🤖 AI Summary
This study addresses the challenge of sustained spacecraft attitude stabilization under unseen actuator faults—specifically gain errors, sign reversals, and constant biases—by introducing a novel evaluation paradigm centered on a “stability gate” metric that prioritizes long-term stability over instantaneous recovery. The proposed approach integrates recurrent neural networks for online estimation, an analytical control law, Nussbaum-gain adaptation, a disturbance observer, and multi-source sensor fusion into a structured estimation-and-control architecture. Validated on the Basilisk six-degree-of-freedom platform, the framework achieves success rates of 97.8% and 94.4% under sign-reversal and gain-fault scenarios, respectively. Notably, the inclusion of the disturbance observer elevates the success rate for constant-bias faults from 0% to 59.4%, substantially outperforming baseline methods such as end-to-end reinforcement learning and conventional PID control.
📝 Abstract
Recent learned fault-tolerant-control (FTC) work reports high success on spacecraft actuator faults, but often in simulation, on narrow fault sets, and with transient metrics that a trajectory need only touch once. We ask what recovers spacecraft pointing when success means holding it on faults never seen in training. We answer with a benchmark built around a settled gate, pointing held within 0.2 deg over a dwell window and scored on the true state, train/test splits disjoint in inertia, gain, sign pattern, and bias, Wilson intervals over n=500 episodes per cell, and one-command reproduction on a 6-DOF Basilisk testbed. Across classical, adaptive, learned end-to-end, and structured controllers, three findings stand out. Fault-unaware PD/PID and from-scratch end-to-end RL score 0%, so learning capacity alone is not the lever. Classical adaptive laws resolve sign faults but handle gain poorly at 55.2%, and a literature-faithful Nussbaum-gain law reaches 45.2% and 3.2%. A structured estimate-then-control design, with a learned recurrent module that infers actuator gain online and feeds an analytic law, wins on sign and gain faults at 97.8% and 94.4%, approaching the privileged oracle while unstructured methods remain at zero. The hard wall is constant additive bias, which is 0% for every controller including the privileged gain oracle, because an integral-free law cannot null a constant disturbance. We close it with a disturbance observer that recovers bias from the dynamics and is self-correcting for gain-estimate error. Composed with the gain estimate, it recovers 59.4% of held-out bias faults with no sign/gain regression, moving that class off zero. We classify sensor-fault regimes similarly, show that sensor bias is unobservable from the corrupted measurement alone and therefore requires fusion rather than an observer, and release the benchmark so the gate is shared.