摘要
随着人工智能的快速发展,救援机器人要能快速准确地实施救援,最重要的就是要能看见,机器视觉的提出解决了这个问题。机器视觉是一项综合技术,主要用计算机来模拟人的视觉功能,从客观事物的图像中提取信息,进行处理并加以理解,最终用于实际检测、测量和控制,这也是救援机器人寻找被困者、处理地形环境以及将图像传送至操作者的根本。机器视觉的加入使救援机器人更加灵活,同时也为救援机器人智能化提供了必要条件。该文将工业相机作为视觉传感器,完成目标识别与定位任务。利用C语言编程在keil5软件平台开发,植入嵌入式实时操作系统FreeRTOS的STM32F427IGHx芯片中,以完成对机器人的实时控制。通过实验验证了视觉识别及定位方法的可行性。
with the rapid development of artificial intelligence,the most important thing for rescue robots to implement rescue quickly and accurately is to see.The proposal of machine vision undoubtedly solves this problem.Machine vision is a comprehensive technology,which mainly uses computer to simulate human visual function,extract information from the image of objective things,process and understand it,and finally use it for actual detection,measurement and control.This is also the fundamental of rescue robot to find the trapped,deal with the terrain environment and transmit the image to the operator.The addition of machine vision not only makes the rescue robot more flexible,but also provides the necessary conditions for the intellectualization of the rescue robot.In this paper,the industrial camera is used as the visual sensor to complete the task of target recognition and positioning.Using C language programming,it is developed on keil5 software platform and implanted into STM32F427IGHx chip of embedded real-time operating system FreeRTOS to complete the real-time control of the robot.The feasibility of visual recognition and positioning method is verified by experiments.
作者
全瑞琴
刘岩
徐媛媛
陈振兵
QUAN Ruiqin;LIU Yan;XU Yuanyuan;CHEN Zhenbing(School of Control Engineering, Xinjiang Institute of Technology, Xinjiang Urumqi 830000,China)
出处
《工业仪表与自动化装置》
2022年第3期64-68,74,共6页
Industrial Instrumentation & Automation
基金
2020年度新疆工程学院科研基金项目(2020xgy242302)
国家级大学生创新创业训练计划项目(202110994011)
新疆工程学院人工智能及智慧矿山工程技术中心。
关键词
救援机器人
图像处理
视觉识别
目标定位
rescue robot
image processing
visual recognition
target location
