Fault-Tolerant Control Framework for Quadcopters With Implementation-Decoupled Fault Diagnosis and Continuum Control Allocation

Fault-tolerant control (FTC) for quadcopters has been drawing increased attention due to flight safety demands. However, a significant challenge in active FTC is that current fault detection and diagnosis (FDD) methods are often tightly coupled with specific controller architectures, which escalates implementation complexity across diverse platforms. Furthermore, the accuracy of diagnosis remains susceptible to actuator parameters, such as motor time constants. Moreover, prevailing FTC strategies frequently overlook the effective utilization of residual thrust from rotors experiencing loss-of-effectiveness (LoE) faults. To address these challenges, this article introduces a novel FTC framework that incorporates implementation-decoupled FDD (ID-FDD) and continuum control allocation (CCA). The ID-FDD approach achieves diagnosis decoupling from controller implementations and actuator dynamics by reconstructing rotor thrust through inversion-based estimation and quantifying LoE faults via extremum-characteristic analysis. CCA dynamically adapts to escalating LoE faults by progressively sacrificing yaw authority to prioritize translational control, thereby enabling full thrust utilization across 0–100% LoE without imposing a high computational burden. Experimental results demonstrate that ID-FDD achieves rapid fault diagnosis (Formula presented) during high-rate body rotation (Formula presented) using only a standard PID controller, with simulations confirming insensitivity to motor time constants (0.05–0.5 s), while CCA maintains trajectory tracking (RMSE < 0.08 m) at 75% LoE by utilizing 25% residual thrust from the faulty rotor while reducing yaw rate through thrust vector coordination. The corresponding demonstration video can be found at https://youtu.be/zDE3Cqhs2oY.