Large transport route planning method, device and system

Abstract

The invention discloses a method, a device and a system for planning a transportation route of a large cargo. The method comprises the following steps: inputting and loading parameters of a transportation vehicle, cargoes and a transportation scene; performing path planning based on a multi-constraint algorithm to generate a feasible path; performing three-dimensional dynamic simulation on the feasible path; verifying whether the three-dimensional dynamic simulation is passed or not; if not, adjusting the parameters or manually modifying the path and then returning to the previous step; and if yes, outputting a path and a simulation report. According to the invention, comprehensive parameters of transport vehicles, goods and transport scenes are stored and updated in real time through the data support layer, so that planning deviation caused by data missing or lagging is avoided; by means of constraint matching analysis and multi-constraint path planning of the processing layer, feasible paths meeting the requirements of the height, the gradient, the turning radius, the bearing capacity and the like can be screened out, the scheme does not need to be repeatedly adjusted to adapt to actual transportation conditions, and time and labor cost are saved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of transportation, and particularly relates to a large piece transportation route planning method, device and system. BACKGROUND

[0002] Most of the transportation on the market relies on navigation software or GIS systems, but the real-time changes of the on-site conditions in the factory cannot rely on such technologies to assist in planning the transportation route. Moreover, the on-site conditions are harsh, relying on manual on-site investigation of roads and records, which is prone to not timely convey information, miss critical information, low efficiency, and thus lead to the occurrence of collision risks in the transportation process.

[0003] For example, the prior art CN112101866A discloses a logistics scheduling system based on GIS and intelligent algorithm, which can automatically plan the shortest path, supplement the empty vehicles, avoid the congested road sections, and detect the missing goods during driving through GIS intelligent analysis and real-time positioning of vehicles, thereby significantly improving the delivery efficiency and accuracy. As can be seen, in this technology, it mainly focuses on the shortest path planning to improve the delivery efficiency, and in this technology, the path planning is still realized through GIS intelligent analysis and real-time positioning of vehicles. The above problems cannot be effectively solved.

[0004] Therefore, in order to solve the above problems, it is urgent to develop related technical solutions. Based on this, the present application is proposed. SUMMARY

[0005] The purpose of the present application is to provide a large piece transportation route planning method, device and system to solve the problems in the prior art.

[0006] To this end, the present application provides a large piece transportation route planning method, comprising: S1, inputting and loading the parameters of a transportation vehicle, goods and a transportation scene; S2, performing path planning based on a multi-constraint algorithm to generate a feasible path; S3, performing three-dimensional dynamic simulation on the feasible path; S4, verifying whether the three-dimensional dynamic simulation is passed; if not, adjusting the parameters or manually modifying the path and returning to step S2; if yes, outputting the path and a simulation report.

[0007] In some embodiments, in step S1, the parameters of the transportation vehicle include the model, carrying weight, size, turning radius, speed and driving mode of the transportation vehicle; The parameters of the goods include the size and weight of the goods; The parameters of the transportation scene include road information, drivable speed limit, obstacle size and position, and ground bearing weight.

[0008] In some embodiments, step S2 comprises: constructing a road grid map; topological planning based on the road grid map, and generating a topological path; performing geometric constraint verification on the topological path, and screening out feasible paths that satisfy all constraints.

[0009] In some embodiments, the process of constructing a road network map comprises: identifying drivable areas in the transportation scenario; extracting key path points within the drivable areas, and connecting the path points to form edges; assigning attribute costs and constraint conditions to each edge.

[0010] In some embodiments, the process of generating a planning path and performing geometric constraint verification on the topological path comprises: performing height verification, bearing capacity verification, turning radius verification, and slope verification on each topological path in sequence; if the planning path satisfies all verification constraints, storing it in a feasible path set; if it does not satisfy, trying the next topological path; until all topological paths are checked, outputting feasible paths that satisfy the constraints.

