Neural network compensation for force tracking control of an autonomous helicopter system

In this paper, neural network is used to compensate for uncertainties in an autonomous aerial helicopter system when a force control technique is applied to the environment. Applying the force control technique to a helicopter system is quite difficult since both position and force are regulated. To perform force tracking tasks well, position control should be done first. The speed of the helicopter is controlled by the LQR method, and the position is controlled by closing the outer loop to form a PD controlled system. The force control is applied to the position controlled system. The adaptive impedance force control algorithm is applied and tested to control the desired force under unknown location and stiffness of the environment. Simulation studies show that neural network rejects outer disturbances successfully.