Mowing robot full-coverage operation system and method based on multi-source information fusion

Abstract

The invention relates to the field of mowers, and discloses a mowing robot full-coverage operation system and method based on multi-source information fusion, and the method comprises the steps: fusing the environment perception information of a mowing robot, multi-robot cooperation global sharing map information and historical operation data; through core steps of operation environment modeling, dynamic operation partitioning, multi-machine task allocation, collaborative path planning, operation execution monitoring, dynamic adjustment and the like, whole-process intelligent management and control of mowing operation is realized. Meanwhile, operation tasks are matched in combination with the operation state and capability parameters of the robot, partitions, tasks and paths are dynamically optimized according to the operation progress, the environment and the state change of the robot, and the matched system achieves automatic execution of all the processes through modular design. The adaptability and efficiency of multi-robot collaborative operation are improved, full coverage and low conflict of mowing operation are guaranteed, and the method is suitable for collaborative operation of mowing robots in various large-area and multi-obstacle scenes.

Description

Technical Field

[0001] This invention relates to the field of lawnmowers, and more specifically to a lawnmower robot full-coverage operation system and method based on multi-source information fusion. Background Technology

[0002] With the development of intelligent equipment and automation technology, lawn mowing robots are increasingly being applied to large-area outdoor operations such as municipal greening, park lawns, golf courses, and orchards. Compared to manual mowing, lawn mowing robots can reduce labor costs and improve work efficiency to a certain extent, showing promising application prospects. Currently, most lawn mowing robots operate in a standalone mode, where a single robot uses its onboard sensors to perceive the working environment and plan its mowing path accordingly. However, in situations involving large working areas or complex environments, a single lawn mowing robot has limitations in terms of operating time, coverage efficiency, and environmental adaptability, making it difficult to meet the actual needs for high-efficiency and full-coverage operations.

[0003] To improve operational efficiency, solutions involving multiple lawnmower robots working collaboratively have gradually emerged in existing technologies. These solutions typically achieve parallel operation by pre-dividing fixed work areas for multiple robots or simply distributing the task evenly among different robots. However, in practical applications, lawnmower environments often exhibit uneven obstacle distribution, significant variations in terrain complexity, and dynamic changes during the operation. Pre-defined or statically divided work areas are difficult to adapt to these changes, easily leading to issues such as repeated mowing or missed mowing in certain areas.

[0004] Furthermore, in multi-robot collaborative operation scenarios, each lawnmower robot typically relies on its own collected environmental perception information to make operational decisions. There may be differences between different information sources in terms of time, space, and perception accuracy. If there is a lack of an effective information integration mechanism, it is easy to cause inconsistencies in the basis of operational decisions, thereby affecting the coordination and stability of the overall operation.

[0005] Some existing solutions attempt to introduce path planning or task scheduling mechanisms, but they mostly focus on static path generation or simple scheduling strategies. They lack the ability to dynamically adjust according to changes in the environment, differences in work progress, and changes in the robot's operating status during the operation, making it difficult to achieve truly comprehensive, efficient, and low-conflict collaborative operations. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the present invention aims to provide a full-coverage operation system and method for lawn mowing robots based on multi-source information fusion. By fusing and processing information from the lawn mowing robot's environmental perception system, globally shared map information generated collaboratively by multiple robots, and historical operation data, the operation area is dynamically partitioned. Task allocation and collaborative path planning are implemented in conjunction with the operating status of multiple lawn mowing robots. During the lawn mowing operation, dynamic adjustments are made according to the operation progress and environmental changes, thereby achieving efficient collaboration and full-coverage execution of lawn mowing operations.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A method for full-coverage lawn mowing robot operation based on multi-source information fusion includes: The multi-source information fusion step acquires and merges multi-source information from the lawnmower robot's environmental perception system, the globally shared map information generated by multi-machine collaboration, and historical operation data to generate fused environmental information for operation decision-making; The operational environment modeling step involves constructing an operational environment model based on the fused environmental information. The dynamic task partitioning step involves dynamically dividing the task area into multiple task sub-areas based on the task environment model. The multi-machine task allocation step combines the operating status parameters and capability parameters of multiple lawnmowers to allocate the work sub-area to the corresponding lawnmower. The collaborative path planning step generates mowing paths for each mowing robot based on the results of the sub-region division of the work area and the work environment model. The operation execution and monitoring steps involve monitoring the progress of the operation and changes in the environment during the lawn mowing robot's lawn mowing operation. The dynamic adjustment process involves dynamically adjusting the division of work sub-areas, task allocation, or mowing path when abnormal work progress, environmental changes, or changes in the status of the mowing robot are detected, until full coverage mowing is completed.

