A method, system, electronic device, and storage medium for multi-robot task scheduling and collaborative operation of photovoltaic cleaning robots.

Abstract

This application provides a method, system, electronic device, and storage medium for multi-robot task scheduling and collaborative operation of photovoltaic cleaning robots. It relates to the technical field of multi-robot task scheduling and collaborative operation. This application acquires the position and speed information of each robot in a robot cluster to construct a dynamic adjacency graph model. Then, it controls each robot to exchange information via a time-division multiple access protocol through a self-organizing network communication relay node, controlling communication delays and broadcasting obstacle information in real time. Next, based on relevant information, it divides the area to be cleaned, determines task sub-regions, and achieves task allocation according to robot type adaptability. Finally, during cleaning, by incorporating swarm control using an artificial potential field method, adjacent robots synchronously clean the boundaries and adjust their trajectories, enabling efficient collaborative operation of multiple photovoltaic cleaning robots and improving cleaning efficiency and flexibility.

Description

Technical Field

[0001] This application relates to the field of multi-machine task scheduling and collaborative operation technology, and in particular to a method, system, electronic device and storage medium for multi-machine task scheduling and collaborative operation of a photovoltaic cleaning robot. Background Technology

[0002] In large-scale photovoltaic power plants, numerous photovoltaic cleaning robots need to work collaboratively to ensure the cleanliness of the photovoltaic panels and improve power generation efficiency. The area and layout of photovoltaic panels vary in different areas, and various obstacles may exist on-site. Furthermore, the power reserves and cleaning capabilities of each robot are not entirely the same. Therefore, there is an urgent need for an efficient multi-robot task scheduling and collaborative operation scheme to achieve the rational allocation of cleaning tasks and orderly cooperation among robots, maximizing cleaning efficiency and energy utilization.

[0003] Currently, existing technologies for photovoltaic cleaning robots mostly involve guiding the robots to clean along pre-set fixed routes. In terms of communication, conventional communication networks are typically used, resulting in relatively independent information exchange between robots and a lack of real-time and efficient collaborative communication mechanisms. Task allocation is largely based on experience or simple average distribution methods, without fully considering the actual state of the robots and the complex environment of the site.

[0004] However, existing technologies have many drawbacks. Fixed-route cleaning cannot handle dynamic obstacles in the site, leading to incomplete cleaning or an increased risk of robot collisions. Conventional communication networks are susceptible to environmental interference, resulting in significant communication delays between robots, untimely information exchange, and difficulty in collaborative operation. Simple task allocation methods neither allow robots to flexibly adjust their tasks based on their remaining battery power and cleaning capacity, nor do they consider the actual conditions of the area to be cleaned. This results in some robots being overloaded while others are idle, leading to low overall cleaning efficiency and failing to meet the high-efficiency cleaning needs of large-scale photovoltaic power plants. Summary of the Invention

[0005] The purpose of this application is to provide a method, system, electronic device and storage medium for multi-machine task scheduling and collaborative operation of photovoltaic cleaning robots, so as to solve the problem of poor multi-machine task scheduling and collaborative operation effect in the prior art.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides a method for multi-machine task scheduling and collaborative operation of a photovoltaic cleaning robot, comprising:

[0007] Obtain the position and speed information of each photovoltaic cleaning robot in the robot cluster in the photovoltaic power station, and construct a dynamic adjacency graph model based on the position and speed information of each photovoltaic cleaning robot;

[0008] The photovoltaic cleaning robots are controlled to interact with each other through the self-organizing network communication relay node using the time division multiple access protocol, so that the communication delay between the photovoltaic cleaning robots in the robot cluster is within a preset delay range. At the same time, obstacle information is broadcast in real time in the robot cluster through the self-organizing network communication relay node.

[0009] Based on the dynamic adjacency graph model and the obstacle information, combined with the remaining power and cleaned area of ​​each photovoltaic cleaning robot, the area to be cleaned in the photovoltaic power station is divided into sections. The task sub-area of ​​each photovoltaic cleaning robot is determined by the robot cluster, so as to realize the dynamic allocation of task load according to the model adaptability.

