基于机械臂的混凝土裂缝修复工艺知识库

构建及应用研究

马文瑾

摘 要

随着极端环境下工程数量逐年增加，工程结构因裂缝导致而发生的事故频率也不断提升，为了避免结构缺陷影响工程设施的功能与安全，保障工程设施的有效维护变得刻不容缓。然而，在恶劣的极端环境下，采用传统人工维护方案时，往往伴随着危险、重复劳动多、效率低下、受限于工作人员主观性等困难。无人设备的飞速发展使得极端环境下替代人工的修复方案成为了可能，经研究发现，实现无人修复的前提首先是解决机器根据现场情况进行自主决策的问题。因此，如何将裂缝修复领域中的知识系统化，研究信息技术如何进行裂缝修复方案的决策过程便具有了重要意义。

本文通过系统分析裂缝修复领域的专业知识，构建基于机械臂的混凝土裂缝修复工艺知识库，提出裂缝轨迹提取方案及裂缝决策推理机制，实现基于实际裂缝的修复方案自动决策和输出。在实验室对轨迹提取方案和裂缝修复决策推理验证，系统能计算出机械臂可识别的裂缝轨迹坐标，并推理输出裂缝修复方案。

通过上述研究，实现了无人情况下裂缝修复方案的自动输出，并提出了裂缝轨迹识别技术，为将来的裂缝自动修复提供了工作路径，对提高极端环境下工程维护信息化水平提供了借鉴意义。

关键词：裂缝修复 机械臂 轨迹提取 知识库 案例推理

Abstract

With the increase in the number of engineering construction in extreme environment year by year, the frequency of accidents caused by engineering structure cracks is also growing. In order to avoid the structural defects affecting the safety and function of engineering facilities, it is imperative to ensure the effective maintenance of engineering facilities. However, in the case of artificially repairing structural cracks in harsh extreme environments, it is often accompanied by difficulties such as danger, repeated labor, inefficiency, and subjectivity of the staff. The rapid development of unmanned equipment makes it possible to replace artificial repair solutions in extreme environments. Research shows that the prerequisite for unmanned repair is firstly to solve the problem of independent decision-making by machines based on site conditions. Therefore, how to systematize the knowledge in the field of crack repair and study how information technology can make the decision-making process of crack repair program is of great significance.

By systematically analyzing the professional knowledge in the field of fracture repair, this paper constructs the knowledge base of crack repair process based on the robotic arm, proposes the crack trajectory extraction scheme and the crack decision-making reasoning mechanism, and realizes the automatic decision-making and output of the actual fracture repair scheme.First of all, the knowledge of crack repair was extracted from the standards and regulations, journal papers and expert experience,and the artificial decision-making process of crack repair was used for reference to analyze the causes of crack formation and the factors influencing the crack repair decision.It is concluded that the morphological characteristics of cracks are important manifestations of crack causes.Secondly, the database of crack characteristics is established to provide data support for crack detection, and the extraction method of crack trajectory coordinates based on hand-eye calibration is proposed.Furthermore, the knowledge base of crack repair process was constructed, and the reasoning mechanism of crack repair based on cbr-rbr mixed model was established.Finally, a decision support system of crack repair was constructed to verify the trajectory extraction scheme and crack repair decision reasoning in the laboratory. The system was able to calculate the crack coordinate recognized by the robotic arm and deduce and output the crack repair scheme.

Through the above research, the automatic output of the crack repairing scheme in the unmanned situation is realized, and the crack trajectory identification technology is proposed, which provides a working path for the future automatic repair of cracks, and provides reference for improving the informationization level of engineering maintenance in extreme environments.

Keywords: Crack repair    Robotic arm    Trajectory extraction    Knowledge base

                   Case-based reasoning