[0011] In some embodiments, in step S3, the three-dimensional dynamic simulation of the feasible path comprises starting a three-dimensional scene simulator, inputting scene configuration data, and generating a three-dimensional scene model; performing collision detection based on the three-dimensional scene model.

[0012] In some embodiments, the generation of the three-dimensional scene model comprises: performing type judgment on the input scene configuration data, and respectively invoking a ground generation simulator, a factory building generation simulator, a road generation simulator, or a device generation simulator; parsing parameters and selecting a model template, and sequentially performing the operations of applying size parameters, producing a model grid, adding feature components, and applying materials, and outputting each sub-model to generate a three-dimensional scene model after scene integration.

[0013] In some embodiments, the collision detection based on the three-dimensional scene model comprises: performing preliminary collision detection on the simulation objects using bounding boxes; if the bounding box intersects with the simulation object, performing accurate geometric body calculation for secondary collision detection; according to the collision detection result, if a collision occurs, recording the collision information, and if no collision occurs, marking the key point state; Determine whether the simulated endpoint has been reached. If not, advance the simulation clock one step, update the vehicle's position and attitude based on the path and speed, and repeat the collision detection process until the endpoint is reached and a simulation report is generated.

[0014] On the other hand, a large-item transportation route planning device is also provided, including: The parameter input module is used to input and load parameters of the transport vehicle, goods, and transport scenario. The multi-constraint path planning module is used to construct a road network based on the parameters, perform topology planning and multi-dimensional geometric constraint verification, and generate feasible paths. The 3D simulation module is used to start a 3D scene simulator, input scene configuration data to generate a 3D scene model, and perform 3D dynamic simulation of the feasible path. The collision detection module is used to verify collision risks and record collision information or mark the status of key points during the 3D dynamic simulation process by using bounding box coarse detection and precise geometry calculation. The results output module is used to output the verified path and simulation report after the simulation reaches its endpoint.

[0015] On the other hand, a system for the aforementioned method of planning routes for large-item transportation is also provided, comprising: The data support layer includes a road infrastructure database, a cargo model library, and an environmental data interface, which are used to store parameters of transport vehicles, cargo, and transport scenarios. The processing layer includes a constraint matching analyzer, a multi-constraint path planning engine, and a transportation process simulation engine, which are used to perform path planning, constraint verification, and simulation calculations based on information from the data support layer. The interaction layer includes a mobile interface, a 3D visualization simulator, and a report generation module, which are used to receive user input, display the 3D simulation process, and output the path and simulation report.

[0016] Beneficial effects: 1. This invention stores and updates comprehensive parameters of transport vehicles, goods, and transportation scenarios in real time through a data support layer, avoiding planning deviations caused by missing or outdated data; with the help of constraint matching analysis and multi-constraint path planning in the processing layer, feasible paths that meet requirements such as height, slope, turning radius, and load-bearing capacity can be selected, eliminating the need to repeatedly adjust the plan to adapt to actual transportation conditions and saving time and labor costs. 2. This invention uses a transportation process simulation engine and 3D visualization to intuitively recreate the dynamic process of a transport vehicle traveling along a path, detect potential collision risks between goods and obstacles in advance, and reduce the possibility of damage to goods and roads during actual transportation. 3. By combining the mobile terminal operation interface of the interactive layer with detailed simulation reports, this invention allows users to quickly operate and obtain reference information such as route details and risk point markings without professional technical skills. This improves the accuracy, efficiency and safety of large-item transportation route planning as a whole, and solves the problems of low efficiency and insufficient risk prediction caused by traditional reliance on manual surveys and two-dimensional drawing analysis. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart of the method for planning large-item transportation routes provided by the present invention.

[0019] Figure 2 The flowchart shows the multi-constraint path planning in the large-item transportation route planning method provided by the present invention.

[0020] Figure 3 A flowchart of the three-dimensional simulation in the large-item transportation route planning method provided by the present invention.

[0021] Figure 4 The flowchart shows the collision detection process in the large-item transportation route planning method provided by this invention.

