MPPI Control-based Adaptive Pursuit Guidance for Path-following Control of Quadrotors in the Presence of Wind Disturbance

Abstract

In this paper, an Model Predictive Path Integral (MPPI) control-based adaptive pursuit guidance for a path-following control of UAVs in the presence of wind disturbances is proposed. Typical pursuit guidance works well for the path-following problem with several design parameters, such as a guidance gain and virtual target point. However, the pursuit guidance law is not suitable for achieving specific objectives, such as minimizing control energy and distance from the reference path. Therefore, the design parameters of the guidance law should be selected differently according to the objectives and the shapes of the reference path. A concept of the proposed algorithm is to adaptively use the baseline guidance law using MPPI control to suit objectives in various path types. It also uses a nonlinear disturbance observer to estimate and reject the effects of wind disturbances. In simulation results, the costs of the proposed algorithm are lower or similar to the costs of the best combination of guidance paramters in the baseline guidance law according to the path and wind type. The result obtained shows that the proposed approach is effective in that it allows to utilize adaptively the baseline guidance law.