SL1-Simplex: Safe Velocity Regulation of Self-Driving Vehicles in Dynamic and Unforeseen Environments
This paper proposes a novel extension of the Simplex architecture with model switching and model learning to achieve safe velocity regulation of self-driving vehicles in dynamic and unforeseen environments. To guarantee the reliability of autonomous vehicles, an $\mathcal{L}_{1}$ adaptive controller that compensates for uncertainties and disturbances is employed by the Simplex architecture as a verified safe controller to tolerate concurrent software and physical failures. Meanwhile, safe switching controller is incorporated into the Simplex for safe velocity regulation through the integration of the traction control system and anti-lock braking system. Specifically, the vehicle's angular and longitudinal velocities asymptotically track the provided references that vary with driving environments, while the wheel slips are restricted to safety envelopes to prevent slipping and sliding. Due to the high dependence of vehicle dynamics on the driving environments, the proposed Simplex leverages the finite-time model learning to timely learn and update the vehicle model for $\mathcal{L}_{1}$ adaptive controller, when any deviation from the safety envelope or the uncertainty measurement threshold occurs in the unforeseen driving environments. Finally, the effectiveness of the proposed Simplex architecture for safe velocity regulation is validated by the AutoRally platform.