[0022] Figure 5 This is an architecture diagram of the large-item transportation route planning system provided by the present invention. Detailed Implementation

[0023] The invention will be more readily understood by referring to the following detailed description of preferred embodiments and included examples. Unless otherwise specified, 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 pertains. In case of conflict, the definitions in this specification shall prevail.

[0024] like Figures 1-5 As shown, a method for planning routes for large-item transportation includes: S1. Input and load parameters for the transport vehicle, cargo, and transport scenario. Transport vehicle parameters include model, load capacity, dimensions, turning radius, speed, and mode of transport. Cargo parameters cover dimensions, weight, and center position. Transport scenario parameters involve road information, such as slope, width, turning radius, obstacle location and dimensions, and foundation bearing capacity. Inputting this data provides comprehensive and accurate foundational data for subsequent path planning and simulation, preventing discrepancies between planning results and actual needs due to missing or inaccurate parameters.

[0025] S2. Path planning is performed based on a multi-constraint algorithm to generate feasible paths. This multi-constraint algorithm simultaneously considers various constraints, including road geometric constraints (such as slope ≤ 8°, turning radius ≥ minimum turning radius of the transport vehicle), load-bearing constraints (road foundation bearing capacity ≥ total weight of transport vehicle and cargo / contact area), and spatial constraints (cargo sweep body does not collide with obstacles). First, a road network graph is constructed, key path points are extracted, and constraint attributes are assigned to each edge. Then, through topology planning and multi-dimensional constraint verification, paths that satisfy all constraints are selected. This ensures that the generated paths meet transportation requirements in terms of geometry, load-bearing capacity, and spatial constraints, solving the problem of infeasible paths caused by considering only single constraints in existing technologies, and improving the practicality of path planning.

[0026] S3. Perform a three-dimensional dynamic simulation of the feasible path. The three-dimensional dynamic simulation involves constructing a 1:1 three-dimensional model of the actual scene in a computer, simulating the entire process of a transport vehicle carrying goods along the planned path, including turning, climbing, and overcoming obstacles. A physics engine is used to simulate the interaction between the vehicle and the road surface, the inertia and gravity of the goods, and other dynamic characteristics, dynamically generating the sweep envelope of the goods and the vehicle. This visually presents the spatial relationships during transportation, identifies potential spatial conflicts in the path in advance, provides a visual basis for subsequent verification, and compensates for the shortcomings of existing two-dimensional analysis methods.

[0027] S4. Verify whether the 3D dynamic simulation passes. If not, adjust the parameters or manually modify the path and return to step S2. If yes, output the path and simulation report. Passing the verification means that in the 3D dynamic simulation, the cargo sweep body does not collide with obstacles or road facilities and meets all constraints such as road slope and load-bearing capacity. Specifically, through bounding box coarse detection and precise geometry calculation, the spatial relationship between the sweep body and scene elements is judged frame by frame. If a collision is detected or constraints are not met, the transport vehicle parameters are adjusted (e.g., increasing the number of axes) or the path points are modified, and the path planning is re-performed. If all simulation processes meet the requirements, a simulation report containing path details, risk point annotations, and passability conclusions is output. Closed-loop verification ensures the feasibility of the path, avoids safety hazards in actual transportation, and the output report provides detailed guidance for the implementation of the transportation plan, improving the safety and efficiency of transportation.

[0028] In one embodiment, such as Figure 2 As shown, construct a road grid map; Based on the road grid map, topology planning is performed and a topology path is generated; wherein, the drivable area in the transportation scenario is identified, specifically the roads, venues, and other areas in the transportation scenario that allow transportation vehicles to pass, excluding obstacles and restricted areas; Extract key path points within the drivable area, such as intersections, ramp start points, and turning points, which play a crucial role in the path, and connect these path points to form edges; Assign attribute costs and constraints to each edge. Attribute costs are quantifiable costs such as transportation time and travel distance. Constraints include limits such as maximum slope, minimum turning radius, and foundation bearing capacity.

