Intelligent building construction path planning method and system based on AI

Abstract

The invention relates to the technical field of intelligent building construction path planning methods, in particular to an AI-based intelligent building construction path planning method and system, and the method comprises the following steps: mapping site construction data into a virtual scene graph which comprises construction machinery information and environment information; on the basis of an AI algorithm and the virtual scene graph, a reasonable construction sequence and path are automatically planned for all machines in construction, and a construction scheme is generated; simulating the construction scheme to verify feasibility and risk points of the construction scheme; the construction scheme is optimized according to the risk point, and an optimal construction scheme is generated, so that the dynamic change of the construction site can be captured in real time, and intelligent path planning and resource allocation can be carried out based on the information; the method provided by the invention has self-learning and optimization capabilities, can continuously accumulate experience from practice, and improves the accuracy and efficiency of planning.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent building construction path planning methods, and in particular to an AI-based intelligent building construction path planning method and system thereof. Background Art

[0002] In recent years, with the acceleration of urbanization and the continuous advancement of construction technology, the scale and complexity of construction projects have continued to increase. Traditional construction management methods have shown many shortcomings in the face of these new challenges. Especially in terms of construction path planning, existing technologies often find it difficult to effectively handle various dynamic factors in large and complex projects, resulting in inefficient resource utilization and frequent delays in construction schedules.

[0003] At present, the construction path planning method commonly used in the industry mainly relies on manual experience and simple computer-aided tools. This method can still cope with small projects, but it seems to be powerless in large and complex projects. First, manual planning is difficult to fully consider all influencing factors, and it is easy to make wrong decisions. Secondly, although simple computer tools can provide certain data support, they lack intelligent analysis and optimization capabilities and cannot cope with real-time changes on the construction site.

[0004] More specifically, the existing technology has the following major problems: First, there is a lack of real-time and accurate grasp of the dynamic situation of the construction site. Traditional methods usually rely on regular manual inspections and reports, which is not only time-consuming and labor-intensive, but also prone to missing important information. Secondly, the scheduling of mechanical equipment is often too rigid and difficult to flexibly adjust according to the actual situation on site, resulting in low equipment utilization and even mutual interference. Thirdly, the existing planning methods are difficult to effectively predict and respond to potential risks, and often respond passively after problems arise, which not only affects the construction progress, but also may bring safety hazards. Finally, the existing technology lacks the ability to self-learn and optimize, and it is difficult to draw experience from past projects and improve the accuracy and efficiency of future planning.

[0005] The existence of these problems not only reduces the efficiency of construction and increases costs, but may also affect the quality and safety of the project. Therefore, the development of a path planning method that can intelligently and dynamically handle complex construction environments has become an urgent need in the construction industry. Summary of the invention

[0006] The present invention is proposed to solve the above problems. It aims to solve the key technical problems existing in the traditional construction path planning method, such as insufficient real-time performance, low intelligence, and weak risk prediction ability. By introducing artificial intelligence technology, combining virtual scene modeling and real-time data collection, the present invention provides a new intelligent building construction path planning method and system.

[0007] The present invention proposes an AI-based intelligent building construction path planning method, which includes the following steps:

[0008] Mapping the on-site construction data into a virtual scene graph, wherein the virtual scene graph includes construction machinery information and environment information;

[0009] Based on the AI ​​algorithm and the virtual scene graph, a reasonable construction sequence and path are automatically planned for each machine in construction, and a construction plan is generated;

[0010] Simulating the construction plan to verify the feasibility and risk points of the construction plan; and

[0011] The construction plan is optimized according to the risk points and an optimal construction plan is generated.

[0012] Preferably, the step of mapping the on-site construction data into a virtual scene graph comprises:

[0013] Digitally map the current venue as a virtual map;

[0014] Acquiring construction data and converting it into components and attributes of the virtual map;

[0015] Constructing a scene model including environmental information and construction machine information, wherein the construction machine is provided with sensors to provide construction data in real time;

[0016] associating scene modeled components with the construction data; and

[0017] The components, the attributes and the construction data are stored in a scene to form the virtual scene graph.