[0010] When the photovoltaic cleaning robot detects that the distance to its neighboring photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold, it adjusts the movement speed of the photovoltaic cleaning robot and the neighboring photovoltaic cleaning robot using a swarm control method. After the two neighboring photovoltaic cleaning robots reach a common boundary line, when an obstacle is detected on the trajectory, the robot adjusts its turning angle to bypass the obstacle and sends a trajectory adjustment signal to the neighboring photovoltaic cleaning robot so that the neighboring photovoltaic cleaning robot adjusts its corresponding trajectory based on the trajectory adjustment signal.

[0011] Optionally, the step of dividing the area to be cleaned in the photovoltaic power station into sections based on the dynamic adjacency graph model and the obstacle information, combined with the remaining power and cleaned area of ​​each photovoltaic cleaning robot, and determining the task sub-region of each photovoltaic cleaning robot through the robot cluster, includes the following steps:

[0012] The remaining power, cleaned area, and energy consumption per unit area of ​​each photovoltaic cleaning robot are obtained. Based on the connection relationship of each node in the dynamic adjacency graph model, photovoltaic cleaning robots with a distance value less than a preset distance threshold are grouped into the same cooperative group.

[0013] The remaining cleaning time is calculated based on the remaining power of each photovoltaic cleaning robot in the same collaborative group, and the cleanable area of ​​each photovoltaic cleaning robot in the same collaborative group is calculated based on the energy consumption per unit area. The proportion of incomplete cleaning is determined based on the cleaned area.

[0014] The remaining area in the photovoltaic power station area to be cleaned, excluding the obstacle information, is divided into several consecutive initial area blocks, wherein the area of ​​each initial area block is not less than the maximum area that a single photovoltaic cleaning robot can clean in one go, and at least two initial area blocks are allocated to each of the same collaborative groups. Based on the cleanable area and the proportion of incomplete cleaning, sub-areas in the initial area blocks are allocated to each photovoltaic cleaning robot in the same collaborative group to form a preliminary task plan.

[0015] The preliminary task plan is sent to each photovoltaic cleaning robot in the same collaborative group, so that each photovoltaic cleaning robot can respond with an acceptance or adjustment request based on its current position and the straight-line distance between itself and the sub-area.

[0016] If any of the photovoltaic cleaning robots requests an adjustment, the sub-region is reassigned based on the acceptance or adjustment requests from all photovoltaic cleaning robots. The sub-region is preferentially assigned to the photovoltaic cleaning robot that is closer to the sub-region, until all photovoltaic cleaning robots in the same cooperative group accept the assignment result, thus forming a task sub-region.

[0017] Optionally, the remaining area within the photovoltaic power station's cleaning area, excluding the obstacle information, is divided into several consecutive initial area blocks, wherein the area of ​​each initial area block is not less than the maximum cleaning area that a single photovoltaic cleaning robot can clean in one go. At least two initial area blocks are allocated to each of the same collaborative groups. Based on the cleanable area and the proportion of incomplete cleaning, sub-regions within the initial area blocks are allocated to each photovoltaic cleaning robot within the same collaborative group to form a preliminary task plan, including the following steps:

[0018] Along the arrangement direction parallel to the photovoltaic panels in the photovoltaic power station, the remaining area is divided into several continuous strips. Along the arrangement direction perpendicular to the photovoltaic panels, each continuous strip is divided into several initial area blocks. The area of ​​each initial area block is not less than the maximum cleaning area that a single photovoltaic cleaning robot can clean in one go, and the initial area block contains an integer number of photovoltaic panels.

[0019] Based on the location information of all photovoltaic cleaning robots in each collaborative group, the straight-line distance between each collaborative group and the center of each initial area block is calculated. Based on the allocation rule that the straight-line distance between the collaborative group and the corresponding initial area block does not exceed the maximum distance between any two photovoltaic cleaning robots in the collaborative group, at least two initial area blocks that are closer to each other are preferentially allocated to the corresponding collaborative group.

[0020] The total cleanable area of ​​all photovoltaic cleaning robots in each collaborative group is counted, the ratio of the area of ​​the initial area block to the total cleanable area is calculated, and the initial area block is divided into sub-areas with the same number of photovoltaic cleaning robots in the collaborative group according to the ratio.