[0029] By using on-site surveys or map data, the drivable area is first determined. Then, key nodes within this area are extracted and connected to form drivable road segments. Simultaneously, travel costs and constraints are marked for each road segment. The technical effect is the creation of a structured road model, ensuring that subsequent route planning is based on the actual drivable area, avoiding the generation of meaningless paths, and improving the rationality of route planning.

[0030] Then, geometric constraint verification is performed on the topological paths to filter out feasible paths that satisfy all constraints. For each topological path, height, load-bearing capacity, turning radius, and slope verification are performed sequentially. Height verification ensures the path's clearance height is greater than the cargo height, guaranteeing cargo passage. Load-bearing capacity verification verifies that the path's foundation bearing capacity can withstand the total weight of the transport vehicle and cargo, preventing road surface damage. Turning radius verification verifies that the path's turning radius is not less than the transport vehicle's minimum turning radius, ensuring smooth turning. Slope verification specifically verifies that the path's slope is not greater than the transport vehicle's maximum climbing gradient, ensuring smooth climbing.

[0031] If the planned path satisfies all validation constraints, it is stored in the feasible path set; If the condition is not met, try the next topological path; Once all topological paths have been checked, output the feasible paths that satisfy the constraints.

[0032] By comparing the geometric parameters of each topological path with the parameters of the transport vehicle, cargo, and road constraints one by one, it is determined whether the requirements are met. Through multi-dimensional constraint verification, paths that are feasible in terms of height, load-bearing capacity, turning, and gradient are selected, avoiding transportation obstacles caused by unmet constraints and ensuring the practicality and safety of the paths.

[0033] In one embodiment, such as Figures 3-4As shown, performing a three-dimensional dynamic simulation of the feasible path includes starting a three-dimensional scene simulator, inputting scene configuration data, and generating a three-dimensional scene model. Generating a three-dimensional scene model includes determining the type of the input scene configuration data and calling a ground generation simulator, a factory generation simulator, a road generation simulator, or an equipment generation simulator, respectively. The aforementioned scene configuration data can include a dataset of information such as the dimensions, location, material, and physical properties of the ground, factory buildings, roads, and fixed equipment in the transportation scene, including road width, factory building height, and equipment coordinates. It determines the category of the scene element corresponding to the data: ground, factory buildings, roads, or fixed equipment. The ground generation simulator is a tool for generating 3D models of the ground in the transportation scene, capable of simulating different road surface materials, smoothness, and slope. The factory building generation simulator is specifically designed for generating 3D models of factory buildings in the transportation scene, simulating the position and shape of factory walls, doors, windows, and columns. The road generation simulator is specifically designed for generating 3D models of roads in the transportation scene, simulating road width, turning radius, and road markings. The equipment generation simulator is specifically designed for generating 3D models of fixed equipment in the transportation scene, such as machine tools and warehouse shelves in a workshop. These simulators can generate diverse environments to simulate real-world conditions.

[0034] The parameters are parsed and a model template is selected. The operations of applying size parameters, generating model mesh, adding feature components and applying materials are performed in sequence. After outputting each sub-model, the scene is integrated to generate a 3D scene model. By splicing and combining each sub-model in 3D space according to the spatial position information in the scene configuration data, a complete transportation scene layout is formed.

[0035] The model template refers to a pre-built, adjustable 3D model framework for the ground, factory buildings, roads, and equipment, avoiding the need to build models from scratch. Size parameters assign the size information from the scene configuration data to the model template, such as setting the road width to 10 meters and the factory span to 25 meters. The production model mesh converts the adjusted model template into a 3D mesh structure composed of polygons, giving the model spatial form and outline. Feature components add detailed features to the model, such as road dividers, factory windows, and ground drainage channels. Applying materials assigns corresponding material properties to the model mesh, such as asphalt for roads and concrete for factory walls, ensuring the model's visual effect matches reality. Sub-models refer to the separately generated independent 3D models of the ground, factory buildings, roads, and equipment.