[0018] Preferably, the step of constructing the scene model includes:

[0019] Divide the current construction map into a number of construction areas, each of which is a component;

[0020] Setting attribute information for each of the components, including type, attribute value, area range, and relationship between components;

[0021] Setting the construction machinery in the current site as a component and setting the construction data as an attribute value;

[0022] Using wireless technology to bind construction components and construction machine components to the relevant machines through unique device identifiers, and associate them with sensors on the machine controllers to sense the current construction environment in real time; and

[0023] By continuously updating the on-site construction data, the virtual scene can present the current construction environment status in real time in a dynamic form.

[0024] Preferably, the method further comprises the following steps:

[0025] Setting the mutual influence relationship between the components, and setting the influence area of ​​the components according to the reaction generated when the components overlap; and

[0026] The construction components are subjected to collision detection according to the operation range, and it is determined whether the components affect each other according to the reaction during the collision.

[0027] Preferably, the step of automatically planning a reasonable construction sequence and path for the construction machinery includes:

[0028] Randomly select a construction area in the current scene, obtain all the construction machinery that have entered the construction site in the area, and obtain the attribute values ​​of the machinery parts;

[0029] Calculating the priority of each construction machine according to the attribute value;

[0030] Plan construction routes within the current construction area and place high-priority machinery at priority positions; and

[0031] Plan the construction sequence of each machine in the construction area, give priority to planning the construction machines with high priority, and plan each construction machine to the construction path.

[0032] Preferably, before randomly selecting the construction area, the following steps are also included:

[0033] Get the project completion date;

[0034] In combination with the completion date, planning the entry and exit sequence of construction machinery in the current virtual scene diagram to form an overall construction plan; and

[0035] The construction phases and execution machinery for each construction area are determined, wherein the overall construction plan includes the time when each area will be constructed.

[0036] Preferably, the step of calculating the priority of each construction machine according to the attribute value includes:

[0037] Set the calculation formula for construction machinery priority, where construction machinery priority = A*urgency+B*importance+C*distance;

[0038] Define the parameters of the construction machinery priority calculation formula, where A is the construction progress urgency coefficient, setting the priority between 0 and 1, the closer to 1, the more urgent, B is the construction result importance coefficient, and C is the distance of the current machine from the planned construction location; and

[0039] The data is trained through a machine learning algorithm to iteratively optimize the formula parameters.

[0040] Preferably, the step of automatically optimizing the construction plan by the AI ​​algorithm includes:

[0041] Set priorities based on the contribution of construction machinery to the project results, with more important construction machinery having higher priorities;

[0042] Using the priorities to plan paths and arrange construction sequences for machines;

[0043] Calculate the priority of the current machinery in each construction area; and

[0044] Plan construction machinery according to regional priorities, giving priority to construction machinery with higher priorities to ensure the progress of the entire project.

[0045] Preferably, the step of simulating the construction plan to verify the feasibility and risk points of the construction plan includes:

[0046] Verify through simulation whether the mechanical planning sequence of the construction plan in the virtual scene graph is executable. If not, re-plan it;

[0047] Verify through simulation whether the mechanical path in the virtual scene graph is smooth. If not, display the unsmooth parts and replan them;

[0048] Verify through simulation whether the machines in the virtual scene graph affect each other. If so, display the components that affect each other and re-plan them; and

[0049] Use neural network learning methods to train and optimize construction path planning algorithms.

[0050] An AI-based intelligent building construction path planning system for implementing the method comprises:

[0051] A data mapping module is used to map the on-site construction data into a virtual scene graph, which contains construction machinery information and environmental information;

[0052] A planning module, for automatically planning a reasonable construction sequence and path for each machine in construction and generating a construction plan based on the AI ​​algorithm and the virtual scene graph; and

[0053] A verification module, used to simulate the construction plan to verify the feasibility and risk points of the construction plan, optimize the construction plan according to the risk points and generate an optimal construction plan;

[0054] The data mapping module, the planning module and the verification module are interconnected via a data interface to form a closed-loop intelligent planning system.

