Design and Implementation of an Artificial Neural Network Wavelet for Load Transportation with two Unmanned Aircraft Systems

Abstract

This paper deals with the problem of enabling a team of two Unmanned Aircraft Systems to perform an autonomous load transportation task. We propose a strategy whose main objective is to ensure a stable flight of the team of agents when cooperatively transporting the load. The strategy is based on the implementation of a neural network controller and a neural network estimator, using wavelet activation functions in combination with a cooperative multi-agent control approach. To show the effectiveness and applicability of the proposed framework, a real time experimental implementation is presented. Additionally, a comparison with respect to a classic control law is also provided, demonstrating the superior performance attained when adopting the proposed controller.