[0036] Understandably, in this embodiment, the type of scene configuration data is first determined to identify the categories of scene elements to be generated, thereby calling the corresponding simulator to ensure that the model generation of each scene element conforms to its own characteristics. Then, based on the pre-built model template, the accuracy of each sub-model is gradually improved by parsing parameters to adjust the size, generating meshes to give shape, adding features to supplement details, and applying materials to restore appearance. Finally, all sub-models are integrated according to the spatial location of the actual scene to form a 3D model consistent with the structure and shape of the real transportation scene. By calling the simulator by category and building the model step by step, the actual shape and attributes of various elements in the transportation scene can be accurately restored, avoiding simulation deviations caused by the inconsistency between the scene model and the real environment. At the same time, the template-based parameter adjustment method reduces the repetitive workload of model building, improves the generation efficiency of the 3D scene model, and provides a realistic scene foundation for the dynamic simulation of feasible paths, ensuring that the simulation process can truly reflect the driving state of the transport vehicle in the actual scene.

[0037] like Figure 4 As shown, collision detection is performed based on the 3D scene model. Specifically, bounding boxes are used to perform preliminary collision detection on the simulated objects; If the bounding box intersects with the simulated object, precise geometric calculations are performed for secondary collision detection. For example, a bounding box refers to a simple geometric shape that encloses the simulated object, such as a cuboid or sphere. This is a preliminary detection tool that quickly determines whether there is a potential collision between objects without complex calculations. For example, intersection means that the bounding boxes of two simulated objects overlap, indicating a potential collision risk that requires further verification.

[0038] Based on the collision detection results, if a collision occurs, the collision information is recorded; if no collision occurs, the key point status is marked. The system determines whether the simulated endpoint has been reached. If not, the simulation clock advances one step, and the vehicle's position and attitude are updated based on the path and speed. The collision detection process is repeated until the endpoint is reached, at which point a simulation report is generated. The simulation report records and summarizes the collision records, key point status data, path clearance status, and final passability conclusions for the entire simulation process. The simulation clock is a time measurement tool that controls the 3D dynamic simulation process; each advance corresponds to a fixed time unit in actual transportation, such as 1 second or 0.5 seconds.

[0039] Understandably, preliminary collision detection is performed using bounding boxes, leveraging the fast computational capabilities of simple geometry to eliminate a large number of scenarios with no collision risk, thus reducing the amount of subsequent computation.

[0040] Only in scenarios with potential collision risks (boundary box intersections) is a secondary detection performed through precise geometric calculations to ensure the accuracy of collision judgment. Simultaneously, the simulation progresses step-by-step over time, updating vehicle status and repeating the detection at each step until the entire transport path is covered.

[0041] The initial detection significantly reduces the overall computational load of collision detection, avoids simulation lag caused by precise calculations throughout the process, and improves simulation efficiency; the secondary detection makes up for the insufficient accuracy of bounding box detection, ensuring that no real collision risks are missed and guaranteeing the reliability of the detection results. Real-time recording of collision information and marking of key point status can provide detailed data support for subsequent analysis of transportation risks. The detection method that proceeds step by step over time can comprehensively cover the entire transportation process, avoiding safety hazards caused by omissions in local detection. The final simulation report can also intuitively present the feasibility of the route, providing a clear reference for actual transportation.

[0042] On the other hand, the present invention also provides a large-item transportation route planning device, comprising: The parameter input module is used to input and load parameters of the transport vehicle, goods, and transport scenario. The multi-constraint path planning module is used to construct a road network based on the parameters, perform topology planning and multi-dimensional geometric constraint verification, and generate feasible paths. The 3D simulation module is used to start a 3D scene simulator, input scene configuration data to generate a 3D scene model, and perform 3D dynamic simulation of the feasible path. The collision detection module is used to verify collision risks and record collision information or mark the status of key points during the 3D dynamic simulation process by using bounding box coarse detection and precise geometry calculation. The results output module is used to output the verified path and simulation report after the simulation reaches its endpoint.

