (FAS) approach. For articulated soft robot (ASR) systems, an elastic force encoding method grounded in energy shaping is employed to transform the under-actuated ASR system into a fully actuated one while preserving its Euler-Lagrange structure. Building on the transformed ASR system, where the unmatched disturbances are generated by an exogenous system, a continuous-time observer-based predictive controller is developed. The proposed controller ensures that the disturbance estimation error asymptotically converges to zero, and the closed-loop system achieves asymptotic stability. Finally, simulations on the ASR system are conducted to validate the effectiveness of the proposed control strategy.