[0009] In this invention, preferably, the multi-source information fusion step includes the following operations: The environmental perception information from the lawnmower robot's environmental perception system, the globally shared map information generated by multi-robot collaboration, and historical operation data are respectively sorted in chronological order and processed to correspond spatial locations. The multi-source information is then weighted and integrated according to the credibility of different information sources. To create a unified format for integrated environmental information.

[0010] In this invention, preferably, the dynamic job partitioning step includes the following operations: Based on the spatial boundaries of the work area reflected in the integrated environmental information, the work area is initially divided. Based on the initial division, and taking into account the distribution of obstacles and the accessibility of the area, the boundaries of each division area were modified to obtain multiple sub-areas suitable for independent operation of the lawnmower robot.

[0011] In this invention, preferably, during the lawn mowing operation, the actual progress of each sub-area is periodically acquired. The actual work progress is then compared with the corresponding planned work progress. When a significant schedule deviation is detected, or when a change in the working environment is detected, Recalculate the boundary positions of the corresponding sub-region. It also performs operations such as merging, splitting, or boundary adjustment on the sub-regions of the operation.

[0012] In this invention, preferably, before performing the multi-machine task allocation step, The remaining battery power, mowing width, travel speed, and historical operational stability information of each lawnmower robot were collected. The information is then collected and organized to form reference information for the capabilities of the lawnmower robot used for subsequent sub-area allocation in operations.

[0013] In this invention, preferably, the multi-machine task allocation step includes the following operations: Based on the lawnmower robot's capability reference information, the work sub-areas are sorted or grouped. The work sub-regions are then assigned to corresponding lawnmower robots according to their size and difficulty. This is to ensure that the workload of each lawnmower robot remains balanced overall.

[0014] In this invention, preferably, the cooperative path planning step includes the following operations: After each lawnmower was assigned to its corresponding work area Based on the spatial boundaries of the work sub-region and the location and shape information of obstacles in the fused environmental information. Generate mowing paths for each lawnmower robot, covering the corresponding sub-area. And when generating the path, it avoids the travel paths of adjacent lawnmower robots.

[0015] In this invention, preferably, during the mowing operation, when an obstacle is detected affecting the mowing path, or when the operation sub-area is adjusted, The affected lawnmower robot is stopped from continuing along its original mowing path, and an alternative mowing path is generated for the robot based on the updated fused environmental information. At the same time, the mowing paths of other unaffected lawnmowers remain unchanged.

[0016] In this invention, preferably, during the lawn mowing operation, the area where the lawn has been mowed is calculated based on the actual trajectory of the lawn mowing robot. The defined area is then compared with a preset sub-area for the operation. When uncovered or overlapping areas are detected, the system will trigger coordinated adjustments to the division of sub-regions, allocation of sub-regions, and mowing paths.

[0017] A lawnmower robot full-coverage operation system based on multi-source information fusion, applying any one of the above-described lawnmower robot full-coverage operation methods, includes: The multi-source information fusion module is used to acquire and fuse multi-source information from the lawnmower robot's environmental perception system, the globally shared map information generated by multi-machine collaboration, and historical operation data to generate fused environmental information; The work environment modeling module is used to construct a work environment model based on the fused environment information. The dynamic job partitioning module is used to dynamically divide the area to be worked into multiple job sub-areas based on the job environment model; The task allocation module is used to combine the operating status parameters and capability parameters of multiple lawnmowers to allocate the work sub-areas to the corresponding lawnmowers. The collaborative path planning module is used to generate mowing operation paths based on the operation sub-region and operation environment model; The operation monitoring module is used to monitor the progress of the lawn mowing operation and environmental changes. The dynamic adjustment module is used to dynamically adjust the work sub-area, task allocation, or mowing path when abnormal work progress, environmental changes, or changes in the status of the mowing robot are detected, so as to complete the full-coverage mowing operation.