[0021] Based on the cleanable area of ​​each photovoltaic cleaning robot, the area of ​​each sub-region is determined so that the ratio of the area of ​​the sub-region to the cleanable area of ​​the corresponding photovoltaic cleaning robot tends to be consistent.

[0022] Based on the incomplete cleaning ratio of each photovoltaic cleaning robot, and the principle that the straight-line distance between the edge of the sub-region corresponding to the photovoltaic cleaning robot with a higher incomplete cleaning ratio and the edge of the cleaned area is smaller, the sub-region is adjusted, and the adjusted sub-region is associated with the corresponding photovoltaic cleaning robot to form a preliminary task plan.

[0023] Optionally, when the photovoltaic cleaning robot detects that the distance to its neighboring photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold, a swarm control method is used to adjust the movement speed of the photovoltaic cleaning robot and the neighboring photovoltaic cleaning robot. After the two adjacent photovoltaic cleaning robots reach a common boundary line, when an obstacle is detected on the trajectory, the turning angle is adjusted to bypass the obstacle, and a trajectory adjustment signal is sent to the neighboring photovoltaic cleaning robot so that the neighboring photovoltaic cleaning robot adjusts its corresponding trajectory based on the trajectory adjustment signal. This includes the following steps:

[0024] When the photovoltaic cleaning robot detects that the distance to its neighboring photovoltaic cleaning robot is less than or equal to a preset collaborative distance threshold, it sends a boundary synchronization request to the neighboring photovoltaic cleaning robot. Based on the boundary synchronization request, the two neighboring photovoltaic cleaning robots exchange their current direction of travel and distance information to the common boundary line, and use a swarm control method to adjust their respective movement speed so that the time difference between the two neighboring photovoltaic cleaning robots reaching the common boundary line is within a preset range.

[0025] After the two adjacent photovoltaic cleaning robots reach the common boundary line, they move synchronously in a direction parallel to the common boundary line, maintaining a preset distance during the movement. When an obstacle is detected on the movement trajectory, the robot adjusts its turning angle to bypass the obstacle based on the obstacle information and the straight-line distance between itself and the boundary of the task sub-area. At the same time, it sends a trajectory adjustment signal to the adjacent photovoltaic cleaning robot, so that the adjacent photovoltaic cleaning robot adjusts its own movement trajectory accordingly based on the trajectory adjustment signal, thereby maintaining the cooperative distance between the two adjacent photovoltaic cleaning robots within a preset distance range.

[0026] After the adjacent photovoltaic cleaning robots avoid obstacles, they resume the original cleaning path or replan their travel route to the next boundary segment based on the boundary information of the task sub-area updated in real time, until the cleaning operation of the common boundary line is completed.

[0027] Optionally, the robot moves synchronously in a direction parallel to the common boundary line, maintaining a preset distance during movement. When an obstacle is detected on the trajectory, the robot adjusts its turning angle to avoid the obstacle based on the obstacle information and the straight-line distance between the photovoltaic cleaning robot and the boundary of the task sub-area. This includes the following steps:

[0028] The vertical distance information between the two adjacent photovoltaic cleaning robots and the common boundary line is collected. The vertical distance information is exchanged through the self-organizing network communication relay node. The travel direction of the two adjacent photovoltaic cleaning robots is adjusted so that the travel direction is parallel to the common boundary line and the distance difference between them and the common boundary line is within a preset safe distance.

[0029] Calculate the straight-line distance between two adjacent photovoltaic cleaning robots. When the straight-line distance between two adjacent photovoltaic cleaning robots is greater than a preset distance, increase the movement speed of the photovoltaic cleaning robot closer to the inside of the common boundary and decrease the movement speed of the photovoltaic cleaning robot on the outside, so that the straight-line distance is within the preset distance.

[0030] When an obstacle is detected in the travel trajectory in front of the two adjacent photovoltaic cleaning robots, the coordinate information and lateral width of the obstacle are recorded. Based on the coordinate information of the obstacle, the straight-line distance between the photovoltaic cleaning robot itself and the boundary of the task sub-area is calculated.

[0031] When the straight-line distance between the photovoltaic cleaning robot and the boundary of the task sub-area is greater than the lateral width of the obstacle, the robot adjusts its turning angle away from the common boundary line based on the principle that the turning angle is proportional to the lateral width. Alternatively, when the straight-line distance between the photovoltaic cleaning robot and the boundary of the task sub-area is less than or equal to the lateral width, the robot adjusts its turning angle towards the common boundary line and generates a new travel trajectory. The minimum distance between the new travel trajectory and the common boundary line is not less than a preset safety distance.

