A Path Control Method for a Multi-Machine Collaborative Mobile Measurement Robot

Abstract

This invention discloses a path control method for a multi-robot collaborative mobile measurement robot, belonging to the field of path planning technology. Specifically, it includes: collecting surface height information of the target area and generating a simplified 3D model by binding the surface height information with geographic coordinates; calculating the cumulative duration distribution map of direct sunlight received by each coordinate point on the ground within the current time period based on the solar trajectory data of the current season; dividing the area into high solar intensity regions and continuous shadow regions according to the cumulative duration distribution map, and topologically associating the coordinates of the target point to be measured with the continuous shadow regions; establishing a dynamic path allocation protocol in the multi-robot communication network, enabling each robot to select a non-overlapping measurement route in the path network formed by the continuous shadow regions based on its real-time position and task queue, and updating the path occupancy status through periodic broadcasts; this invention optimizes the path planning of mobile measurement robots in urban high-temperature environments.

Description

Technical Field

[0001] This invention relates to the field of path planning technology, and more specifically to a path control method for a multi-machine collaborative mobile measurement robot. Background Technology

[0002] With the continued acceleration of global urbanization, the demand for urban infrastructure construction and renovation is increasing daily, placing higher demands on the precision of urban surveying. Traditional surveying methods mainly rely on manual mapping or fixed surveying equipment, which are not only inefficient but also difficult to adapt to the complex and varied terrain and dense building distribution in modern urban environments. Against this backdrop, mobile surveying robots, with their autonomous mobility, flexible deployment, and efficient operation modes, are gradually becoming an important technological tool in the field of urban surveying. Especially in application scenarios such as 3D modeling of large urban areas, inspection of municipal facilities, and road network mapping, mobile surveying robots demonstrate significant technological advantages.

[0003] However, in practical applications, mobile measurement robots face severe challenges in high-temperature environments during summer. When robots operate for extended periods in direct sunlight, their internal electronic components and core sensors experience temperature increases due to continuous heat accumulation. This can lead to not only performance degradation and data drift, but in severe cases, even system malfunctions or hardware damage. This problem is particularly pronounced in southern my country, where summer heat is prolonged and solar radiation is intense. Furthermore, the prevalent heat island effect in urban environments further exacerbates the risk of overheating. Therefore, effectively reducing the thermal load on equipment by optimizing the robot's movement path, while ensuring the successful completion of measurement tasks, has become a critical issue that urgently needs to be addressed in the development of mobile measurement robot technology. Summary of the Invention

[0004] The purpose of this invention is to provide a path control method for a multi-machine collaborative mobile measurement robot, solving the following technical problems:

[0005] When the robot's path is mostly in direct sunlight, it may affect the operation of the equipment.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A path control method for a multi-machine collaborative mobile measurement robot includes the following steps:

[0008] Collect surface height information of the target area, bind the surface height information with geographic coordinates, and generate a simplified 3D model that includes building edges, tree canopies, and the tops of fixed equipment;

[0009] Based on the solar trajectory data for the current season, calculate the cumulative duration of direct sunlight received at each coordinate point on the ground within the current time period;

[0010] Based on the cumulative duration distribution map, high solar radiation intensity areas and continuous shadow areas are divided, and the coordinates of the target point to be measured are topologically correlated with the continuous shadow areas;

[0011] A dynamic path allocation protocol is established in the multi-robot communication network, enabling each robot to select a non-overlapping measurement route in the path network formed by continuous shaded areas based on its real-time position and task queue, and to update the path occupancy status through periodic broadcasts.

[0012] As a further aspect of the present invention: the process of generating the simplified 3D model is as follows:

[0013] The surface height information of the target area is collected by the first robot and the second robot. The first robot moves along the center line of the road and collects the vertical height profile of the buildings on both sides. The second robot rotates and scans at a fixed point at the intersection to obtain the horizontal projection shape of the tree canopy.

[0014] When using polygonal prisms for spatial fitting of building outlines, the bottom boundary of the prism is aligned with the building base in the satellite map, and the top height is corrected layer by layer based on vertical scan data. When calculating the minimum enclosing cone for the projected shape of tree canopies, the diameter of the cone's base is determined based on the maximum projection extension distance, and the vertex height is taken from the measured value of the highest point of the canopy. After superimposing the prism and cone models according to geographic coordinates, the spatial overlap area between the models is detected, the redundant volume of the overlapping part of the prism and cone is deleted, and the independent structures are retained to form a three-dimensional spatial topology with object type identification.