[0055] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0056] The core of this method lies in its powerful data integration and intelligent decision-making capabilities. It can not only capture the dynamic changes of the construction site in real time, but also perform intelligent path planning and resource allocation based on this information. More importantly, the method of the present invention has the ability of self-learning and optimization, can continuously accumulate experience from practice, and improve the accuracy and efficiency of planning.

[0057] From the perspective of algorithm mechanism, the present invention cleverly solves multiple seemingly contradictory technical problems. For example, the contradiction between global optimization and local flexibility is effectively solved through a hierarchical planning strategy. The system can not only ensure the consistency of the overall construction plan, but also make flexible adjustments according to local conditions. At the same time, the present invention also achieves an organic combination of efficiency improvement and risk control. By introducing multi-dimensional evaluation indicators and real-time simulation verification, the system can effectively prevent potential risks while pursuing high efficiency.

[0058] In addition, the method of the present invention achieves complementary and synergistic technical effects in many aspects. For example, the combination of real-time data acquisition and intelligent planning algorithms not only improves the accuracy of planning, but also greatly enhances the system's ability to respond to emergencies. The dynamic mapping of virtual scene modeling and actual construction processes provides new possibilities for safety management and quality control.

[0059] In general, the intelligent building construction path planning method and system of the present invention achieves comprehensive benefits that are difficult to achieve with traditional methods by comprehensively using advanced technologies such as artificial intelligence, big data analysis, and virtual simulation. It can not only significantly improve construction efficiency and reduce costs, but also effectively improve safety and reduce environmental impact. The application of this method is expected to bring revolutionary changes to the construction industry and promote the entire industry to develop in a smarter, more efficient, and more sustainable direction. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is the overall logic block diagram of the system of the present invention.

[0061] Figure 2 It is a logic block diagram of the data mapping module of the present invention.

[0062] Figure 3 It is a logic block diagram of the planning module of the present invention.

[0063] Figure 4 It is a logic block diagram of the verification module of the present invention. DETAILED DESCRIPTION

[0064] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects thereof are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0065] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0066] See also Figure 1-4 The purpose of the present invention is to provide an AI-based intelligent building construction path planning method and system to solve the problems of low path planning efficiency and low resource utilization in traditional building construction.

[0067] To achieve the above-mentioned objectives, the present invention provides the following technical solutions: First, an AI-based intelligent building construction path planning method includes the following steps: mapping on-site construction data into a virtual scene graph, wherein the virtual scene graph contains construction machinery information and environmental information; based on the AI ​​algorithm and the virtual scene graph, automatically planning a reasonable construction sequence and path for each machine in construction, and generating a construction plan; simulating the construction plan to verify the feasibility and risk points of the construction plan; and optimizing the construction plan according to the risk points and generating the optimal construction plan.

[0068] In one embodiment of the present invention, the step of mapping on-site construction data into a virtual scene graph specifically includes: digitally mapping the current site into a virtual map; acquiring construction data and converting it into components and attributes of the virtual map; constructing a scene model containing environmental information and construction machinery information, wherein the construction machinery is provided with sensors to provide construction data in real time; associating the components of the scene modeling with the construction data; and storing the components, the attributes and the construction data in the scene to form the virtual scene graph.

[0069] For example, in a large commercial construction project, we can use laser scanning technology and drone photography to obtain accurate 3D data of the construction site. This data is then converted into a virtual map that includes information such as building structures, material storage areas, and the location of construction machinery. For example, an excavator may be represented as a 3D model with specific dimensions, working radius, and current position. At the same time, environmental information such as terrain, weather conditions, etc. can also be incorporated into the virtual scene map. The beneficial effect of this approach is that it can provide a comprehensive and real-time updated virtual environment for subsequent AI planning without interfering with actual construction.

