Font Size: a A A

Target Following System Based On Embedded Mobile Vision

Posted on:2019-11-10Degree:MasterType:Thesis
Country:ChinaCandidate:X YangFull Text:PDF
GTID:2428330542972963Subject:Detection Technology and Automation
Abstract/Summary:PDF Full Text Request
Moving target tracking refers to the system and the moving target in the absence of a physical connection,the system follows a specific mechanism to move with the target,and maintain the appropriate distance,to complete follow,loading and video recording and other specific functions.With the development of machine vision technology,the way of visual detection of targets gradually matures,In this paper,a target tracking system based on embedded mobile vision is proposed,which adopts AprilTags as the system follow-up target.The machine vision sensor is used to identify and locate the moving target.The embedded system is used to develop a moving target tracking system with dual-mode control system.The system hardware part is composed of main control module,visual sensor module,remote control auxiliary unit,communication module,motor drive module and power module.The main control module adopts embedded processor STM32F4 and Atmega328 dual mode to cooperate with each other to accomplish target identification and follow-up control.Design dual-mode control,autonomous follow mode and manual remote control mode.In autonomous follow mode,by analyzing the visual information collected by the machine vision sensor module,the recognition and positioning of AprilTags mark is realized,the target position information is obtained,and the drive system autonomously follows the target and completes the adaptive tracking.In manual remote control mode,the 2.4G RF module n RF24L01 can be used to send commands to control the movement of the robot to realize the control of the robot in special situations such as unrecognized targets.Software design using visual sensor onboard Micro Python interpreter,Used Python language achieve the target recognition and positioning algorithm,C language writed the system movement section,Adopt differential follow-up algorithm to drive the system to follow the target movement,and uses the processing software based on JAVA language to write Android mobile application,send the operation mode selection command to the system through the WIFI module.Design a mobile target tracking system prototype,completed the test of system functions.The test results show,In autonomy follow mode,unobstructed,light conditions,The recognition rate of the AprilTags tag in the visual range is high,the farthest recognition distance 10 M,identification of deviation ± 5cm,and can follow the distance control within 1m to 2m,completed the system autonomy to follow a specific goal,also can be followed by manual remote control,wireless communication is stable,achieve dual-mode control,satisfy the system design requirements.
Keywords/Search Tags:Machine vision, AprilTags, Target recognition, Autonomous follow
PDF Full Text Request
Related items