[0015] As a further aspect of the present invention: during the scanning process of the horizontal projection shape of the tree canopy, a multi-height-layer scanning strategy is adopted for densely arranged tree groups, specifically:

[0016] The first layer of scanning is performed on a plane below the height of the lowest branch of the canopy to obtain the distribution position of the trunk and the outer contour of the tree group; the second layer of scanning is performed at the height of the middle layer of the canopy to capture the extension direction of the main branches and the distribution of canopy gaps; the third layer of scanning is performed at the height of the top of the canopy to record the concave and convex features of the canopy surface; a composite cone model is constructed based on the three layers of scanning data, decomposing the actual canopy into a superimposed structure of multiple sub-cones. The axial position of the sub-cones is offset according to the extension direction of the branches, and the bottom diameter is set in segments according to the maximum extension value of the contour of each layer of scanning, so that the projection range of the composite model matches the actual shadow shape.

[0017] As a further aspect of the present invention: the calculation process of the cumulative duration distribution map is as follows:

[0018] The target area is divided into equally spaced grids, and each grid node is associated with geographic coordinate information. For each grid node, the sequence of changes in solar azimuth and elevation angles within the target time period is traversed. Based on the spatial position and shape parameters of each object in the 3D model, the time interval in which the node is covered by the projection of different objects is calculated.

[0019] For the prism model of a building, parallel light rays are projected along the direction of the solar azimuth angle, and a dynamic projection trajectory is generated based on the geometric parameters of the side edge of the prism. The coverage time of the projection trajectory on the ground grid nodes is recorded. For the cone model of a tree, the ground projection area is dynamically adjusted according to the geometric relationship between the cone apex angle and the solar altitude angle. The fuzzy transition range of the projection edge is calculated by combining the canopy density coefficient. The projection coverage time of each object is superimposed and statistically analyzed according to the grid nodes to generate the cumulative duration data of each node not being covered by any projection, forming a ground illumination heat map that changes over time.

[0020] As a further aspect of the present invention: during the generation of the illumination heat map, real-time meteorological data is introduced to dynamically correct the projection calculation: when cloud cover forecast occurs within the target time period, the transmittance parameter is matched according to the cloud type code, and the cumulative duration data of the grid nodes is attenuated within the corresponding time interval; for sudden local cloud changes, the cloud movement speed and coverage area are obtained through the meteorological data interface, the movement trajectory of the cloud projection is simulated in the three-dimensional model space, the projection is regarded as a temporary virtual object and added to the projection calculation queue, and superimposed with the original fixed object projection for analysis, updating the cumulative duration data of the affected grid nodes, triggering the reconstruction of the local heat map and the real-time adjustment of the path network.

[0021] As a further aspect of the present invention: the process of topological association is as follows:

[0022] Virtual measurement path nodes are established within a continuous shaded area. The node spacing is dynamically adjusted according to the geometry of the shaded area. For strip-shaped shaded areas, main nodes are equidistantly arranged along their long axis, and diagonal connecting branches are inserted between the main nodes. For measurement points that must pass through areas with high solar radiation intensity, the distribution pattern of the surrounding shaded areas is analyzed, and multiple candidate connecting paths are constructed between adjacent shaded areas. The path priority is adjusted according to the spatial relationship between the current solar altitude angle and the direction of the candidate paths.

[0023] When the solar altitude angle exceeds the set threshold, the path segment perpendicular to the solar azimuth angle direction in the connecting path is selected as the priority route; when the solar altitude angle is below the threshold, the path segment consistent with the building projection extension direction in the connecting path is selected, and a temporary shadow coverage time window is associated to dynamically switch paths.

[0024] As a further aspect of the present invention: the construction of the candidate connection path specifically includes:

[0025] Analyze the boundary orientation of adjacent shaded areas within a high-sunlight region. When the extension trend of the two shaded area boundaries meets the angle condition, a virtual shaded corridor is created in the area where the extension directions intersect. The effective duration of the virtual shaded corridor is calculated based on the relationship between the rate of change of the solar azimuth angle and the corridor orientation. When the angle between the direction of change of the solar azimuth angle and the corridor orientation decreases, the effective duration of the corridor is extended. During path selection, the virtual shaded corridor is topologically connected with the actual shaded area to generate a composite path that crosses the sunlight region. Before the effective duration of the corridor ends, a path failure warning is sent to the robot, triggering dynamic switching of subsequent path segments.