[0070] Next, when constructing the scene model, the present invention adopts a modular approach. Specifically, the current construction map is divided into several building areas, each of which is a component; attribute information is set for each component, including type, attribute value, area range and relationship between components; the construction machinery in the current site is set as a component, and the construction data is set as the attribute value; the construction components and construction machinery components are respectively bound to the relevant machinery through a unique device identifier through wireless technology, and are associated with the sensors on the device controller to perceive the current construction environment in real time; and through the continuous updating of on-site construction data, the virtual scene presents the current construction environment status in real time in a dynamic form.

[0071] To give a specific example, suppose we are building a multi-story office building. We can divide the entire construction site into multiple building areas such as the foundation area, the main structure area, and the interior decoration area. Each area is set as an independent component with its own specific attributes. For example, the attributes of the main structure area may include information such as the current number of construction floors, concrete strength, and steel bar layout. At the same time, construction machinery such as tower cranes and concrete pump trucks are also set as independent components, and their attributes include current location, working status, remaining fuel, etc.

[0072] The benefits of this modular approach are obvious. First, it enables the system to handle complex construction environments more flexibly. Second, by binding construction machinery to actual equipment, we can obtain and update construction data in real time, which is essential for subsequent path planning and risk assessment. Finally, the dynamically updated virtual scene provides an intuitive monitoring platform for construction managers, helping to identify and solve potential problems in a timely manner.

[0073] Preferably, the present invention also includes setting the mutual influence relationship between the components, setting the influence area of ​​the components according to the reaction generated when the components overlap, and performing collision detection between the construction components according to the operation range, and determining whether the components influence each other according to the reaction when the collision occurs. The introduction of this step greatly improves the safety and reliability of construction planning.

[0074] For example, on a construction site, a mobile crane and an excavator may overlap in certain areas. By pre-setting the working ranges and possible mutual influence of the two machines, the system can automatically avoid potential collision risks when planning the path. Specifically, a circular influence area with a radius of 50 meters can be set for the crane, and a rectangular influence area of ​​20 meters by 30 meters can be set for the excavator. When the two areas overlap, the system automatically performs a collision detection and takes this factor into account in the path planning.

[0075] In practical applications, this approach can effectively prevent construction accidents and improve construction efficiency. For example, the system may suggest scheduling the crane's work time in the morning and the excavator's work time in the afternoon, thereby avoiding the two machines working in overlapping areas at the same time. This not only improves safety, but also optimizes the efficiency of machine use.

[0076] Next, a core innovation of the present invention lies in its method of automatically planning a reasonable construction sequence and path for construction machinery. Specifically, this process includes the following steps: randomly selecting a construction area in the current scene, obtaining all construction machinery that have entered the construction site in the area, and obtaining the attribute values ​​of the machinery components; calculating the priority of each construction machinery according to the attribute values; planning the construction path in the current construction area, setting the machinery with high priority in the priority position; and planning the construction sequence of each machinery in the construction area, giving priority to the construction machinery with high priority, and planning each construction machinery on the construction path.

[0077] For example, let's say we are constructing a large commercial center. During the foundation construction phase, the system may first select the foundation excavation area as the current construction area. Then, it will identify all the machinery in the area, such as excavators, dump trucks, rollers, etc. Next, the system will calculate the priority of each machine based on its attribute values ​​(such as work efficiency, remaining fuel, maintenance status, etc.).

[0078] In one embodiment of the present invention, the priority is calculated using an innovative formula: priority = A* urgency + B* importance + C* distance. A is the urgency coefficient of the construction progress, ranging from 0 to 1, with the closer to 1, the more urgent it is; B is the importance coefficient of the construction result; and C is the distance of the current machine from the planned construction location. This formula continuously optimizes parameters through a machine learning algorithm to adapt to different construction environments and requirements.

[0079] For example, for an excavator in a critical position, its urgency may be 0.9, its importance is 0.8, and its distance is 10 meters. Assuming that after machine learning optimization, A = 0.5, B = 0.3, C = -0.01, then the priority of the excavator is: 0.5*0.9+0.3*0.8-0.01*10=0.64. In contrast, an ordinary dump truck may have lower urgency and importance, but is closer, and its priority may be 0.4.

