Fixed-Time Prescribed Performance-Based Adaptive Neural Safe Control for QUAV Under Flight Environment Constraints

The real-time obstacle avoidance in unpredictable environments has consistently been a practical engineering challenge that needs to be addressed in the safe flight control of quadrotor unmanned aerial vehicles (QUAVs). This article proposes a fixed-time prescribed performance-based adaptive neural safe control scheme for QUAV under flight environment constraints, system uncertainties, and external disturbances. By integrating a fixed-time command filter, a novel fixed-time boundary protection algorithm is proposed to generate a safe desired flight path timely. On this basis, a fixed-time prescribed performance function and error transformation are utilized to ensure both transient and steady state performance of the QUAV system. Additionally, a radial basis function neural network and an adaptive neural disturbance observer are codesigned to address the effects of system uncertainties and external disturbances. The boundedness and safe flight performance of the closed-loop QUAV system within a specified time frame can be guaranteed through Lyapunov stability analysis. Experimental results are presented to demonstrate the effectiveness of the proposed control scheme.