Safe Learning-based Tracking Control for Quadrotors under Wind Disturbances
This work addresses safety-critical control for aerial robotics in unpredictable conditions, representing an incremental improvement by integrating learning and control barrier functions into existing cascaded QP frameworks.
The paper tackled safe trajectory tracking for quadrotors under wind disturbances by proposing a learning-based safety-preserving cascaded quadratic programming control (SPQC) that uses Gaussian Processes to estimate uncertainties and control barrier functions to enforce safety constraints, with performance validated through numerical simulations in scenarios with wind disturbances and cluttered environments.
Enforcing safety on precise trajectory tracking is critical for aerial robotics subject to wind disturbances. In this paper, we present a learning-based safety-preserving cascaded quadratic programming control (SPQC) for safe trajectory tracking under wind disturbances. The SPQC controller consists of a position-level controller and an attitude-level controller. Gaussian Processes (GPs) are utilized to estimate the uncertainties caused by wind disturbances, and then a nominal Lyapunov-based cascaded quadratic program (QP) controller is designed to track the reference trajectory. To avoid unexpected obstacles when tracking, safety constraints represented by control barrier functions (CBFs) are enforced on each nominal QP controller in a way of minimal modification. The performance of the proposed SPQC controller is illustrated through numerical validations of (a) trajectory tracking under different wind disturbances, and (b) trajectory tracking in a cluttered environment with a dense time-varying obstacle field under wind disturbances.