[0026] As a further aspect of the present invention: when the dynamic path allocation protocol is executed, each robot sends a path update message to the central coordinator containing the current shadow zone number, task progress, and request for the next target shadow zone; the central coordinator maintains a global shadow zone occupancy status table, which records the current occupying robot identifier and the expected release time of each shadow zone;

[0027] When multiple robots request to enter the same shaded area, the remaining task queues of each requesting robot are extracted, the number of consecutive connections in the shaded area in its subsequent path is calculated, and robots with fewer shaded area connections are given priority to enter the target area. Alternative path instructions are sent to robots that have not been assigned a shaded area. The alternative path is generated by the central coordinator by recombining the connection nodes of adjacent free shaded areas according to the current shaded area occupancy status, and includes time synchronization information for path switching.

[0028] As a further aspect of the present invention: the generation process of the alternative path instruction is as follows:

[0029] The central coordinator searches for the set of currently unoccupied shaded areas. Based on the current position of the requesting robot and the distribution of the target points to be measured, it searches for continuous path segments in the vacant shaded areas that meet the task progress requirements. When a continuous path cannot be formed in the vacant shaded areas, the path bridging mode is activated. The system searches for transitional path segments in the sunlit area that meet the temporary shade coverage time window. The selection of transitional path segments is based on the geometric relationship between the current solar altitude angle and the height of surrounding buildings. The theoretical shade coverage rate of the transitional path segments within the specified time window is calculated. Only road segments that meet the coverage rate are selected to be added to the alternative path sequence, forming a composite passage route with mixed shade and temporary shade coverage.

[0030] The beneficial effects of this invention are:

[0031] This invention effectively solves the technical challenge of overheating in mobile measurement robots operating in high-temperature urban environments by combining multi-robot collaborative measurement with intelligent path planning. First, an innovative method of collecting vertical contours of buildings and horizontal projections of tree canopies is used to construct a simplified 3D model containing precise object type identifiers. Second, based on the sun's trajectory and the cumulative duration distribution map calculated from the 3D model, dynamic and accurate division of ground shadow areas is achieved, overcoming the inaccuracy of existing technologies in calculating projections of irregular objects such as trees. Finally, through the establishment of a dynamic path allocation protocol and a shadow area topology association mechanism, multiple robots can intelligently allocate shadow resources and plan non-overlapping measurement routes, effectively avoiding the problems of multi-robot path conflicts and shadow resource competition in traditional methods. In particular, the multi-height layer tree scanning strategy and composite cone modeling method employed in this invention significantly improve the accuracy of tree shadow calculation; while innovative mechanisms such as virtual shadow corridors and path backoff compensation ensure optimal path selection when traversing sunlit areas. The synergistic effect of these key technical features enables the measurement robot to minimize direct sunlight time while ensuring task integrity, thereby significantly reducing equipment heat load and improving system stability. The overall solution achieves a significant improvement in equipment thermal management through intelligent path optimization without requiring additional heat dissipation equipment, demonstrating outstanding technical advantages and application value. Attached Figure Description

[0032] The invention will now be further described with reference to the accompanying drawings.

[0033] Figure 1 This is a flowchart illustrating the present invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Please see Figure 1 As shown, this invention provides a path control method for a multi-machine collaborative mobile measurement robot, comprising the following steps:

[0036] First, a multi-robot collaborative model was adopted for surface height information acquisition and model building in the target area. The first robot moved along the road centerline, using its high-precision LiDAR and image acquisition equipment to vertically scan the buildings on both sides of the road, obtaining their vertical height contour data. The second robot, positioned at an intersection, precisely captured the horizontal projection shape of the tree canopy through a 360-degree rotation scan. After accurately binding the acquired surface height information with geographic coordinates, the building contours were spatially fitted using polygonal prisms. The prism's base boundary was strictly aligned with the building's base in the high-resolution satellite map, and the top height was corrected layer by layer based on the vertical scan data to ensure the model matched the actual building height. For the tree canopy projection shape, its minimum bounding cone was calculated. The cone's base diameter was determined based on the maximum projection extension distance, and the vertex height was taken from the measured value of the highest point of the canopy. After superimposing the prism and cone models according to geographic coordinates, spatial analysis algorithms were used to detect overlapping areas between models, removing redundant volumes and retaining independent structures to form a simplified 3D model containing accurate object type identifiers, providing a precise data foundation for subsequent calculations.

