With the objective of exposing hidden design deficiencies in complex nonlinear systems, this paper presents the use of reinforcement learning techniques for application in flight control law development and testing. Following the idea of worst-case testing, a deep Q-network agent is trained to identify input sequences that lead to detrimental system behavior. Because the analysis is based directly on the repeated interaction between the agent and the investigated system, no model simplifications are required, making the presented method applicable to highly complex systems. The capability of the learning-based worst-case analysis is demonstrated for the speed protection function of the hover flight control law of an electric vertical takeoff and landing (eVTOL) aircraft. The analysis discovers possible piloted maneuvers that violate the implemented protection algorithm. A root cause analysis of the emerging behavior reveals the neglect of an important flight mechanical coupling term in the design of the protection algorithm and ultimately leads to the revision and improvement of the controller. This demonstrates the benefits of the presented testing method for the design, verification, and validation of complex systems. The application to a high-fidelity system used for control law development of an actual eVTOL prototype currently under construction demonstrates the relevance of the method beyond academia.

This study proposes a novel, nonlinear trajectory/path-following controller based on jerk-level error dynamics. Therefore, at first the nonlinear acceleration-based kinematic equations of motion of a dynamic system are differentiated with respect to time to obtain a representation connecting the translation jerk with the (specific) force derivative. Furthermore, the path deviation, i.e., the difference between the planned and the actual path, is formulated as nonlinear error dynamics based on the jerk error. Combining the derived equations of motion with the nonlinear error dynamics as well as employing nonlinear dynamic inversion, a control law can be derived that provides force derivative commands, which may be commanded to an inner loop for trajectory control. This command ensures an increased smoothness and faster reaction time compared to traditional approaches based on a force directly. Furthermore, the nonlinear parts in the error dynamic are feedforward components that improve the general performance due to their physical connection with the real dynamics. The validity and performance of the proposed trajectory/path-following controller are shown in an aircraft-related application example.