[0018] The beneficial effects of this invention are: This invention features deep fusion of multi-source information, enabling more precise operational decision-making. By integrating real-time environmental perception, a globally shared map, and historical operational data, and through spatiotemporal matching and weighted integration, it overcomes the locality and lag issues of single information sources, improving the completeness and consistency of environmental information and laying a reliable data foundation for subsequent end-to-end operational decisions. Secondly, dynamic operational zoning based on the operational environment model, combined with obstacle distribution and accessibility, real-time correction of zoning boundaries solves the adaptability problem of traditional fixed zoning in multi-obstacle scenarios. This ensures a high degree of matching between sub-region division and robot operational capabilities. The zoning boundaries are corrected based on the operational environment model, obstacle distribution, and accessibility, and can be adjusted in real-time according to operational progress and environmental changes, significantly improving the adaptability of multi-robot collaborative operations in complex, multi-obstacle lawn scenarios. Thirdly, precise multi-robot task allocation achieves balanced workload. By collecting information such as remaining robot battery power, operational capabilities, and historical stability, the operational sub-regions are precisely matched with robot capabilities, avoiding the problem of some robots being idle and others overloaded due to uneven task allocation, fully leveraging the advantages of multi-robot collaborative operations. Fourth, collaborative path planning effectively reduces operational conflicts. The planner designs full-coverage paths that closely match sub-region boundaries and obstacle distribution, while also avoiding paths between adjacent robots. This reduces path intersections and repeated mowing among multiple robots, improving overall operational efficiency. Fifth, full-process monitoring and dynamic adjustment ensure full operational coverage. By monitoring operational progress, environment, and robot status in real time, precise adjustments to zones, tasks, and paths are achieved. Furthermore, relying on a linkage adjustment mechanism based on coverage assessment, issues such as missed mowing and repeated mowing are effectively avoided. This ensures stable completion of full-coverage mowing operations even when the environment changes or robot status malfunctions, thus improving the overall intelligence and efficiency of mowing operations. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall system architecture in this invention; Figure 2 This is a flowchart of the multi-source information fusion process in this invention; Figure 3 This is a schematic diagram of dynamic job partitioning and task allocation in this invention; Figure 4 This is a schematic diagram of collaborative path planning and dynamic adjustment in this invention; Figure 5 This is a flowchart of the full-coverage operation monitoring and linkage adjustment process in this invention. Detailed Implementation

[0020] 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.

[0021] It should be noted that when a component is described as "fixed to" another component, it can be directly on the other component or may have a component in between. When a component is considered "connected to" another component, it can be directly connected to the other component or may have a component in between. When a component is considered "set on" another component, it can be directly set on the other component or may have a component in between. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0022] Unless otherwise defined, 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. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0023] Please also see Figures 1 to 5 This embodiment provides a full-coverage lawn mowing robot operation system and method based on multi-source information fusion. Addressing the technical problems of low single-machine efficiency, insufficient multi-machine collaboration, and rigid work area division that are difficult to adapt to dynamic environmental changes in large-area, multi-obstacle lawn mowing scenarios, this embodiment provides a full-coverage lawn mowing robot operation method based on multi-source information fusion. This method is applicable to scenarios where multiple lawn mowing robots collaborate to perform lawn mowing operations. The core of this method is to construct a realistic work environment model through deep fusion of multi-source information, followed by dynamic zoning, precise task allocation, collaborative path planning, and dynamic monitoring and adjustment throughout the entire process. This achieves full coverage, high efficiency, and low conflict in lawn mowing operations. The entire method forms a closed-loop control system from information fusion to operation execution, and from monitoring feedback to dynamic adjustment. Each link is interconnected and data is shared. Simultaneously, the supporting operation system achieves automated execution of each process through modular design, significantly improving the intelligence level and operational adaptability of multi-machine collaborative lawn mowing.

