Randomized algorithms for probabilistic analysis of parametric uncertainties with unmanned helicopters

Abstract

This paper aims at the development of a randomized algorithm based probabilistic analysis approach for parametric uncertainties in unmanned helicopter systems. The proposed approach is developed considering the stochastic characterization of bounded uncertainty in the system assuming that the plant dynamics are exactly known. This provides a new paradigm for synthesizing the controller gain to solve the problem of trajectory tracking for unmanned. Further, to assess the operation of the proposed randomization algorithm-based probabilistic controller and achieve the controller synthesis, a two degrees of freedom (2DoF) helicopter system is modelled and operated with uncertainties for different trajectories. Besides, the robustness of the controller operating under uncertainties is verified with reachability analysis developed on reach tubes and reach sets of the ellipsoidal method. The results identified the efficiency of the proposed approach with time domain characteristics for both simulation and real-time experiments. Moreover, a comparative assessment of the projected approach with conventional techniques is carried out to validate the controller response on the helicopter model.