移动式起重机的吊装作业是施工现场常见且重要的施工活动，适当的吊装路径规划不仅能够提高吊装效率，还能提高吊装作业的安全水平。然而，面对动态、复杂的施工现场环境，项目经理或起重机操作人员通常主要依据个人经验制定吊装作业规划，难以满足现场实际操作的需求，时常导致吊装效率低下、引发安全事故。因此，如何实现高效、安全、动态的吊装路径规划是吊装过程需要重点考虑的问题。 近年来，BIM（Building Information Modeling）、虚拟施工(Virtual Construction, VC)等信息技术的兴起为吊装路径的智能规划提供了有力的支持。当前借助计算机在虚拟场景下进行吊装模拟一定程度上辅助了实际施工中的吊装路径规划，但多是考虑特定的或静态的施工作业环境，较少考虑动态实时变化的现场环境，同时相关算法的搜索效率还有待于提高。此外，大多数的模拟中许多参数仍要手动设置，尚未能充分指导实际吊装、满足智能规划的需求。 本文通过对移动式起重机吊装路径规划的重新定义，建立了数学模型，选取RRT（Rapidly-exploring Random Tree）为基础算法并分析其优劣，基于此提出了纵向“C空间预优化——RRT改进——路径后处理”和横向“效率角度、安全角度、动态角度”的算法整体改进框架，并提出了具体改进策略，形成了完整的吊装路径智能规划方法，以提高吊装的规划效率、动态环境应变能力和安全程度。同时，通过静态和动态测试，验证了该方法的可行性与有效性。进一步，结合该方法设计了移动式起重机吊装路径智能规划原型系统架构及主要功能模块，并进行了初步开发，初步实现了虚拟施工场景下吊装路径的智能规划。 本文提出的移动式起重机吊装路径智能规划方法，为吊装路径规划的理论研究与实践提供了参考价值。同时，这也将有助于提高工程实践中吊装操作的智能化程度，即以“模拟成果”支持“智能吊装”，从而促进施工过程的智能化，推动建筑行业的智能化发展。
The lifting of mobile cranes is a commonly-seen vital construction activity on construction sites. An appropriate lifting path plan is able to improve not only the efficiency of lifting but safety performance. Lifting path planning is conventionally conducted by operators or project managers based on their experience. However, due to the dynamics and complexity of construction sites, the traditional method often results in inefficient, inflexible and unsafe lifting operations. Therefore, how to achieve a fast, safe, dynamic lifting path planning is a problem that needs to be considered in the lifting process. In recent years, the development of information technologies such as Building Information Modeling (BIM) and Virtual Construction (VC) has provided powerful support for intelligent lifting path planning. So far, the lifting simulation in Virtual Construction with the aid of the computer has assisted the lifting path planning in practical construction to a certain degree, but mainly considers the specific or static construction environment, less considers the dynamic, real-time changing environment. Meanwhile, the efficiency of related algorithm and the quality of the planned path still need to be improved. In addition, many parameters in most simulations still need to be manually set, far from meeting the requirements for guiding actual lifting and intelligent planning. This paper establishes a mathematical model by redefining the lifting path planning of mobile cranes, selects RRT (Rapidly-exploring Random Tree) as the basic algorithm, analyzes its advantages and disadvantages and proposes a longitudinal “C-space pre-optimization—RRT improvement—Path post-processing” and horizontal “efficiency, safety, and dynamic” algorithm optimization framework, and put forward many specific improvement strategies, formed a complete intelligent planning method for lifting path, improving the planning efficiency, reaction ability to dynamic environmental and safety. At the same time, the feasibility and effectiveness of the method are verified by static and dynamic tests. Further, according to this method, a mobile crane-lifting path planning prototype system based on the “data layer-application layer-user layer” prototype system architecture and “input-computing-output” three functional modules are designed. In the virtual construction scenario, the intelligent planning of the lifting path was successfully implemented using this system. The intelligent planning method of mobile crane hoisting path proposed in this paper provides reference value for theoretical research and practice of lifting path planning. At the same time, it will also help to increase the lifting intelligence in engineering practice, that is, to support “smart lifting” with “simulation results”, thereby facilitating the intelligent construction process and promoting the intelligence of the construction industry.