[0080] Based on these priorities, the system will plan the optimal working path and sequence for the excavator to ensure that it can efficiently complete the most critical tasks. At the same time, the dump truck's path will be arranged to maximize its transportation efficiency without interfering with the excavator's work.

[0081] This AI-based planning approach brings significant benefits. First, it is able to dynamically adjust the construction plan according to real-time conditions, improving overall construction efficiency. Second, by prioritizing the work of key machinery, it ensures the steady progress of the construction schedule. Finally, this approach can also optimize resource utilization, reducing machinery idle time and unnecessary fuel consumption.

[0082] In practical applications, the present invention also considers a more macro construction plan. Before randomly selecting a construction area, the system will first obtain the completion date of the project; combined with the completion date, the order of entry and exit of construction machinery is planned in the current virtual scene map to form an overall construction plan; and the construction phase and execution machinery of each construction area are determined, wherein the overall construction plan includes the time for construction in each area.

[0083] The introduction of this step enables the system to ensure local efficiency while ensuring that the overall project is completed on schedule. For example, in a large residential project that is expected to be completed in 18 months, the system may work backwards based on the completion date to determine the time nodes of each construction phase. It may be scheduled to complete the foundation and main structure in the first 6 months, the next 9 months for interior decoration, and the last 3 months for exterior decoration and landscaping. Based on this overall plan, the system will arrange the entry and exit times of various construction machinery accordingly to ensure that an appropriate amount of machinery is available at each stage, so as to avoid waste of resources and delays in construction due to insufficient machinery.

[0084] Finally, the present invention also includes an important verification and optimization step. Specifically, the step of simulating the construction plan to verify the feasibility and risk points of the construction plan includes: verifying through simulation whether the mechanical planning sequence of the construction plan in the virtual scene graph is executable, if not, re-planning; verifying through simulation whether the mechanical path in the virtual scene graph is unobstructed, if not, displaying the unobstructed components and re-planning; verifying through simulation whether the machines in the virtual scene graph have mutual influence, if so, displaying the mutually influential components and re-planning; and using a neural network learning method to train and optimize the construction path planning algorithm.

[0085] This step is important. By simulating in a virtual environment, we can identify potential issues and risks before actually executing. For example, in a complex high-rise project, the simulation may find that the movement path of some large equipment is blocked by temporary scaffolding. In this case, the system will automatically re-plan the path, perhaps suggesting that part of the scaffolding be removed first, and then the equipment can be moved and re-erected.

[0086] Furthermore, the present invention uses a neural network learning method to continuously optimize the planning algorithm. This means that the system can learn from each planning and execution and continuously improve its performance. For example, if the system repeatedly finds that a certain type of machinery is prone to causing jams under certain circumstances, it will automatically take this factor into account in future planning and make adjustments in advance.

[0087] In order to implement the above method, the present invention also provides an AI-based intelligent building construction path planning system. The system includes: a data mapping module 1, which is used to map the on-site construction data into a virtual scene graph, which contains construction machinery information and environmental information; a planning module 2, which is used to automatically plan a reasonable construction sequence and path for each machine in the construction based on the AI ​​algorithm and the virtual scene graph, and generate a construction plan; and a verification module 3, which is used to simulate the construction plan to verify the feasibility and risk points of the construction plan, optimize the construction plan according to the risk points, and generate the optimal construction plan. Among them, the data mapping module 1, the planning module 2 and the verification module 3 are interconnected through a data interface to form a closed-loop intelligent planning system.

[0088] This modular system design has many advantages. First, it improves the maintainability and scalability of the system. For example, if we want to upgrade the data mapping technology, we only need to update the data mapping module 1 without changing other parts. Second, the data interface between modules ensures the smooth flow of information, allowing the entire system to operate as an organic whole. Finally, the closed-loop design means that the system can continuously learn and optimize from actual construction, gradually improving the accuracy and efficiency of its planning.

