Quadrotor transporting cable-suspended load using iterative Linear Quadratic regulator (iLQR) optimal control

Abstract

This paper presents an iterative linear quadratic regulator (iLQR) optimal controller of a high level dynamic model using Euler-Lagrange equations for a single quadrotor UAV with a cable-suspended heavy rigid body. The control algorithm is constructed in order to improve two possible outputs: firstly, precisely tracking a given desired trajectory for a quadrotor with load, and, secondly, considering an anti-swing load through a transporting task. Two challenges have been taken into account: aggressive trajectory tracking and stability in behaviour; these must be guaranteed for the full UAV carrying the payload. The high nonlinear dynamic model of the quadrotor with a consiseration of the cable suspended payload is represented in eleven degrees of freedom. Subsequently, the nonlinear dynamic model is linearised in two modes, a vehicle mode without a load effect during a taking-off task and a switching-to-quad-load system mode considering the load effect used in a simulation. The simulation results are presented to illustrate the system stability improvement and precise implementation through verifying the ILQR parameters in comparison with the LQR controller.