[0037] Next, based on the solar trajectory data for the current season, the target area is divided into equally spaced grids, with each grid node associated with detailed geographic coordinate information. For each grid node, by traversing the sequence of changes in the solar azimuth and elevation angles within the target time period, and combining this with the spatial position and shape parameters of each object in the 3D model, the time intervals during which the node is covered by projections from different objects are calculated. For the prism model of a building, parallel light rays are projected along the direction of the solar azimuth, and a dynamic projection trajectory is generated based on the geometric parameters of the prism's side edges. For the conical model of a tree, the ground projection area is dynamically adjusted according to the geometric relationship between the cone's apex angle and the solar elevation angle, and the fuzzy transition range of the projection edge is calculated by combining the canopy density coefficient. The projection coverage time of each object is superimposed and statistically analyzed by grid node to generate the cumulative duration data of each node not being covered by any projection, thereby drawing a distribution map of the cumulative duration of direct sunlight received by each ground coordinate point over time.

[0038] Subsequently, based on the cumulative duration distribution map, a threshold determination method was used to divide the area into high solar intensity regions and continuous shadow regions. Virtual measurement path nodes were established within the continuous shadow regions, with the node spacing dynamically adjusted according to the geometry of the shadow area. For measurement points that must pass through high solar intensity regions, the distribution pattern of the surrounding shadow areas was analyzed, multiple candidate connection paths were constructed, and the path priority was adjusted based on the spatial relationship between the current solar altitude angle and the direction of the candidate paths, thus achieving a topological association between the coordinates of the target point and the continuous shadow region.

[0039] Finally, a dynamic path allocation protocol is established in the multi-robot communication network. Each robot sends a path update message to the central coordinator in real time, containing the current shaded area number, task progress, and request for the next target shaded area. The central coordinator maintains a global shaded area occupancy status table, recording the currently occupying robot identifier and estimated release time for each shaded area. When multiple robots request entry into the same shaded area, the central coordinator extracts the remaining task queues of each requesting robot, calculates the number of consecutive connections in the shaded areas of its subsequent path, and prioritizes allocating to robots with fewer connections. Simultaneously, the central coordinator generates alternative path instructions for robots that are not allocated, ensuring that each robot selects a non-overlapping measurement route in the path network composed of consecutive shaded areas based on its real-time position and task queue, and updates its path occupancy status through periodic broadcasts, achieving efficient multi-robot collaborative operation.

[0040] In a preferred embodiment of the present invention, the process of generating the simplified 3D model is as follows:

[0041] During the data acquisition phase, a dual-robot collaborative mode was adopted. The first robot, acting as the main force for building outline acquisition, moved steadily along the road centerline, using its high-precision LiDAR and image acquisition equipment to perform a comprehensive vertical scan of the buildings on both sides of the road. During the scanning process, the equipment continuously collected distance and position information of various points on the building surface, accurately outlining the vertical height contour of the buildings. The second robot was stationed at the intersection, performing a 360-degree rotating scan with itself as the center. Its professional scanning device could quickly and comprehensively acquire horizontal projection shape data of tree canopies, recording the extension range of the canopy from different angles. The two robots tightly bound the collected surface height information with geographic coordinates, laying a solid data foundation for subsequent modeling.

[0042] In the model building phase, polygonal prisms were used for spatial fitting of the building outlines. The base boundary of the prisms was precisely aligned with the building base in the high-precision satellite map to ensure the accuracy of the model's planar position. Based on detailed data obtained from vertical scanning, the height of the prism's top was meticulously corrected layer by layer to ensure that the model's height matched the actual building height. For the tree canopy, a model was constructed by calculating the minimum bounding cone. The diameter of the cone's base was determined based on the maximum extension distance of the canopy projection, and the vertex height was directly adopted from the measured value of the highest point of the canopy, thus forming a preliminary tree model. Subsequently, the building prism model and the tree cone model were superimposed and integrated according to geographic coordinates. A specialized detection algorithm was used to identify the spatial overlap areas between the models, and redundant volumes in the overlapping parts were decisively deleted, retaining only independent and accurate structures, ultimately forming a three-dimensional spatial topology structure with clear object type identification.

[0043] Faced with densely packed tree clusters, an innovative multi-height layer scanning strategy was adopted when scanning the horizontal projection shape of the tree canopy. The first layer of scanning was conducted at a plane below the height of the lowest branch of the canopy, clearly capturing the distribution of the trunks and the overall outline of the tree cluster. The second layer of scanning was performed at the middle level of the canopy, focusing on capturing the extension direction of the main branches and the distribution of canopy gaps. A third layer of scanning was performed at the top of the canopy, meticulously recording the surface features of the canopy. Based on the data obtained from these three layers of scanning, a composite cone model was constructed, cleverly decomposing the actual tree canopy into a superimposed structure of multiple sub-cones. The axial positions of the sub-cones were flexibly offset according to the branch extension direction, and the base diameter was set segmentally according to the maximum extension value of the scanned contours of each layer. This ensured that the projection range of the combined model closely matched the actual shadow shape, significantly improving the model's accuracy and fidelity in reproducing complex tree morphologies.