[0024] In practical applications, this method first performs a multi-source information fusion step. The system synchronously acquires three types of core information: real-time radar and visual perception data collected by the environmental perception systems of each lawnmower robot, globally shared map information generated collaboratively by multiple robots, and historical operation data of the work area. The globally shared map information includes core content such as the overall spatial boundary of the work area and the global distribution of obstacles, while the historical operation data covers reference information such as zoning strategies, path planning schemes, and operation efficiency of past operations. The system first organizes the three types of information in terms of time sequence and accurately matches their spatial locations to ensure that the information in the same spatial area remains synchronized in the time dimension. Then, it assigns differentiated weights according to the credibility of different information sources and performs weighted integration. Generally speaking, the credibility weight of the globally shared map is the highest, followed by the single-robot environmental perception information, and the historical operation data serves as an auxiliary reference. After weighted integration, a unified, complete, and consistent fused environmental information is generated. This processing method effectively compensates for the locality of single-robot perception and the lag of historical operation data, and significantly improves the completeness and accuracy of the environmental information on which the operation decision is based. Based on the aforementioned integrated environmental information, the system continues to execute the operational environment modeling step. It systematically organizes and digitally represents the spatial boundary coordinates of the operational area, the specific location and morphological parameters of obstacles, and the accessibility levels of different areas from the integrated information. This constructs an operational environment model that includes geospatial information, obstacle information, and accessibility information. The model categorizes and labels obstacles and quantifies regional accessibility indicators. This model serves as the core basis for subsequent dynamic zoning, task allocation, and path planning, ensuring that all operational decisions align with the objective conditions of the actual operational environment.

[0025] After completing the operational environment modeling, the system enters the dynamic operational zoning step. First, it performs a preliminary division based on the overall area and shape of the area to be operated in the operational environment model. Then, it refines the boundaries of the initially divided areas by combining the distribution of obstacles and the accessibility of the areas marked in the model. During the refinement process, it avoids the obstruction of obstacles and ensures that each operational sub-area has an independent accessible path. Finally, it obtains multiple operational sub-areas suitable for independent operation of the lawnmower robot. The area and shape of the sub-areas are adapted to the conventional operational capabilities of the lawnmower robot, avoiding situations where the sub-areas are too large, causing overload of single-machine operation, or too small, causing conflicts in multi-machine operation. Compared with the traditional fixed zoning method, this dynamic zoning strategy greatly improves the adaptability of multi-machine collaborative operation in complex and multi-obstacle environments. The system then performs a multi-robot task allocation step. First, it collects the operating status and capability parameters of each mowing robot in real time, including remaining battery power, mowing width, actual travel speed, and historical operation stability information. Historical operation stability is quantified by indicators such as past operation completion rate and fault-free operation time. The system collects and organizes this information to form capability reference information for each mowing robot. Based on this information, the system sorts or groups the divided operation sub-areas according to area size and operation difficulty, and assigns different operation sub-areas to the corresponding mowing robots in sequence. During the allocation process, the capabilities of the mowing robots are matched with the operation requirements of the sub-areas, while also taking into account the balance of the workload of each robot, avoiding some robots being overloaded with tasks and some robots being idle, thereby improving the overall efficiency of multi-robot collaborative operation.