[0032] Optionally, the step of constructing a dynamic adjacency graph model based on the position and speed information of each photovoltaic cleaning robot includes the following steps:

[0033] Based on the movement speed and direction in the location and speed information, the straight-line distance between every two photovoltaic cleaning robots in the robot cluster is calculated.

[0034] Using the location point corresponding to the coordinate information of each photovoltaic cleaning robot as a node, when the straight-line distance between the two photovoltaic cleaning robots is within the preset adjacency determination range and the angle between the movement direction and the boundary line of their respective current task sub-regions meets the preset angle condition, a connecting edge is established between the location points of the two photovoltaic cleaning robots, the edge weight is determined based on the movement speed, and a dynamic adjacency graph model is constructed.

[0035] Optionally, the control of each photovoltaic cleaning robot to interact with each other via a self-organizing network communication relay node using a time-division multiple access protocol, so that the communication delay between each photovoltaic cleaning robot in the robot cluster is within a preset delay range, and the obstacle information is broadcast in real time in the robot cluster via the self-organizing network communication relay node, including the following steps:

[0036] According to the preset time segment division rules, each photovoltaic cleaning robot in the robot cluster is assigned a dedicated information sending time segment. The duration of the information sending time segment is set according to the amount of data of location information, speed information and task status information required for a single information interaction. The information to be interacted includes obstacle information, coordinate information of the collection location and time identifier of the collection time.

[0037] Each photovoltaic cleaning robot is controlled to send the information to be interacted to the self-organizing network communication relay node within its own dedicated information sending time segment. The self-organizing network communication relay node adds a corresponding forwarding sequence number to each of the information to be interacted according to the receiving order, and broadcasts the information to be interacted to all other photovoltaic cleaning robots.

[0038] After receiving a delay anomaly signal sent by any other photovoltaic cleaning robot under preset conditions, the dedicated time segments corresponding to the two photovoltaic cleaning robots with the delay anomaly are readjusted, and the interval between the dedicated time segments is shortened. The preset condition is that the total time difference exceeds the preset delay range. The total time difference is the time difference between when the information to be interacted is sent from itself to when it is received by the target photovoltaic cleaning robot.

[0039] Secondly, this application provides a multi-machine task scheduling and collaborative operation system for photovoltaic cleaning robots, including:

[0040] The module is used to obtain the position and speed information of each photovoltaic cleaning robot in the robot cluster in the photovoltaic power station, and to construct a dynamic adjacency graph model based on the position and speed information of each photovoltaic cleaning robot.

[0041] The control module is used to control each photovoltaic cleaning robot to interact with each other through the self-organizing network communication relay node using the time division multiple access protocol, so that the communication delay between each photovoltaic cleaning robot in the robot cluster is within a preset delay range, and at the same time, the obstacle information is broadcast in real time in the robot cluster through the self-organizing network communication relay node.

[0042] The partitioning module is used to divide the area to be cleaned in the photovoltaic power station into sections based on the dynamic adjacency graph model and the obstacle information, combined with the remaining power and cleaned area of ​​each photovoltaic cleaning robot. The robot cluster determines the task sub-region of each photovoltaic cleaning robot so as to realize the dynamic allocation of task load according to the model adaptability.

[0043] The adjustment module is used to adjust the movement speed of the photovoltaic cleaning robot and the adjacent photovoltaic cleaning robot by means of swarm control when the photovoltaic cleaning robot detects that the distance between itself and the adjacent photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold. After the two adjacent photovoltaic cleaning robots reach the common boundary line, when an obstacle is detected on the trajectory, the turning angle is adjusted to bypass the obstacle. At the same time, a trajectory adjustment signal is sent to the adjacent photovoltaic cleaning robot so that the adjacent photovoltaic cleaning robot adjusts its corresponding trajectory according to the trajectory adjustment signal.

[0044] Thirdly, this application provides an electronic device, comprising:

[0045] Memory, used to store computer programs;

[0046] A processor is configured to execute the computer program to implement the steps of a multi-machine task scheduling and collaborative operation method for a photovoltaic cleaning robot as described in the first aspect above.

