Geometric Adaptive Control for a Quadrotor-Slung Load System with Neural Network Estimation

This paper presents a geometric adaptive control strategy for an underactuated quadrotor-slung load system that enables accurate trajectory tracking of the load. This work is motivated by the increasing demand for slung-load transportation in low-altitude logistics, leveraging the demonstrated capabilities of quadrotors in diverse applications. A comprehensive mathematical model is derived to account for dynamic coupling effects and external disturbances. The load position is regulated by exploiting the system’s differential flatness, whereas the quadrotor attitude and cable direction are governed by the controller dynamics. The proposed geometric adaptive control strategy comprises a feedback control law for load position tracking and a radial basis function neural network for compensating external disturbance. The tracking performance and robustness of the proposed control strategy are validated via numerical simulations.