[0089] In practical applications, this system can provide powerful decision support for construction managers. For example, in a large commercial complex project, the data mapping module 1 may collect data from hundreds of sensors in real time, including the location, working status, surrounding environmental conditions, etc. of various construction machinery. These data are integrated into the virtual scene graph, providing a comprehensive and accurate information basis for the planning module 2.

[0090] Planning module 2 will develop the best work plan for each construction machine based on this information and the overall progress requirements of the project. For example, it may suggest scheduling the work time of a large crane in the morning to avoid the strong wind in the afternoon; or it may suggest adjusting the working areas of several excavators to maximize their synergy.

[0091] Verification Module 3 is used to simulate and optimize these plans. By simulating the entire construction process in a virtual environment, it can identify potential problems in advance. For example, it may find that according to the current plan, there will be traffic congestion for material transport vehicles during a certain period of time. Based on this discovery, the system will automatically adjust the plan and may recommend adding a temporary road or adjusting the travel time of certain vehicles.

[0092] It is worth noting that an important feature of this system is its self-learning ability. By constantly comparing the planning results with the actual construction situation, the system is able to gradually improve the accuracy of its predictions and planning. For example, if the system finds that the actual completion time of a certain type of construction task is always 10% longer than the estimated time, it will automatically take this factor into account in future planning, thereby improving the accuracy of the overall planning.

[0093] In specific implementation, we can set some threshold parameters for the system to control its decision-making process. For example, we can set a safety margin threshold, such as 15%. This means that the system will automatically reserve 15% extra time for each task during planning to deal with possible unexpected situations. At the same time, we can also set an optimization trigger threshold. For example, when the machine utilization rate in a certain area is less than 60%, the system will automatically start the optimization process and reallocate resources. The setting of these thresholds needs to be based on a lot of practical experience and should be adjusted according to different project types and construction environments.

[0094] Another important feature of the present invention is its scalability. Although we have mainly discussed the path planning of construction machinery, this system can actually be extended to the management of the entire construction process. For example, we can incorporate human resource management into the system to achieve the coordinated scheduling of workers and machinery. The system can assign the most suitable work tasks and positions to each worker based on factors such as their skill level and working hours, thereby further improving the overall construction efficiency.

[0095] Finally, it is worth mentioning the potential of this system to improve construction safety. By monitoring and predicting various risk factors on the construction site in real time, the system can issue an alarm before danger occurs. For example, if the system detects that the noise level or dust concentration in a certain area is about to exceed the safety threshold, it will automatically adjust the work plan of the relevant machinery or recommend additional protective measures. This can not only reduce the risk of accidents, but also ensure that the construction process complies with various safety regulations and standards.

[0096] In general, the AI-based intelligent building construction path planning method and system provided by the present invention, by combining advanced artificial intelligence technology with traditional building construction management, has brought revolutionary changes to the construction industry. It can not only significantly improve construction efficiency and reduce resource waste, but also improve construction safety and reduce environmental impact. In the future construction of smart cities, such a system will undoubtedly play an increasingly important role.

[0097] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. The AI-based intelligent building construction path planning method is characterized by: The following steps are involved: Mapping the on-site construction data into a virtual scene graph, wherein the virtual scene graph includes construction machinery information and environment information; Based on the AI ​​algorithm and the virtual scene graph, a reasonable construction sequence and path are automatically planned for each machine in construction, and a construction plan is generated; Simulating the construction plan to verify the feasibility and risk points of the construction plan; and The construction plan is optimized according to the risk points and an optimal construction plan is generated.

2. The method according to claim 1, characterized in that The step of mapping the on-site construction data into a virtual scene graph comprises: Digitally map the current venue as a virtual map; Acquiring construction data and converting it into components and attributes of the virtual map; Constructing a scene model including environmental information and construction machine information, wherein the construction machine is provided with sensors to provide construction data in real time; associating scene modeled components with the construction data; and The components, the attributes and the construction data are stored in a scene to form the virtual scene graph.

