摘要
设计了一种面向实验教学的单目视觉与惯性测量单元(IMU)融合的开源同时定位与建图(SLAM)仪器。完成了摄像头和IMU在实时机器人操作系统(ROS)下的数据传输、联合标定、视觉里程计前端、IMU的预积分、视觉与IMU的紧耦合、回环检测和后端优化软件设计,计算得到了仪器实时运动轨迹。该实验教学仪器可以应用于移动机器人的SLAM任务。实验仪器经大量测试,能够为学生提供算法开源的视觉SLAM实验平台,有效提高人工智能、机器人工程的本科人才培养水平,增强学生知识综合运用能力和实践动手能力,获得了良好的实践教学效果。
An open source simultaneous localization and mapping(SLAM)instrument based on monocular vision and inertial measurement unit(IMU)is designed for experimental teaching.The data transmission,joint calibration,visual odometer front-end,IMU pre integration,tight coupling of visual information and attitude information,loop detection and back-end optimization software design of camera and IMU in real-time robot operating system(ROS)are completed,and the real-time trajectory of the instrument is calculated.After extensive testing,the experimental instrument can provide students with an open-source visual SLAM experimental platform for students,effectively improve the training level of undergraduate talents in artificial intelligence and robotics,enhance students'comprehensive application of knowledge and practical skills,and obtain good practical teaching effects.
作者
刘涛
杜宏旺
LIU Tao;DU Hongwang(College of Intelligent Systems Science and Engineering,Harbin Engineering University,Harbin 150001,China)
出处
《实验技术与管理》
CAS
北大核心
2021年第7期111-115,119,共6页
Experimental Technology and Management
基金
国家自然科学基金项目(51409059)
哈尔滨工程大学2020年校级本科教学改革研究项目(JG2020B15)。
关键词
同时定位与建图(SLAM)
惯性测量单元(IMU)
信息融合
实验仪器
simultaneous localization and mapping(SLAM)
inertial measurement unit(IMU)
information integration
experimental instrument
