基于规则的巡检机器人自动越障专家系统研究

发布时间：2018-06-03 18:14

本文选题：专家系统 + 巡检机器人　；参考：《山东科技大学》2017年硕士论文

【摘要】：随着科学技术的发展,尤其是机器人技术的发展,输电线路的巡检工作逐渐由人工巡检方式向机器人巡检方式过渡,巡检机器人在输电线路巡检工作中的重要性逐渐得到认可。由于输电线路环境复杂,障碍多样,巡检机器人自动越障的问题日益凸显,主要表现在越障稳定性差、越障效率低等。因此,研究一种新型的提高巡检机器人自动越障稳定性和效率的方法具有十分重要的意义。本文在深入研究了产生式规则和专家系统理论的基础上,充分调研了巡检机器人控制系统的发展现状,探讨了课题研究的目标与可行性,提出了一种基于规则的专家系统的方法,将产生式规则和专家系统理论应用到巡检机器人自动越障控制过程中。本文结合巡检机器人结构、机器人自由度和传感器信号特点,对传感器信号采集电路、安装布局和机器人自动越障流程进行了深入的研究,分析了巡检机器人自动越障的功能要求,设计了自动越障专家系统的总体结构、功能模块和工作流程。分析了知识的概念,知识的表示方式和获取方式,采用直接获取和间接获取相结合的知识获取方式和产生式规则的知识表示方式建立了专家系统的知识库,并设计了知识库的维护流程。在此基础上探讨了推理机的推理方式、推理方向、冲突消除策略和Rete匹配算法,分析比较了盲目性搜索和启发性搜索两种搜索方式,采用确定性推理方式、正向推理方向、盲目性搜索方式和Rete改进算法,设计了推理机的推理流程。在建立了知识库和推理机的基础上,以Windows7为操作系统,采用Visual Studio 2010作为开发工具,以C#语言编写推理程序,以SQL Server 2008作为数据库,开发了基于规则的巡检机器人自动越障专家系统。它具有良好的人机界面,实现了自动越障、知识库管理、帮助等功能。在完成理论分析的基础上,运用巡检机器人自动越障对本专家系统进行了实验,验证了越障过程理论分析的正确性与可靠性,实验结果表明了将专家系统理论应用于巡检机器人自动越障过程控制的可行性。
[Abstract]:With the development of science and technology, especially the development of robot technology, the inspection of transmission lines is gradually transferred from manual inspection to robot inspection, and the importance of inspection robot in transmission line inspection is gradually recognized. Due to the complex environment and various obstacles of transmission lines, the problem of automatic obstacle crossing of inspection robot is increasingly prominent, which is mainly reflected in the poor stability and low efficiency of obstacle crossing. Therefore, it is of great significance to study a new method to improve the stability and efficiency of automatic obstacle crossing. On the basis of in-depth study of production rules and expert system theory, this paper makes a full investigation of the current situation of the control system of the inspection robot, and discusses the goal and feasibility of the research. In this paper, a rule-based expert system method is proposed. The production rule and expert system theory are applied to the automatic obstacle control of the inspection robot. Combined with the structure of the robot, the degree of freedom of the robot and the characteristics of the sensor signal, this paper makes a deep research on the sensor signal acquisition circuit, the installation layout and the robot automatic obstacle crossing flow. This paper analyzes the functional requirements of the automatic obstacle crossing of the inspection robot, and designs the overall structure, function module and work flow of the automatic obstacle crossing expert system. This paper analyzes the concept of knowledge, the representation and acquisition of knowledge, and establishes the knowledge base of expert system by the combination of direct acquisition and indirect acquisition and knowledge representation of production rules. The maintenance flow of knowledge base is designed. On this basis, the reasoning mode, direction, conflict resolution strategy and Rete matching algorithm of the inference engine are discussed. Two search methods, blind search and heuristic search, are analyzed and compared. The blind search method and the improved Rete algorithm are used to design the reasoning flow of the inference machine. Based on the establishment of knowledge base and inference engine, Windows7 is used as operating system, Visual Studio 2010 is used as development tool, reasoning program is written in C # language, and SQL Server 2008 is used as database. A rule-based automatic obstacle crossing expert system for inspection robot is developed. It has good man-machine interface, automatic obstacle surmounting, knowledge base management, help and other functions. On the basis of the theoretical analysis, the experiment of the expert system is carried out with the automatic obstacle surmounting of the inspection robot, and the correctness and reliability of the theoretical analysis of the obstacle surmounting process are verified. The experimental results show the feasibility of applying the expert system theory to the automatic obstacle control of the inspection robot.
【学位授予单位】：山东科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242

【参考文献】