[0044] In another preferred embodiment of the present invention, the calculation process of the cumulative duration distribution map is as follows:

[0045] First, the target area is divided into a regular grid matrix with a resolution of 5 meters × 5 meters, and each grid node is precisely associated with geographic coordinates and elevation information. For each grid node, the sequence of changes in solar azimuth and elevation angles within the target time period (e.g., 8:00-18:00) is traversed at 10-minute time steps. Using the precise geometric parameters of each object in the 3D model, a ray tracing algorithm is used to calculate the time interval within which the node is covered by the projection of different objects within each time step.

[0046] For the prism model of a building, parallel light beams are projected along the solar azimuth direction. Based on the geometric parameters of the prism's side edges, a dynamic projection trajectory is generated using spatial analytical geometry methods. The system records the coverage time of the projection trajectory on the ground grid nodes in real time, forming a spatiotemporal distribution dataset of the building's shadow. For the cone model of a tree, the major and minor axis parameters of the ground projection ellipse are dynamically adjusted according to the geometric relationship between the cone's apex angle and the solar altitude angle. Simultaneously, combined with the canopy density coefficient (obtained through laser point cloud echo intensity inversion), a Gaussian blur algorithm is used to calculate the illumination attenuation transition range at the projection edge, making the shadow simulation more closely resemble the real environment.

[0047] After calculating the projection of individual objects, the system performs spatiotemporal overlay statistics on the projection coverage time of each object at each grid node. By integrating building and tree shadow data through Boolean operations, it generates cumulative duration data for each grid node that was not covered by any projection within the target time period. Using heatmap visualization technology, the cumulative duration data is mapped to different color gradients, forming a heatmap that intuitively reflects the distribution of ground illumination.

[0048] In the process of generating the solar radiation heat map, an innovative real-time meteorological data dynamic correction mechanism is introduced. When the meteorological forecasting system detects cloud cover within the target time period, the system automatically matches the corresponding transmittance parameter (range 0.2-0.8) based on the cloud type code (such as cumulus, stratus, etc.). Within the time interval of cloud cover, the cumulative duration data of the corresponding grid nodes is attenuated and corrected, with the correction coefficient being inversely proportional to the cloud transmittance.

[0049] In response to sudden localized cloud changes, the system acquires parameters such as cloud movement speed, direction, and coverage in real time through a meteorological data interface. Within the 3D model space, cloud projections are abstracted as virtual objects with dynamic boundaries, and a particle system is used to simulate the movement trajectory of the cloud projections. The cloud projections are spatiotemporally overlaid with existing fixed object projections to update the cumulative duration data of affected grid nodes. When changes in local illumination conditions exceed a preset threshold (e.g., cumulative duration change rate > 15%), the system automatically triggers rapid reconstruction of the local heatmap and adjusts the path network in real time to adapt to the dynamic changes in illumination conditions.

[0050] By employing this multi-dimensional spatiotemporal analysis and dynamic data fusion method, this invention significantly improves the accuracy of shadow area segmentation. Especially in complex urban environments, it can effectively address the dynamic changes in building and tree shadows as well as cloud interference, providing mobile measurement robots with precise lighting environment information and supporting their intelligent path planning decisions under high-temperature conditions.

[0051] In another preferred embodiment of the present invention, the topological association process is as follows:

[0052] I. Path Node Layout Strategy for Continuous Shaded Areas

[0053] Within continuous shaded areas, an adaptive node spacing algorithm is used:

[0054] Geometric feature recognition: Through morphological analysis algorithms, the geometric parameters such as aspect ratio and compactness of the shaded area are identified to determine whether it is a strip (aspect ratio > 3:1), block, or irregular shape.

[0055] Dynamic deployment of master nodes:

[0056] For strip-shaped shaded areas, main nodes are placed at 3-5 meter intervals along the long axis to ensure continuous robot movement within the shade. For block-shaped shaded areas, a triangular mesh method is used, with nodes placed at the center and edges where curvature is greater to optimize path selection flexibility.

[0057] Branch connection optimization: Diagonal branches are inserted between main nodes to form a redundant path network. The branch angle is dynamically adjusted according to the curvature of the shaded area boundary to ensure that the path angle between adjacent nodes is ≥60°, avoiding the decrease in motion efficiency caused by sharp angle turns.