[0026] After completing the allocation of the work sub-areas, the system executes the collaborative path planning step. For each lawnmower robot assigned a work sub-area, the system combines the spatial boundary coordinates of the sub-area with the position and shape information of obstacles in the work environment model, and uses a coverage path planning algorithm to generate a lawnmower path that can cover the corresponding sub-area for each lawnmower robot. During the path planning process, a safe working distance is reserved around the obstacles, and special avoidance processing is carried out on the travel paths of adjacent lawnmower robots to avoid work conflicts caused by path intersections during the operation of multiple machines. The planned path will fit the shape and accessibility distribution of the sub-area, reduce the ineffective detours of the lawnmower robots, and effectively reduce the probability of repeated lawnmowing in multi-machine operation. Next, the operation execution and monitoring steps begin. Each lawnmower robot synchronously performs lawnmowing operations according to the planned mowing path. During the operation, the system will monitor the entire operation process in real time according to a preset cycle. On the one hand, it periodically obtains the actual operation progress of each sub-area and compares it with the preset planned operation progress in real time. On the other hand, it continuously monitors the operating status of each lawnmower robot and the dynamic changes in the operating environment, including the robot's remaining power consumption, operational malfunctions, and whether new obstacles have appeared in the operating area or whether existing obstacles have changed position. The monitoring cycle can be flexibly adjusted according to the size of the operating area and the complexity of the environment. All monitored data will be transmitted back to the system's core processing unit in real time, providing accurate and timely data support for subsequent dynamic adjustments.

[0027] When the system detects a significant deviation between the actual and planned progress of any sub-region, or detects changes in the working environment or an abnormal operation of a lawnmower, it will immediately execute dynamic adjustment steps. The system will recalculate the boundary position of the corresponding sub-region based on real-time monitoring data and merge, split, or adjust the boundaries of the sub-regions. For example, if a lawnmower has insufficient battery power to complete the assigned sub-region, the system will split the unfinished portion of that sub-region and assign it to other lawnmowers that are progressing faster and in good condition. If new obstacles appear in the working area, the system will promptly adjust the boundaries of the sub-regions to avoid the obstacle and ensure the continuity of the operation. While dynamically adjusting sub-regions, the system simultaneously performs dynamic adjustment of mowing paths. When it detects that the planned path of a mowing robot is affected by an obstacle or that its assigned work sub-region has been adjusted, the system immediately controls the mowing robot to stop moving along the original path and regenerates an alternative mowing path for it based on the updated fusion environment information. Meanwhile, other unaffected mowing robots continue to work on their original paths. This local adjustment method avoids the efficiency loss caused by replanning the entire system's path and greatly improves the timeliness and efficiency of the adjustment.

[0028] To ensure full coverage of the mowing operation, the system will continuously perform coverage assessment and linkage adjustment operations. Based on the positioning module of each mowing robot, it obtains the actual travel trajectory of the robot, counts the area that has been mowed in real time, and compares this area with the preset sub-areas. If any uncovered or overlapping areas are found, the system will immediately trigger linkage adjustments to the division and allocation of sub-areas and the mowing path. The system will re-optimize the sub-area boundaries, adjust the task allocation scheme, and plan new operation paths until all areas to be mowed are fully covered. A multi-source information fusion-based full-coverage lawn mowing robot operation system adapted to this method achieves automated execution of the entire process through modular design of a multi-source information fusion module, an operation environment modeling module, a dynamic operation zoning module, a task allocation module, a collaborative path planning module, an operation monitoring module, and a dynamic adjustment module. The modules maintain real-time data interaction. The multi-source information fusion module transmits the processed fused environment information to the operation environment modeling module. The operation environment model of the modeling module provides decision-making basis for the dynamic operation zoning module and the collaborative path planning module. The operation monitoring module feeds back real-time monitoring data to the dynamic adjustment module, and the adjustment instructions from the dynamic adjustment module are then synchronized to the corresponding zoning, allocation, and path planning modules, forming a highly efficient and collaborative operation system that effectively ensures the orderly and efficient execution of multi-machine lawn mowing operations.

