Adaptive Trajectory Tracking Control of a UAV using Gaussian Processes

Abstract: Unmanned aerial vehicles are a popular choice for various transport and monitoringapplications. In many outdoor applications frequent variations of thevehicle conguration and time-varying disturbances originating from the environmentare expected. In order to counteract resulting deviations from aspecied trajectory both model-based and data-driven methods are utilised.With the performance expectations encoded in a designer chosen trajectory,the implementation of a nonlinear controller based on geometric methods onthe nonlinear Lie group SE(3) guarantees stability for the nominal system. Incombination with an adaptive compensation term using a radial basis functionnetwork (RBFN) and a learning-based term using a Gaussian Process, alearning-based framework is proposed. The approach is evaluated with respectto its real time applicability and numerical results are provided for the adaptiveRBFN control approach.