[0058] II. Candidate Path Construction Methods in High Sunlight Areas

[0059] For measurement points that must traverse areas with high solar radiation, a three-level candidate path system is constructed:

[0060] Shaded area boundary analysis:

[0061] Extract the feature point set of the boundary of adjacent shaded areas and calculate the rate of change of curvature of the boundary curve.

[0062] When the angle between the extension directions of the two shaded areas is within the range of 45°-135°, it is determined that the angle condition is met.

[0063] Virtual shadow corridor generation:

[0064] Create a virtual shadow corridor with a width 1.5 times that of the robot in the intersection area of ​​the boundary extension direction.

[0065] The corridor length L is dynamically adjusted according to the solar altitude angle. The calculation formula is: L=H×tan(90°-α)×K, where H is the building height, α is the solar altitude angle, and K is the safety factor (taken as 1.2-1.5).

[0066] Path effectiveness assessment:

[0067] The system calculates the cosine of the angle between the rate of change of the solar azimuth angle and the corridor direction in real time. When the cosine value increases, the effective duration of the corridor is extended. A sliding time window is used to predict the corridor failure time, and a path replanning warning is triggered 10 minutes before failure.

[0068] III. Dynamic Path Priority Adjustment Mechanism

[0069] Dual-mode path selection strategy based on solar altitude angle:

[0070] High solar mode (α≥45°):

[0071] Prioritize path segments perpendicular to the sun's azimuth angle, utilizing the "shadow walls" created by the lateral shadows of buildings for protection. Enable a fast traversal algorithm to reduce exposure time by accelerating passage through high-sunlight areas.

[0072] Low solar mode (α < 45°):

[0073] Choose a path segment that aligns with the building's projected direction to fully utilize long-distance projected shadows. Associate temporary shadow coverage time windows, accurately calculate passable time windows, and initiate the traversal action the instant shadow coverage occurs.

[0074] Path switching decision:

[0075] When the expected exposure time of a path segment exceeds the heat tolerance threshold, a path switching is triggered. An improved version of the Dijkstra algorithm is used, incorporating a temperature accumulation penalty term into the path planning. The calculation formula is: Cost = Distance × (1 + k × T), where Cost represents the total cost of the path, is a path evaluation index calculated based on the improved Dijkstra algorithm, k is the temperature influence coefficient, Distance is the length of the path segment, and T is the average temperature of the path segment.

[0076] IV. Composite Path Generation and Dynamic Maintenance

[0077] The virtual shadow corridor is topologically connected to the actual shadow area to form a composite path that spans the sunlit area:

[0078] Corridor entrance optimization: Set buffer nodes at the connection points between the virtual corridor and the actual shadow area to ensure a smooth transition.

[0079] Real-time status monitoring:

[0080] A network of light sensors deployed on the building's rooftops verifies the actual shadow effects of the corridors in real time. When the measured light intensity deviates from the predicted value by more than 20%, a reassessment of the corridor's effectiveness is triggered.

[0081] Failure response strategies:

[0082] Five minutes before the corridor fails, an alternative path plan is activated to guide the robot to turn in advance. A bidirectional search algorithm is used to simultaneously calculate the shortest path to the target point and the thermally optimal path, and dynamically select the path based on the real-time thermal status.

[0083] In another preferred embodiment of the invention, the dynamic path allocation protocol, through an efficient communication and intelligent scheduling mechanism, enables multiple mobile measurement robots to rationally allocate and utilize shaded area resources. During operation, each robot continuously sends path update messages to the central coordinator. These messages record in detail the robot's current shaded area number, task completion progress, and the target shaded area requested for the next stage, allowing the central coordinator to monitor the working status of each robot in real time. The central coordinator is responsible for maintaining a global shaded area occupancy status table, which clearly indicates which robot is currently occupying each shaded area. Simultaneously, based on the robot's travel speed and task planning, it estimates the expected release time of each shaded area, providing comprehensive data support for subsequent path allocation.

[0084] When multiple robots simultaneously request entry into the same shaded area, the central coordinator quickly retrieves the remaining task queues of each requesting robot and analyzes the number of consecutive shaded areas they need to traverse in their subsequent measurement tasks. Based on this analysis, the target shaded area is prioritized for allocation to robots with fewer consecutive shaded area connections in their subsequent paths. This reduces the likelihood of the robot frequently adjusting its route due to path conflicts, ensuring smoother progress of its measurement tasks. For robots that are not allocated a shaded area, the central coordinator promptly sends alternative path instructions. The alternative path included in these instructions is generated by the coordinator based on the real-time occupancy status of the current shaded area and by reorganizing the connection nodes of adjacent free shaded areas. It also provides precise path switching time synchronization information to ensure that the robot can seamlessly switch to the new path and continue to execute its measurement tasks efficiently.