[0029] In practical applications, this method allows for flexible adjustment of relevant parameters based on the area of ​​the work area, obstacle density, and the number of lawnmowers deployed. These parameters include information weighting, monitoring cycle, the scale of sub-region division, and safe distances between multiple robots, adapting to different types of large-area lawnmower scenarios. The information weighting is based on the size of the work area and the complexity of obstacles. Multiple sets of operational tests are conducted using a three-dimensional information source evaluation system that considers real-time performance, global coverage, and accuracy. In large-area, multi-obstacle scenarios, the weight of the globally shared map is increased; in small-area, less-obstacle scenarios, the weight of single-robot environmental perception information can be appropriately increased. Historical operational data is always used as a supplementary reference, and the weighting dynamically adapts to the reliability of the scenario's information source. The monitoring cycle is determined based on the number of lawnmowers deployed, the degree of dynamic interference in the work environment (including the frequency of dynamic obstacles such as pedestrians and pets), and the data processing capabilities of the cloud-based core processing unit. For scenarios with a large number of robots and frequent dynamic interference, the monitoring cycle is set to 5-10 seconds; for stable environments and fewer robots, it can be extended to 15-20 seconds, balancing the practicality of operational monitoring. The timeliness and cloud data processing load are considered. The division of the work sub-area is based on the full-charge endurance of a single lawnmower robot and the actual lawnmower operation efficiency, combined with the obstacle distribution density of the work area. The core reference is the area that a single robot can complete on a full charge. Areas with high obstacle density are divided into smaller sub-areas, while open areas with sparse obstacles can be appropriately merged into sub-areas, while ensuring that the overall workload of multiple robots in the sub-areas is balanced. The safe distance between multiple robots is set according to the actual width of the lawnmower robot body, the lawnmower width, and the operating speed. The distance is set to 0.3-0.5m for large-sized lawnmower robots and high-speed operation scenarios, and 0.2-0.3m for regular-sized robots and low-speed operation scenarios around obstacles. At the same time, a safety margin is reserved for the operation of the blade head to avoid collisions between the robot body and the blade head. By setting the parameters as described above, it can accurately adapt to different types of large-area lawn mowing operation scenarios, improve the accuracy of environmental cognition through multi-source information fusion, achieve optimal matching of multi-machine capabilities through dynamic zoning and task allocation, and reduce operation conflicts and missed mowing probability through collaborative path planning and dynamic adjustment, ultimately achieving full coverage, high efficiency and intelligence in lawn mowing operations.

[0030] This example uses a 10,000-square-meter irregular lawn in an urban park as the operational scenario. The lawn contains static obstacles such as trees, curved flower beds, and stone benches, while some areas are subject to dynamic disturbances such as temporary pedestrians and pets. Four intelligent lawnmower robots, numbered R1, R2, R3, and R4, are deployed to perform full-coverage lawnmower operations. A cloud server is used as the core processing unit of the system. The lawnmower robots and the cloud communicate in real time via 5G+WiFi6 dual-mode wireless communication with a communication latency of ≤100ms. All four lawnmower robots are equipped with 360° LiDAR, a high-definition airborne vision module, a GPS / IMU combined positioning module, and an embedded AI chip, enabling real-time environmental perception and positioning. In this example, the following core parameters are preset: the weighted weight of multi-source information is 0.6 for the global shared map, 0.3 for single-machine environmental perception, and 0.1 for historical operation data; the operation monitoring cycle is 10s; the safe distance between multiple machines is 0.3m; and the lawnmower robot's travel speed is 0.2-0.3m / s. Among them, R1 and R2 are large-sized lawnmower robots with a mowing width of 0.5m, while R3 and R4 are standard-sized lawnmower robots with a mowing width of 0.4m. Before operation, the initial status of the four robots is collected: R1 has 90% remaining battery and 98% historical operation stability; R2 has 85% remaining battery and 95% historical operation stability; R3 has 80% remaining battery and 92% historical operation stability; and R4 has 95% remaining battery and 96% historical operation stability.

[0031] The cloud system first performs a multi-source information fusion step, simultaneously acquiring real-time radar and visual perception data from four robots, a globally shared map of a 10,000-square-meter lawn generated collaboratively by multiple robots, and historical data on past lawn mowing operations in the park. After weighted integration according to preset weights, it generates unified fused environmental information including lawn spatial boundary coordinates, obstacle locations and shapes, and area accessibility. Based on this information, a work environment model is constructed, accurately marking the specific coordinates and dimensions of 28 trees, 3 curved flower beds, and 12 sets of stone benches within the lawn, while also marking accessible and temporarily restricted areas. Subsequently, the system performs a dynamic work zoning step, initially dividing the 10,000-square-meter lawn into 8 work sub-regions based on its overall irregular shape. Then, based on the obstacle distribution and accessibility status in the model, boundary correction is performed, eliminating areas containing obstacles and merging small and scattered areas, ultimately resulting in 6 work sub-regions suitable for independent single-robot operation, with sub-region areas ranging from 1200 to 2000 square meters.

