EXPERIMENTAL SYSTEM AND SOFTWARE FOR INVESTIGATION OF IMAGE CAPTURING AND FOLLOWING OF MOVING OBJECTS IN TWO DIMENSIONS

Abstract

Establishing optimal parameters of the photo capture and tracking system for moving objects is promising; creating an experimental system and software for studying the UAV control system using an on-board computer and flight controllerAn experimental system and software have been developed to generate and study commands for the motors of unmanned aerial vehicle (UAV) and its automatic rotation during following of a moving object using image capturing in order to optimize process of UAV controlling. A quadcopter with a Q450 frame has been used, along with D2212-920kv engines and 10 × 1.5cm propellers. An open source Pixhawk PX-4 autopilot and a Raspberry-3 on-board computer with a Pi V2 camera with a resolution of 1080 × 720 pixels have been installed. The quadcopter was suspended so that the propellers were in a horizontal position. The experimental setup includes a two-channel oscilloscope ISDS205A for measuring the width of the motors control pulses and a tachometer UT-372 for measuring the speed of rotation of the motors. We have compiled a program that calculates the coordinates of the center of the image X of the object in pixels and the corresponding pulse durations to control the rotation of the quadcopter motors using the fourth channel of the autopilot receiver. The flight system uses Dronekit, datetime, errno, platform libraries, and the video conversion program uses sys, OpenCV, Picamera libraries. Theoretical developments and practical recommendations made on the basis of the work performed have both independent theoretical value for prospective research in the field of photocapture and photo-tracking of small and distant moving objects, and have practical value in connection with ensuring a real high indicator and measurement of the parameters of the movement of unmanned aerial vehicles.