[0085] When generating alternative path instructions, the central coordinator first searches the set of currently unoccupied shaded areas. Then, combining the requesting robot's current location and the distribution of the target points, it filters and combines these free shaded areas, attempting to find a continuous path segment that meets the robot's task progress requirements. However, if a continuous path cannot be formed in a free shaded area, the central coordinator activates a path bridging mode, turning its attention to the sunlit area. In the sunlit area, the coordinator accurately searches for transitional path segments that fit the temporary shade coverage time window, based on the relationship between the current solar altitude angle and the height of surrounding buildings. Only those road segments that meet the theoretical shade coverage standard within the specified time window are selected into the alternative path sequence, ultimately forming a composite route that integrates regular shaded area paths and temporary shaded coverage segments. This ensures that the robot can complete the measurement task while minimizing heat load, even when it cannot rely entirely on continuous shaded areas.

[0086] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A path control method for a multi-machine collaborative mobile measurement robot, characterized in that, Includes the following steps: Collect surface height information of the target area, bind the surface height information with geographic coordinates, and generate a simplified 3D model that includes building edges, tree canopies, and the tops of fixed equipment; Based on the solar trajectory data for the current season, calculate the cumulative duration of direct sunlight received at each coordinate point on the ground within the current time period; Based on the cumulative duration distribution map, high solar radiation intensity areas and continuous shadow areas are divided, and the coordinates of the target point to be measured are topologically correlated with the continuous shadow areas; A dynamic path allocation protocol is established in the multi-robot communication network, enabling each robot to select a non-overlapping measurement route in the path network formed by continuous shaded areas based on its real-time position and task queue, and to update the path occupancy status through periodic broadcasts. The process of generating the simplified 3D model is as follows: The surface height information of the target area is collected by the first robot and the second robot. The first robot moves along the center line of the road and collects the vertical height profile of the buildings on both sides. The second robot rotates and scans at a fixed point at the intersection to obtain the horizontal projection shape of the tree canopy. When using polygonal prisms for spatial fitting of building outlines, the bottom boundary of the prism is aligned with the building base in the satellite map, and the top height is corrected layer by layer based on vertical scan data. When calculating the minimum enclosing cone for the projected shape of tree canopies, the diameter of the cone's base is determined based on the maximum projection extension distance, and the vertex height is taken from the measured value of the highest point of the canopy. After superimposing the prism and cone models according to geographic coordinates, the spatial overlap area between the models is detected, the redundant volume of the overlapping part of the prism and cone is deleted, and the independent structures are retained to form a three-dimensional spatial topology with object type identification.

2. The path control method for a multi-machine collaborative mobile measurement robot according to claim 1, characterized in that, During the scanning process of the horizontal projection shape of the tree canopy, a multi-height-layer scanning strategy is adopted for densely arranged tree groups, specifically: The first layer of scanning is performed on a plane below the height of the lowest branch of the canopy to obtain the distribution position of the trunk and the outer contour of the tree group; the second layer of scanning is performed at the height of the middle layer of the canopy to capture the extension direction of the main branches and the distribution of canopy gaps; the third layer of scanning is performed at the height of the top of the canopy to record the concave and convex features of the canopy surface; a composite cone model is constructed based on the three layers of scanning data, decomposing the actual canopy into a superimposed structure of multiple sub-cones. The axial position of the sub-cones is offset according to the extension direction of the branches, and the bottom diameter is set in segments according to the maximum extension value of the contour of each layer of scanning, so that the projection range of the composite model matches the actual shadow shape.

3. The path control method for a multi-machine collaborative mobile measurement robot according to claim 1, characterized in that, The calculation process for the cumulative duration distribution map is as follows: The target area is divided into equally spaced grids, and each grid node is associated with geographic coordinate information. For each grid node, the sequence of changes in solar azimuth and elevation angles within the target time period is traversed. Based on the spatial position and shape parameters of each object in the 3D model, the time interval in which the node is covered by the projection of different objects is calculated. For the prism model of a building, parallel light rays are projected along the direction of the solar azimuth angle, and a dynamic projection trajectory is generated based on the geometric parameters of the side edge of the prism. The coverage time of the projection trajectory on the ground grid nodes is recorded. For the cone model of a tree, the ground projection area is dynamically adjusted according to the geometric relationship between the cone apex angle and the solar altitude angle. The fuzzy transition range of the projection edge is calculated by combining the canopy density coefficient. The projection coverage time of each object is superimposed and statistically analyzed according to the grid nodes to generate the cumulative duration data of each node not being covered by any projection, forming a ground illumination heat map that changes over time.