[0032] In the multi-robot task allocation step, the system allocates tasks based on the capability reference information of the four robots. The two largest sub-areas are assigned to the most capable robots R1 and R4, the two medium-sized sub-areas are assigned to R2, and the two smallest sub-areas are assigned to R3, ensuring that the workload of each robot matches its capabilities and that the overall workload is balanced. After allocation, the collaborative path planning module generates full-coverage mowing paths for each of the four robots' sub-areas. The paths use a parallel bow-shaped coverage planning algorithm, reserving a 0.3m safety distance between adjacent paths of each robot and a 0.2m working distance around obstacles, effectively avoiding multi-robot conflicts and robot collisions with obstacles.

[0033] Four lawnmower robots synchronously perform lawnmower operations according to the planned path. The cloud-based operation monitoring module collects the operation progress, running status, and changes in the operating environment of each robot in real time at 10-second intervals. When the operation has been going on for 15 minutes, the monitoring module finds that a temporary obstacle, a construction fence, has appeared in the sub-area assigned to R2, and that R2's actual operation progress is 12% behind the planned progress. The system immediately triggers a dynamic adjustment step, splitting the sub-area assigned to R2 and allocating the 800-square-meter area affected by the construction fence to R4, which has an earlier operation progress. At the same time, the boundaries of the remaining sub-area are adjusted. An alternative lawnmower path is regenerated for R2 based on the updated integrated environmental information. R1 and R3 continue to operate on their original paths, while R4 performs lawnmower operations in the split area after completing its initial sub-area operation.

[0034] During the operation, the system continuously counts the mowed area based on the actual movement trajectory of the four robots and compares it with the preset operation sub-areas in real time. In the later stage of the operation, it was found that there was an uncovered area of ​​about 50 square meters at the edge of the sub-area assigned to R3. The system immediately triggered linkage adjustment, divided the missed area into a temporary sub-area, assigned it to R1 which had completed its own operation, and generated a mowing path for the temporary area for R1. R1 quickly completed the re-mowing operation in the area.

[0035] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for full-coverage lawn mowing robot operation based on multi-source information fusion, characterized in that, include: The multi-source information fusion step acquires and merges multi-source information from the lawnmower robot's environmental perception system, the globally shared map information generated by multi-machine collaboration, and historical operation data to generate fused environmental information for operation decision-making; The operational environment modeling step involves constructing an operational environment model based on the fused environmental information. The dynamic task partitioning step involves dynamically dividing the task area into multiple task sub-areas based on the task environment model. The multi-machine task allocation step combines the operating status parameters and capability parameters of multiple lawnmowers to allocate the work sub-area to the corresponding lawnmower. The collaborative path planning step generates mowing paths for each mowing robot based on the results of the sub-region division of the work area and the work environment model. The operation execution and monitoring steps involve monitoring the progress of the operation and changes in the environment during the lawn mowing robot's lawn mowing operation. The dynamic adjustment process involves dynamically adjusting the division of work sub-areas, task allocation, or mowing path when abnormal work progress, environmental changes, or changes in the status of the mowing robot are detected, until full coverage mowing is completed.

2. The method for full-coverage lawn mowing robot operation based on multi-source information fusion according to claim 1, characterized in that, The multi-source information fusion step includes the following operations: The environmental perception information from the lawnmower robot's environmental perception system, the globally shared map information generated by multi-robot collaboration, and historical operation data are respectively sorted in chronological order and processed to correspond spatial locations. The multi-source information is then weighted and integrated according to the credibility of different information sources. To create a unified format for integrated environmental information.

