Resilient Estimation and Safe Planning for UAVs in GPS-Denied Environments

Unmanned aerial vehicles (UAVs) suffer from intolerable sensor drifts in global positioning system (GPS)-denied environments, leading to potentially dangerous situations. This chapter proposes a safety-constrained control framework that adapts UAVs at a path re-planning level to support resilient state estimation in GPS-denied environments. The proposed framework consists of an anomaly detector, a resilient state estimator, a robust controller, a pursuer location tracker (PLT), and an escape controller (EsC). The detector ensures anomaly detection and provides a switching criterion between the robust control and emergency control modes. PLT is developed to track the pursuer’s location by the unscented Kalman filter with sliding window outputs. Using the estimates from PLT, we design an EsC based on the model predictive controller such that the UAV escapes from the effective range of the spoofing device within the escape time that is defined as the safe time within which the estimation errors remain in a tolerable region with high probability. Subsequently, the proposed framework is extended for the multi-UAV systems that perform the time-critical coordination task.