4. The path control method for a multi-machine collaborative mobile measurement robot according to claim 3, characterized in that, In the process of generating the illumination heat map, real-time meteorological data is introduced to dynamically correct the projection calculation: when cloud cover is forecasted within the target time period, the transmittance parameter is matched according to the cloud type code, and the cumulative duration data of the grid nodes is attenuated within the corresponding time interval; for sudden local cloud changes, the cloud movement speed and coverage area are obtained through the meteorological data interface, the movement trajectory of the cloud projection is simulated in the three-dimensional model space, the projection is regarded as a temporary virtual object and added to the projection calculation queue, and superimposed with the original fixed object projection for analysis, updating the cumulative duration data of the affected grid nodes, triggering the reconstruction of the local heat map and the real-time adjustment of the path network.

5. The path control method for a multi-machine collaborative mobile measurement robot according to claim 1, characterized in that, The process of the topological association is as follows: Virtual measurement path nodes are established within a continuous shaded area. The node spacing is dynamically adjusted according to the geometry of the shaded area. For strip-shaped shaded areas, main nodes are equidistantly arranged along their long axis, and diagonal connecting branches are inserted between the main nodes. For measurement points that must pass through areas with high solar radiation intensity, the distribution pattern of the surrounding shaded areas is analyzed, and multiple candidate connecting paths are constructed between adjacent shaded areas. The path priority is adjusted according to the spatial relationship between the current solar altitude angle and the direction of the candidate paths. When the solar altitude angle exceeds the set threshold, the path segment perpendicular to the solar azimuth angle direction in the connecting path is selected as the priority route; when the solar altitude angle is below the threshold, the path segment consistent with the building projection extension direction in the connecting path is selected, and a temporary shadow coverage time window is associated to dynamically switch paths.

6. The path control method for a multi-machine collaborative mobile measurement robot according to claim 5, characterized in that, The construction of the candidate connection path specifically includes: Analyze the boundary orientation of adjacent shaded areas within a high-sunlight region. When the extension trend of the two shaded area boundaries meets the angle condition, a virtual shaded corridor is created in the area where the extension directions intersect. The effective duration of the virtual shaded corridor is calculated based on the relationship between the rate of change of the solar azimuth angle and the corridor orientation. When the angle between the direction of change of the solar azimuth angle and the corridor orientation decreases, the effective duration of the corridor is extended. During path selection, the virtual shaded corridor is topologically connected with the actual shaded area to generate a composite path that crosses the sunlight region. Before the effective duration of the corridor ends, a path failure warning is sent to the robot, triggering dynamic switching of subsequent path segments.

7. The path control method for a multi-machine collaborative mobile measurement robot according to claim 1, characterized in that, When the dynamic path allocation protocol is executed, each robot sends a path update message to the central coordinator, which includes the current shadow zone number, task progress, and request for the next target shadow zone. The central coordinator maintains a global shadow zone occupancy status table, which records the current occupant robot identifier and the expected release time for each shadow zone. When multiple robots request to enter the same shaded area, the remaining task queues of each requesting robot are extracted, the number of consecutive connections in the shaded area in its subsequent path is calculated, and robots with fewer shaded area connections are given priority to enter the target area. Alternative path instructions are sent to robots that have not been assigned a shaded area. The alternative path is generated by the central coordinator by recombining the connection nodes of adjacent free shaded areas according to the current shaded area occupancy status, and includes time synchronization information for path switching.

8. The path control method for a multi-machine collaborative mobile measurement robot according to claim 7, characterized in that, The process of generating the alternative path instruction is as follows: The central coordinator searches for the set of currently unoccupied shaded areas. Based on the current position of the requesting robot and the distribution of the target points to be measured, it searches for continuous path segments in the vacant shaded areas that meet the task progress requirements. When a continuous path cannot be formed in the vacant shaded areas, the path bridging mode is activated. The system searches for transitional path segments in the sunlit area that meet the temporary shade coverage time window. The selection of transitional path segments is based on the geometric relationship between the current solar altitude angle and the height of surrounding buildings. The theoretical shade coverage rate of the transitional path segments within the specified time window is calculated. Only road segments that meet the coverage rate are selected to be added to the alternative path sequence, forming a composite passage route with mixed shade and temporary shade coverage.

Images