CASIA OpenIR  > 毕业生  > 硕士学位论文
Alternative TitleTracked mobile robot control and unstructured road navigation
Thesis Advisor赵晓光
Degree Grantor中国科学院大学
Place of Conferral中国科学院自动化研究所
Degree Discipline控制工程
Keyword履带式移动机器人 运动控制系统 非结构化道路检测 路径跟踪 视觉导航 Tracked Mobile Robot Motion Control System Unstructured Road Detection Path Tracking Visual Navigation
Abstract履带式移动机器人,因其特殊的履带式结构,具有高度的环境适应能力,适合在恶劣环境或野外作业,尤其在越野、爬坡等方面,相比其他结构的移动机器人具有明显优势,广泛应用于国防、工业、医疗救助等方面。随着人工智能技术的迅速发展,机器人的智能化程度受到越来越多的关注,更多的智能技术应用到机器人的路径规划、轨迹跟踪、视觉导航等任务中,因此,履带式移动机器人智能化技术的研究具有重大的理论和工程应用价值。 本文以一款自主研发的履带式移动机器人为背景,完成了机器人控制系统的搭建、非结构化道路检测及路径跟踪算法的研究。本文的工作及成果主要分为以下几个部分: 首先,设计了基于ARM和FPGA处理器的总体控制方案,搭建了硬件平台,并完成了底层驱动、控制器、传感器等部件的选型和实现;基于该平台完成了主控制器和运动控制器的软件设计;分析了履带式移动机器人运动控制原理,设计了速度闭环系统,对转弯、直行等基本的运动控制做了分析和算法设计。 然后,对非结构道路检测算法进行了研究,利用Gabor滤波器提取道路纹理特征实现非结构化道路消失点的检测,并利用消失点和方向一致性、颜色特征划分道路区域;通过对图像进行降采样及对局部软投票方法实现优化,在保证道路检测精度的基础上降低了时间复杂度,并用实验验证了算法的有效性和实时性。 最后,研究了路径跟踪算法,提出了利用机器人当前位置与期望位置的距离偏差和航向角偏差直接控制两轮速度的模糊控制算法,设计了模糊控制器实现履带式移动机器人的路径跟踪;并设计了履带式移动机器人视觉导航系统,对本文的非结构化道路检测和路径跟踪算法做了验证和分析。
Other AbstractSince the special mechanical structure of the tracked mobile robot, it has a large contact surface with the earth. Thus, it has a superior ability of climbing and off-road and can work in harsh environments or in the field, which is difficult to achieve for other mobile robots. Thanks to its advantages, it is be widely used in the military, industry, medical assistance and so on. With the fast development of artificial intelligence, the intelligent degree of mobile robot has received more and more attention and many intelligent technologies has been introduce in the field of route planning, trajectory tracking, visual navigation. Therefore, conducting research on the intelligent tracked mobile robot has great theoretical and practical signification. We have established a control system for a self-developed tracked mobile robot and developed the unstructured road detection and path tracking algorithms. These works will be introduced in this paper in three parts: Firstly, we have designed and built a platform based on ARM and FPGA. Appropriate bottom-driven, controller and sensors have been selected and the software system, including the primary controller and motion controller, had been designed. We have also analyzed the motion control principle of tracked mobile robot and developed the speed closed-loop system. What’s more, the algorithms of basic motion, such as turning and straight walking, have also been accomplished. Secondly, we have researched the algorithm of unstructured road detection. Texture characteristics were adopted to detect the vanishing point of unstructured road, which is extracted by Gabor filters. Then, we divided the road based on the OCR and color feature. To ensure the detection precision and reduce the time complexity, we employed the strategy of down-sampling for image and optimized the method of local soft voting. Experimental results have shown the feasibility and effectiveness of our algorithm. At last, we have studied the algorithm of path tracking. A novel fuzzy control algorithm is proposed, which is direct control the two-wheeler speed by the distance and course deviation between the current position and desired position. We have implemented the fuzzy controller for path tracking. Moreover, we have also designed a visual navigation system for our tracked mobile robot and testified our algorithm of unstructured road detection and path tracking.
Other Identifier2012e8014661087
Document Type学位论文
Recommended Citation
GB/T 7714
王静. 履带式移动机器人控制与非结构化道路导航[D]. 中国科学院自动化研究所. 中国科学院大学,2015.
Files in This Item:
File Name/Size DocType Version Access License
CASIA_2012e801466108(3460KB) 暂不开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[王静]'s Articles
Baidu academic
Similar articles in Baidu academic
[王静]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[王静]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.