Adaptive trajectory tracking of UAV with a cable-suspended load using vision-inertial-based estimation

Unmanned aerial vehicle (UAV) carrying cable-suspended load is a basic method for load transportation. Tracking a given trajectory without the global positioning system or a prior environmental map with limited computation power and limited sensory is a crucial problem. In this paper, a novel paradigm/structure combining an adaptive state estimator and a trajectory tracking controller for the unmanned aerial vehicle with a cable-suspended load system is designed. An advantage of this work is the adaptive method to estimate the system state and environmental feature positions at the same time. The asymptotic convergence of our system is theoretically proved by the Lyapunov theory. The visual–inertial estimation algorithm we propose has lower computational consumption than the optimization-based methods and provides high-precision, low latency state estimation for the system executing aggressive flight. Several simulations are designed and carried out within an ODE-based simulator to verify the proposed method. Comparisons with other visual–inertial-based algorithms are also carried out in the simulator. Simulation results demonstrate the proposed algorithm’s convergence, robustness, and efficiency.