3. The method for full-coverage lawn mowing robot operation based on multi-source information fusion according to claim 2, characterized in that, The dynamic job partitioning step includes the following operations: Based on the spatial boundaries of the work area reflected in the integrated environmental information, the work area is initially divided. Based on the initial division, and taking into account the distribution of obstacles and the accessibility of the area, the boundaries of each division area were modified to obtain multiple sub-areas suitable for independent operation of the lawnmower robot.

4. The method for full-coverage lawn mowing robot operation based on multi-source information fusion according to claim 3, characterized in that, During the lawn mowing operation, the actual progress of each sub-area is periodically acquired. The actual work progress is then compared with the corresponding planned work progress. When a significant schedule deviation is detected, or when a change in the working environment is detected, Recalculate the boundary positions of the corresponding sub-region. It also performs operations such as merging, splitting, or boundary adjustment on the sub-regions of the operation.

5. The method for full-coverage lawn mowing robot operation based on multi-source information fusion according to claim 1, characterized in that, Before performing the multi-machine task allocation step. The remaining battery power, mowing width, travel speed, and historical operational stability information of each lawnmower robot were collected. The information is then collected and organized to form reference information for the capabilities of the lawnmower robot used for subsequent sub-area allocation in operations.

6. The method for full-coverage lawn mowing robot operation based on multi-source information fusion according to claim 5, characterized in that, The multi-machine task allocation steps include the following operations: Based on the lawnmower robot's capability reference information, the work sub-areas are sorted or grouped. The work sub-regions are then assigned to corresponding lawnmower robots according to their size and difficulty. This is to ensure that the workload of each lawnmower robot remains balanced overall.

7. The method for full-coverage lawn mowing robot operation based on multi-source information fusion according to claim 1, characterized in that, The collaborative path planning steps include the following operations: After each lawnmower was assigned to its corresponding work area Based on the spatial boundaries of the work sub-region and the location and shape information of obstacles in the fused environmental information. Generate mowing paths for each lawnmower robot, covering the corresponding sub-area. And when generating the path, it avoids the travel paths of adjacent lawnmower robots.

8. The method for full-coverage lawn mowing robot operation based on multi-source information fusion according to claim 7, characterized in that, During lawn mowing operations, if an obstacle is detected affecting the mowing path, or if the work sub-area is adjusted, The affected lawnmower robot is stopped from continuing along its original mowing path, and an alternative mowing path is generated for the robot based on the updated fused environmental information. At the same time, the mowing paths of other unaffected lawnmowers remain unchanged.

9. The method for full-coverage lawn mowing robot operation based on multi-source information fusion according to claim 8, characterized in that, During the lawn mowing operation, the area where the lawn has been mowed is calculated based on the actual trajectory of the lawn mowing robot. The defined area is then compared with a preset sub-area for the operation. When uncovered or overlapping areas are detected, the system will trigger coordinated adjustments to the division of sub-regions, allocation of sub-regions, and mowing paths.

10. A lawnmower robot full-coverage operation system based on multi-source information fusion, employing the lawnmower robot full-coverage operation method based on multi-source information fusion as described in any one of claims 1-9, characterized in that, include: The multi-source information fusion module is used to acquire and fuse multi-source information from the lawnmower robot's environmental perception system, the globally shared map information generated by multi-machine collaboration, and historical operation data to generate fused environmental information; The work environment modeling module is used to construct a work environment model based on the fused environment information. The dynamic job partitioning module is used to dynamically divide the area to be worked into multiple job sub-areas based on the job environment model; The task allocation module is used to combine the operating status parameters and capability parameters of multiple lawnmowers to allocate the work sub-areas to the corresponding lawnmowers. The collaborative path planning module is used to generate mowing operation paths based on the operation sub-region and operation environment model; The operation monitoring module is used to monitor the progress of the lawn mowing operation and environmental changes. The dynamic adjustment module is used to dynamically adjust the work sub-area, task allocation, or mowing path when abnormal work progress, environmental changes, or changes in the status of the mowing robot are detected, so as to complete the full-coverage mowing operation.

Images