基于ROS的室内移动服务机器人定位与导航系统的研究与开发
发布时间：2019-04-03 07:20
【摘要】：近年来,由于信息化、工业化的不断推进,机器人产业在高新技术产业中不断发展,正处于一个蓬勃兴起的阶段,在中国制造业进行转型升级的浪潮中,机器人技术的发展尤其重要。2015年年末,北京国家会议中心举办了世界机器人大会,中国在战略上也将机器人与智能制造纳入了国家科技创新的优先重点领域。机器人SLAM技术是机器人技术中核心技术之一,SLAM是即时定位与地图构建技术,其中建图定位与路径规划是SLAM算法主要的攻克方向。SLAM相关算法需要一个完整的通信系统将传感器数据与算法结合。ROS又称机器人操作系统,它由斯坦福大学与Willow Garage公司共同研发,ROS提供了一个标准的操作系统环境,可以实现硬件与PC中进程、PC中各进程间通信的功能,通过ROS可以将算法计算结果通过节点传递给底层控制层进行运动控制。围绕上述几个方面,本文围绕机器人定位与路径规划两个问题,在现有研究的基础上,做出以下几点工作:(1)研究机器人领域国内外研究发展趋势通过当前相关科研进展来阐述本课题的意义,从当前学术研究方向引出相关技术。(2)运动学模型及测量模型的研究对实验平台的速度模型与里程计模型进行研究并建模。同时分析机器人实际运行过程中传感器可能受到的环境噪声,对观测模型进行研究,用于削减环境噪声带来的影响。(3)机器人定位算法本文所涉及的定位算法均是基于概率的算法,首先先根据机器人建立运动学模型及观测模型,然后将模型融入于滤波算法中,实现相关位置估计算法,通过MATLAB来验证算法有效性,并将算法融合到ROS中。(4)机器人导航算法导航算法分为全局与局部路径规划两个方向。针对这两个方向,本课题各提出两种算法,并在MATLAB中验证算法有效性,进而对路径进行优化处理,最终在ROS中实现对实验平台的控制。(5)整体实验平台搭建本文分别对定位与导航模块进行MATLAB仿真并进行试验平台搭建,本课题使用双轮车搭载Intel处理器、激光雷达等传感器,通过STM32嵌入式开发板完成对小车的控制,ROS系统及算法在独立小型低功耗PC上进行算法运算与数据交互,最终完成对实验室环境的建图定位与自主导航的任务。
[Abstract]:In recent years, due to the continuous advancement of informatization and industrialization, the robot industry is constantly developing in the high-tech industry and is in a flourishing stage. In the tide of transformation and upgrading of China's manufacturing industry, The development of robotics technology is particularly important. In late 2015, the Beijing National Convention Center held the World Robot Conference, and China has strategically included robotics and intelligent manufacturing in the priority areas of national scientific and technological innovation. Robot SLAM technology is one of the core technologies of robot technology, and SLAM is real-time positioning and map building technology. Mapping location and path planning are the main attack direction of SLAM algorithm. Lam correlation algorithm needs a complete communication system to combine sensor data and algorithm. Ros is also called robot operating system. Developed by Stanford University and Willow Garage, ROS provides a standard operating system environment that enables hardware to communicate with processes in PC and between processes in PC. Through ROS, the results of the algorithm can be passed to the bottom control layer for motion control. Around the above-mentioned aspects, this paper focuses on the robot positioning and path planning, on the basis of the existing research, The following work is done: (1) the research and development trend of robot field at home and abroad expounds the significance of this topic through the current related scientific research progress. The related technologies are derived from the current academic research direction. (2) the velocity model and odometer model of the experimental platform are studied and modeled by the research of kinematics model and measurement model. At the same time, the possible environmental noise of the sensor during the actual operation of the robot is analyzed, and the observation model is studied. It is used to reduce the impact of environmental noise. (3) the localization algorithms involved in this paper are all probability-based algorithms. Firstly, the kinematics model and observation model are established according to the robot. Then the model is incorporated into the filtering algorithm, and the correlation position estimation algorithm is implemented. The effectiveness of the algorithm is verified by MATLAB, and the algorithm is fused into ROS. (4) the robot navigation algorithm is divided into two directions: global and local path planning. In view of these two directions, two kinds of algorithms are proposed in this paper, and the validity of the algorithm is verified in MATLAB, and then the path is optimized. Finally, the control of the experiment platform is realized in ROS. (5) the whole experiment platform is built. In this paper, the positioning and navigation module is simulated by MATLAB and the test platform is built. In this paper, the dual-wheeled vehicle is used to carry on the Intel processor. Lidar and other sensors, through the STM32 embedded development board to complete the control of the car, ROS system and algorithm on the independent small low-power PC algorithm operation and data interaction, Finally, the task of mapping location and autonomous navigation of laboratory environment is completed.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242
本文编号：2453020
[Abstract]:In recent years, due to the continuous advancement of informatization and industrialization, the robot industry is constantly developing in the high-tech industry and is in a flourishing stage. In the tide of transformation and upgrading of China's manufacturing industry, The development of robotics technology is particularly important. In late 2015, the Beijing National Convention Center held the World Robot Conference, and China has strategically included robotics and intelligent manufacturing in the priority areas of national scientific and technological innovation. Robot SLAM technology is one of the core technologies of robot technology, and SLAM is real-time positioning and map building technology. Mapping location and path planning are the main attack direction of SLAM algorithm. Lam correlation algorithm needs a complete communication system to combine sensor data and algorithm. Ros is also called robot operating system. Developed by Stanford University and Willow Garage, ROS provides a standard operating system environment that enables hardware to communicate with processes in PC and between processes in PC. Through ROS, the results of the algorithm can be passed to the bottom control layer for motion control. Around the above-mentioned aspects, this paper focuses on the robot positioning and path planning, on the basis of the existing research, The following work is done: (1) the research and development trend of robot field at home and abroad expounds the significance of this topic through the current related scientific research progress. The related technologies are derived from the current academic research direction. (2) the velocity model and odometer model of the experimental platform are studied and modeled by the research of kinematics model and measurement model. At the same time, the possible environmental noise of the sensor during the actual operation of the robot is analyzed, and the observation model is studied. It is used to reduce the impact of environmental noise. (3) the localization algorithms involved in this paper are all probability-based algorithms. Firstly, the kinematics model and observation model are established according to the robot. Then the model is incorporated into the filtering algorithm, and the correlation position estimation algorithm is implemented. The effectiveness of the algorithm is verified by MATLAB, and the algorithm is fused into ROS. (4) the robot navigation algorithm is divided into two directions: global and local path planning. In view of these two directions, two kinds of algorithms are proposed in this paper, and the validity of the algorithm is verified in MATLAB, and then the path is optimized. Finally, the control of the experiment platform is realized in ROS. (5) the whole experiment platform is built. In this paper, the positioning and navigation module is simulated by MATLAB and the test platform is built. In this paper, the dual-wheeled vehicle is used to carry on the Intel processor. Lidar and other sensors, through the STM32 embedded development board to complete the control of the car, ROS system and algorithm on the independent small low-power PC algorithm operation and data interaction, Finally, the task of mapping location and autonomous navigation of laboratory environment is completed.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242
【引证文献】
相关硕士学位论文 前2条
1 高日;基于多传感器信息融合的机器人定位技术研究[D];北京建筑大学;2018年
2 卢双;多功能家庭床椅服务机器人研究[D];江南大学;2018年
，本文编号：2453020
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2453020.html