[0043] like Figure 5 As shown, the present invention also provides a system for applying a large-item transportation route planning method, comprising: The data support layer includes a road infrastructure database, a cargo model library, and an environmental data interface, used to store parameters of transport vehicles, cargo, and transport scenarios. Specifically, the road infrastructure database can store geometric parameters of roads such as width, length, and turning radius; physical parameters such as foundation bearing capacity and road surface slope; and location parameters such as the coordinate range of the road within the transport scenario. The cargo model library can store 3D model data of cargo such as length, width, and height; physical parameters such as weight and center of gravity; and attribute information such as whether it is fragile or requires special fixing methods. The environmental data interface can connect to on-site sensors or a map update system to obtain dynamic parameters in the transport scenario in real time, such as the location of temporary obstacles, the road surface friction coefficient in rainy weather, and the scope of construction areas.

[0044] The processing layer includes a constraint matching analyzer, a multi-constraint path planning engine, and a transportation process simulation engine. These are used for path planning, constraint verification, and simulation calculations based on information from the data support layer. Specifically, the constraint matching analyzer can, for example, convert parameters in the data support layer into quantifiable constraint rules. For instance, converting the road foundation bearing capacity of 50 tons / m² into a condition that the total weight of the transport vehicle and cargo ÷ contact area ≤ 50 tons / m², and converting the cargo height of 4 meters into a constraint standard that the path clearance height ≥ 4 meters. The multi-constraint path planning engine can, for example, call road data from a road infrastructure database to construct a path network, combine it with the rules generated by the constraint matching analyzer, filter out candidate paths that do not violate constraints, and optimize the path by calculating indicators such as transportation distance and time. The transportation process simulation engine can, for example, extract cargo data from a cargo model library and scene data from a road infrastructure database to construct a virtual transportation scenario, simulating the dynamic state of the transport vehicle traveling along the planned path, such as the vehicle's tilt angle when turning and the speed change when climbing.

[0045] The interaction layer includes a mobile interface, a 3D visualization simulator, and a report generation module. These modules receive user input, display the 3D simulation process, and output the route and simulation report. The mobile interface, for example, provides a visual operation window allowing users to select a transport vehicle model from a drop-down menu, enter the cargo weight in an input box, and click a button to trigger route planning, while simultaneously displaying the simulation progress, such as "Simulation progress: 60%". The 3D visualization simulator, for example, presents the dynamic state generated by the transportation process simulation engine in a 3D view, such as showing the transport vehicle driving on the road and the spatial distance between the cargo and the factory pillars. Users can zoom and rotate the screen to view details. The report generation module, for example, automatically summarizes key data from the simulation process, such as whether a collision occurred and the speed at turning points, generating a document containing a route diagram, risk point markings (e.g., the speed at turning point B needs to be reduced to 5 km / h), and passability conclusions (e.g., the planned route meets all transportation constraints and is ready for execution).

[0046] Understandably, in this embodiment, the data support layer first stores the static and dynamic parameters of transport vehicles, goods, and transport scenarios by category to ensure that the data is orderly and can be updated in real time, thus guaranteeing the comprehensiveness and accuracy of the data used in subsequent calculations and avoiding deviations in path planning due to missing data.

[0047] The processing layer calls the parameters of the data support layer, first transforms the constraint rules through the constraint matching analyzer, then generates and optimizes the path by the multi-constraint path planning engine, and finally simulates the transportation dynamics through the transportation process simulation engine to ensure that the planned path meets the actual transportation constraints and that the simulation process can reflect the real transportation status, reducing adjustments in actual transportation.

[0048] The interaction layer serves as the connection channel between the user and the system, receiving user operation commands, displaying the calculation results of the processing layer, and generating reports that can be directly referenced. This allows users to quickly understand the transportation plan without professional technical skills, improving the ease of use of the system and the reference value of the plan.

[0049] Finally, 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. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for planning routes for transporting large items, characterized in that, include: S1. Input and load parameters for the transport vehicle, goods, and transport scenario; S2. Path planning is performed based on multi-constraint algorithms to generate feasible paths; S3. Perform a three-dimensional dynamic simulation of the feasible path; S4. Verify whether the three-dimensional dynamic simulation is successful; if not, adjust the parameters or manually modify the path and return to step S2; if yes, output the path and simulation report.