3. The method according to claim 2, characterized in that The steps of constructing the scene model include: Divide the current construction map into a number of construction areas, each of which is a component; Setting attribute information for each of the components, including type, attribute value, area range, and relationship between components; Setting the construction machinery in the current site as a component and setting the construction data as an attribute value; Binding construction components and construction machine components to the relevant machines through unique device identifiers through wireless technology, and associating them with sensors on the equipment controller to sense the current construction environment in real time; and By continuously updating the on-site construction data, the virtual scene can present the current construction environment status in real time in a dynamic form.

4. The method according to claim 3, characterized in that The following steps are also included: Setting the mutual influence relationship between the components, and setting the influence area of ​​the components according to the reaction generated when the components overlap; and The construction components are subjected to collision detection according to the operation range, and it is determined whether the components affect each other according to the reaction during the collision.

5. The method according to claim 1, characterized in that The step of automatically planning a reasonable construction sequence and path for the construction machinery includes: Randomly select a construction area in the current scene, obtain all the construction machinery that have entered the construction site in the area, and obtain the attribute values ​​of the machinery parts; Calculating the priority of each construction machine according to the attribute value; Plan construction routes within the current construction area and place high-priority machinery at priority positions; and Plan the construction sequence of each machine in the construction area, give priority to planning the construction machines with high priority, and plan each construction machine to the construction path.

6. The method according to claim 5, characterized in that Before randomly selecting the construction area, the following steps are also included: Get the project completion date; In combination with the completion date, planning the entry and exit sequence of construction machinery in the current virtual scene diagram to form an overall construction plan; and The construction phases and execution machinery for each construction area are determined, wherein the overall construction plan includes the time when each area will be constructed.

7. The method according to claim 5, characterized in that The step of calculating the priority of each construction machine according to the attribute value comprises: Set the calculation formula for construction machinery priority, where construction machinery priority = A*urgency+B*importance+C*distance; Define the parameters of the construction machinery priority calculation formula, where A is the construction progress urgency coefficient, setting the priority between 0 and 1, the closer to 1, the more urgent, B is the construction result importance coefficient, and C is the distance of the current machine from the planned construction location; and The data is trained through a machine learning algorithm to iteratively optimize the formula parameters.

8. The method according to claim 1, characterized in that: The steps of automatically optimizing the construction plan by the AI ​​algorithm include: Set priorities based on the contribution of construction machinery to the project results, with more important construction machinery having higher priorities; Using the priorities to plan paths and arrange construction sequences for machines; Calculate the priority of the current machinery in each construction area; and Plan construction machinery according to regional priorities, giving priority to construction machinery with higher priorities to ensure the progress of the entire project.

9. The method according to claim 1, characterized in that: The step of simulating the construction plan to verify the feasibility and risk points of the construction plan includes: Verify through simulation whether the mechanical planning sequence of the construction plan in the virtual scene graph is executable. If not, re-plan it; Verify through simulation whether the mechanical path in the virtual scene graph is smooth. If not, display the unsmooth parts and replan them; Verify through simulation whether the machines in the virtual scene graph affect each other. If so, display the components that affect each other and re-plan them; and Use neural network learning methods to train and optimize construction path planning algorithms.

10. An AI-based intelligent building construction path planning system for implementing the method according to any one of claims 1 to 9, characterized in that: include: A data mapping module is used to map the on-site construction data into a virtual scene graph, which contains construction machinery information and environmental information; A planning module, which is used to automatically plan a reasonable construction sequence and path for each machine in construction and generate a construction plan based on the AI ​​algorithm and the virtual scene graph; as well as A verification module, used to simulate the construction plan to verify the feasibility and risk points of the construction plan, optimize the construction plan according to the risk points and generate an optimal construction plan; The data mapping module, the planning module and the verification module are interconnected via a data interface to form a closed-loop intelligent planning system.

Images