Abstract. In trajectory control problems for airplane-type unmanned aerial vehicles (UAVs), it is important to ensure the smoothness of reference trajectories. Feedback control design generally requires knowledge of their derivatives. If reference actions arrive as real-time signals fr om an autonomous source, the values of their derivatives must be reconstructed considering constraints on the UAV velocity and acceleration. For the combined smoothing and differentiation of reference actions, an autonomous dynamic generator with corrective actions in the form of bounded and smooth sigma functions with constant gains is developed. They are tuned universally under specified worst-case constraints, i.e., the maximum admissible absolute values of the UAV velocity and acceleration. In this case, a nonsmooth reference trajectory is tracked and smoothed by the output variables of the generator under specified design constraints, albeit with significant lower bounds for the tracking error. As shown, the approximation accuracy of a reference trajectory can be improved without violating the design constraints on those segments wh ere the factual rate of change of the reference trajectory is less than the maximum admissible. The scientific novelty of this paper consists in the development of a reference trajectory generator with adaptively tuned gains without dynamic order expansion. Simulation results are presented to confirm the effectiveness of the algorithms developed.
Keywords: dynamic smoothing and differentiation of signals, sigmoid control, adaptive gain tuning, UAV.
Acknowledgments. This work was supported in part by the Russian Science Foundation (project no. 25-11-00067), https://rscf.ru/project/25-11-00067/.
An Adaptive Generator of Reference Trajectories Satisfying UAV’s Dynamic Constraints