2. The planning method according to claim 1, characterized in that, In step S1, the parameters of the transport vehicle include the model, load capacity, size, turning radius, speed, and driving mode of the transport vehicle; The cargo parameters include cargo dimensions and cargo weight; The parameters of the transportation scenario include road information, driving speed limit, obstacle size and location, and the weight that the foundation can bear.

3. The planning method according to claim 1, characterized in that, Step S2 includes: Construct a road grid map; Based on the road grid map, topology planning is performed, and topology paths are generated; Perform geometric constraint verification on the topological path to filter out feasible paths that satisfy all constraints.

4. The planning method according to claim 3, characterized in that, The process of constructing the road network map includes: Identify the drivable areas in the transportation scenario; Extract the key path points within the drivable area and connect the path points to form edges; Assign attributes, costs, and constraints to each edge.

5. The planning method according to claim 3, characterized in that, The process of generating a planned path and performing geometric constraint verification on the topological path includes: For each topological path, the height, bearing capacity, turning radius, and slope are verified sequentially. If the planned path satisfies all validation constraints, it is stored in the feasible path set; If the condition is not met, try the next topological path; Once all topological paths have been checked, output the feasible paths that satisfy the constraints.

6. The planning method according to claim 1, characterized in that, In step S3, performing a three-dimensional dynamic simulation of the feasible path includes starting a three-dimensional scene simulator, inputting scene configuration data, and generating a three-dimensional scene model. Collision detection is performed based on the aforementioned 3D scene model.

7. The planning method according to claim 6, characterized in that, The generated 3D scene model includes: The system determines the type of the input scene configuration data and calls the ground generation simulator, factory generation simulator, road generation simulator, or equipment generation simulator accordingly. The parameters are parsed and a model template is selected. Then, the operations of applying size parameters, generating model mesh, adding feature components and applying materials are performed in sequence. After outputting each sub-model, the scene is integrated to generate a 3D scene model.

8. The planning method according to claim 6, characterized in that, The collision detection based on the 3D scene model includes: Preliminary collision detection of the simulated object is performed using bounding boxes; If the bounding box intersects with the simulated object, then precise geometry calculations are performed for secondary collision detection; Based on the collision detection results, if a collision occurs, the collision information is recorded; if no collision occurs, the key point status is marked. Determine whether the simulated endpoint has been reached. If not, advance the simulation clock one step, update the vehicle's position and attitude based on the path and speed, and repeat the collision detection process until the endpoint is reached and a simulation report is generated.

9. A device for planning routes for large-item transportation, characterized in that, include: The parameter input module is used to input and load parameters of the transport vehicle, goods, and transport scenario. The multi-constraint path planning module is used to construct a road network based on the parameters, perform topology planning and multi-dimensional geometric constraint verification, and generate feasible paths. The 3D simulation module is used to start a 3D scene simulator, input scene configuration data to generate a 3D scene model, and perform 3D dynamic simulation of the feasible path. The collision detection module is used to verify collision risks and record collision information or mark the status of key points during the 3D dynamic simulation process by using bounding box coarse detection and precise geometry calculation. The results output module is used to output the verified path and simulation report after the simulation reaches its endpoint.

10. A system for planning large-item transportation routes according to any one of claims 1-8, characterized in that, include: The data support layer includes a road infrastructure database, a cargo model library, and an environmental data interface, which are used to store parameters of transport vehicles, cargo, and transport scenarios. The processing layer includes a constraint matching analyzer, a multi-constraint path planning engine, and a transportation process simulation engine, which are used to perform path planning, constraint verification, and simulation calculations based on information from the data support layer. The interaction layer includes a mobile interface, a 3D visualization simulator, and a report generation module, which are used to receive user input, display the 3D simulation process, and output the path and simulation report.

Images