An Adaptive Neural Network-Based Fixed Allocation Scheme for a UAV Octorotor

This study introduces a passive faulttolerant control scheme that integrates an adaptive neural network (NN)-based modulation gain with an Integral Sliding Mode Control (ISMC) law to enhance system robustness and adaptability. The key contribution lies in leveraging the NN-based adaptive law to dynamically adjust the sliding gain, ensuring optimal performance under varying system dynamics and mitigating the chattering effect commonly associated with conventional SMC approaches. A fixed control allocation scheme, independent of actuator effectiveness levels, is utilized to convert virtual control signals into physical actuator demands under faulty conditions. The efficacy of the scheme is validated on a numerical model octocopter UAV model. Various simulations under different fault and failure scenarios validate the efficacy of the proposed method.