Algorithm for controlling the movement of a multicopter as part of a group based on signals from the technical vision system

Abstract

The paper analyzes the control of a group of multicopters during autonomous flight along an arbitrary trajectory determined by the leader. Under such a situation, the most acceptable option is the leader- follower model. Determining the relative location of the leader and the follower, as well as determining the parameters for controlling the movement of the follower, is proposed to be carried out using a technical vision system. For this, the leader is marked with a special label. A general flight control algorithm has been developed, which includes all its main stages from pre-flight preparation to landing. Controlling the movement of the follower is carried out by adjusting the angles of yaw, roll, pitch and changing the movement speed depending on the distance to the leader. The paper proposes formulas and algorithm for determining these parameters according to the system of technical vision and inertial navigation system of the follower. To test the proposed algorithms, a model was created in the Simulink software environment of the Matlab system. Experimental studies using a camera with a resolution of 1280x720 confirmed the correctness of the obtained ratios and algorithms. The maximum values of the error of full scale in determining the change in parameters for the distance, pitch and roll angles did not exceed 4%, and for the yaw angle - 6.1%. Increasing the accuracy of parameter determination is possible due to the use of a camera with a higher resolution.