智能小车的感知与决策控制(含CAD图)

智能小车的感知与决策控制(含CAD图)(任务书,开题报告,文献摘要,外文翻译,论文说明书30000字,CAD图4张)
摘 要
随着计算机科学和控制工程与技术的飞速发展，智能移动机器人技术逐步地在工业生产和日常居家出行的生活中扮演了重要的角色，包括室内服务机器人、自动驾驶技术、AGV，作为机器人学和人工智能领域的一个关键组成部分，移动机器人是最近几年来研究和发展的热门。移动机器人研究中，最重要的为同步定位与建图以及导航技术。本文将利用实验室现有的Dashgo D1两轮式差速小车为实验平台，通过机器人操作系统构建软件框架的，进行本课题的研究。
首先，搭建了小车的差分驱动模型、里程计运动模型以及激光雷达模型。在确定激光雷达模型时，考虑对比了基于TOF和三角法测距两种原理的激光雷达，选择了综合性能更佳的TOF激光雷达。
其次，提出了SLAM的概率模型，将问题简化为“估计”——“观测”——“校正”的研究历程。通过对两种具有代表性的开源建图算法——基于粒子滤波的Gmapping和基于图优化的Cartographer算法进行原理分析和实验对比，比较两者的特点和优劣性。
然后针对小车的导航问题，以全局路径规划和局部路径规划作为小车的路径规划算法两方面的问题进行介绍和分析。在全局路径规划中讨论了两种经典的全局路径规划算法Dijkstra和A*，通过一个仿真程序的执行速度和空间占用率对两者进行分析比较。针对局部路径规划，本文介绍了一种通过对一个全局路径规划器所产生的初始全局轨迹进行后续修改来优化小车的期望运行轨迹的TEB算法。最后利用teb_local_planner_tutorials中的插件进行仿真实验比较EB与TEB两种算法的区别。

最后在实际环境中通过Cartographer完成了实验场所真实环境的建图，基于已知地图，利用A*全局路径规划和TEB局部路径规划在实际环境中进行导航，经实验验证，该方案有效且具备一定稳定性。
关键词：移动机器人；SLAM；路径规划；ROS

Abstract
With the rapid development of computer science and control engineering, intelligent mobile robot technology gradually plays an important role in industrial production and daily travel, including indoor service robot autonomous AGV technology, as a key component of the robotics and artificial intelligence, mobile robot is popular in recent years’ research and development of mobile robot study, the most important is simultaneous localization and map building and navigation technology. This article will use the laboratory existing DashgoD1 two-wheel differential vehicle as the experimental platform, and the software framework is constructed through the robot operating system to carry out the research of this topic.
First, the differential drive model, odometer motion model and lidar model of the car were built. In the determination of the lidar model, the laser radar based on the two principles of TOF and triangulation is considered, and the TOF laser radar with better comprehensive performance is selected.

[资料来源：www.doc163.com]

Secondly, the probability model of SLAM is proposed, which simplifies the problem into the process of “estimation” – “observation” – “correction”. By comparing the principle analysis and experimental comparison of two representative open source mapping algorithms, Gmapping based on particle filtering and Cartographer basedon graph optimization, the characteristics and pros and cons of the two are compared.
Then the global path planning and local path planning are introduced and analyzed for the navigation problem. Two classical global path planning algorithms Dijkstra and A* are discussed in the global path planning. For local path planning, this paper introduces a TEB algorithm to optimize the desired trajectory of the car by modifying the initial global trajectory generated by a global path planner. Finally, the plug-in of teb_local_planner_tutorials is used to do the simulation experiment and compare the difference between EB and TEB algorithms.

[资料来源：www.doc163.com]

Finally, in the actual environment, the Cartographer was used to complete the environmental construction of the experimental site. Based on the known maps, the A* global path planning and the TEB local path planning were used for navigation. The experimental results show that the scheme is effective and has certain stability.
Key Words：mobile robot; SLAM; path planning; ROS
 

[资料来源：http://www.doc163.com]

目录
第1章绪论    1
1.1 研究背景及意义    1
1.2 国内外研究现状    1
1.2.1 国外研究现状    1
1.2.2 国内研究现状    3
1.3 研究内容和目标    3
第2章移动机器人系统建模    5
2.1 引言    5
2.2小车运动学模型    5
2.2.1 差分驱动模型    5
2.2.2 里程计模型    7
2.3 激光雷达模型    11
2.3.1 概述    13
2.3.2 数据分析    14
2.4 本章小结    14
第3章实时定位与地图构建    16
3.1 引言    16
3.2 基于RBPF（Rao-Blackwellized Particle Filter）的方法    17 [版权所有：http://DOC163.com]
3.2.1 粒子滤波PF    17
3.2.2 FastSLAM    21
3.2.3 Gmapping建图    24
3.3  基于图优化的方法    25
3.3.1 GraphSLAM 原理    25
3.3.2 非线性最小二乘    26
3.3.3 非线性最小二乘在GraphSLAM中的应用    27
3.3.4 Cartographer建图    30
3.4 实验验证    30
3.5 本章小结    32
第4章路径规划    33
4.1 引言    33
4.2 全局路径规划    33
4.2.1 Dijkstra算法介绍    33
4.2.2 A*算法介绍    35
4.2.3 估价函数    37
4.2.4 对比仿真实验    38
4.3 局部路径规划    40
4.3.1 EB算法    40
4.3.2 TEB算法    42
4.3.3 g2o图优化    49
4.3.4 仿真结果    50
4.4 本章小结    51

[资料来源：http://www.doc163.com]

第5章实验平台及实验结果    53
5.1 硬件平台    53
5.1.1 实验环境    53
5.1.2 硬件平台    53
5.2 软件平台    54
5.2.1 ROS系统简介    55
5.2.2 ROS基本术语    55
5.2.3 ROS特性    56
5.2.4 其他工具    57
5.3 真实场景的SLAM与导航    59
5.3.1 同步定位与建图（SLAM）    59
5.3.2 基于已知地图的导航    60
5.4 本章小结    62
总结与展望    63
参考文献    65
致谢    67