Research And Design Of UAV Tracking System Based On DSP And OpenMV Vision Module

In recent years,with the rapid development of quadrotor UAV and machine vision technology,the visual image acquisition and analysis system,which integrates the functions of environment perception,planning decision and automatic flight,has gradually become the research focus and hot spot in this field.Based on this hot spot,this paper chooses the topic of combining UAV based on DSP with machine vision module.This topic can make UAV complete various flight actions and track specific targets.It has a very simple structure,low cost,can achieve extremely high tracking accuracy,and can be extended in the future.It can be used for indoor low-altitude flight and outdoor open environment flight,and has high practical value.This topic mainly uses dual-core DSP as the control core,uses Open MV machine vision module to collect image signals,uses Python language to write programs to process the collected image signals,and designs a series of algorithms to identify paths,plan paths,and identify and track targets,so as to effectively control the UAV to fly along the designated trajectory,and can identify and process the collected image information,distinguish some features of objects and track moving targets in real time during flight.This paper mainly describes the working principle of UAV and the design of hardware architecture,matching flight control program and visual module program.The UAV system is mainly composed of DSP chip,various sensors and embedded vision module.After collecting image data through dual cameras,the embedded image processing library is used to fit the flight attitude of the target with various target detection algorithms based on neural network algorithm,which is sent to the control chip through serial port.The control chip controls the brushless motor under the control of the flight control program to realize various flight actions.The flight control program is optimized in this paper,and various state tests are carried out in the laboratory.Finally,the target task of UAV autonomous recognition and tracking preset target object is completed.