[0047] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of a multi-machine task scheduling and collaborative operation method for a photovoltaic cleaning robot as described in the first aspect above.

[0048] This application provides a method for multi-robot task scheduling and collaborative operation of photovoltaic cleaning robots. It acquires the position and speed information of each photovoltaic cleaning robot in a robot cluster within a photovoltaic power station, and constructs a dynamic adjacency graph model based on this information. The method controls the photovoltaic cleaning robots to interact via a self-organizing network communication relay node using a time-division multiple access protocol, ensuring that the communication delay between the robots in the cluster remains within a preset range. Simultaneously, obstacle information is broadcast in real-time within the robot cluster via the self-organizing network communication relay node. Based on the dynamic adjacency graph model and the obstacle information, combined with the remaining battery power and cleaned area of ​​each photovoltaic cleaning robot, the method then... The area to be cleaned in the photovoltaic power station is divided into sections. The task sub-area of ​​each photovoltaic cleaning robot is determined by the robot cluster to achieve dynamic allocation of task load according to model adaptability. When a photovoltaic cleaning robot detects that the distance to an adjacent photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold, a swarm control method is used to adjust the movement speed of the photovoltaic cleaning robot and the adjacent photovoltaic cleaning robot. After the two adjacent photovoltaic cleaning robots reach a common boundary line, when an obstacle is detected on the trajectory, the turning angle is adjusted to bypass the obstacle. At the same time, a trajectory adjustment signal is sent to the adjacent photovoltaic cleaning robot so that the adjacent photovoltaic cleaning robot adjusts its corresponding trajectory according to the trajectory adjustment signal.

[0049] The technical solution of this application has the following beneficial effects:

[0050] This application integrates the position and speed information of various photovoltaic cleaning robots to clearly present the positional relationships and motion states between robots, providing a basic basis for subsequent task allocation and collaborative operations, and facilitating the rapid identification of collaborative robot combinations. It employs a time-division multiple access protocol with the help of ad hoc network communication relay nodes to achieve information exchange, effectively avoiding information transmission conflicts, controlling communication delays within a preset range, and ensuring timely information transmission. Simultaneously, it broadcasts obstacle information in real time, allowing each robot to be aware of obstacle conditions in advance, providing a reference for path planning. Combining a dynamic adjacency graph model, obstacle information, and the robot's remaining battery power and cleaned area, the area is rationally divided and task sub-regions are allocated, achieving dynamic allocation of task load according to robot model adaptability, ensuring that the workload of each robot matches its capabilities, and improving overall cleaning efficiency. Using a swarm control method incorporating artificial potential fields, synchronous cleaning of the area boundary is triggered when adjacent robots enter a preset collaborative distance range, ensuring the consistency of boundary area cleaning. By adjusting the travel trajectory through real-time updated task sub-region boundary information, it can flexibly respond to environmental changes and ensure continuous cleaning operations.

[0051] Furthermore, this application obtains the remaining power, cleaned area, and energy consumption per unit area of ​​each photovoltaic cleaning robot. Based on the connection relationship of nodes in the dynamic adjacency graph model, robots with a distance less than a preset threshold are grouped into the same collaborative group. Then, the cleanable area is calculated based on the remaining power and energy consumption per unit area of ​​the robots in the collaborative group, and the proportion of incomplete cleaning is determined in combination with the cleaned area. Subsequently, the area to be cleaned after removing obstacles is divided into initial area blocks and allocated to each collaborative group. Based on the cleanable area and the proportion of incomplete cleaning, sub-regions are allocated to the robots in the group to form a preliminary task plan. Finally, the sub-regions are redistributed based on the robots' feedback on the preliminary plan until all robots accept it, forming the task sub-regions.

[0052] This application establishes a collaborative group, accurately calculates the cleaning capacity by combining parameters such as robot energy consumption, power consumption, and cleaning progress, scientifically divides the area and allocates sub-areas, and optimizes task allocation through a feedback adjustment mechanism to ensure that the task sub-areas are compatible with the robot's position and capabilities. This achieves a reasonable allocation of task load within the collaborative group and improves the coordination and efficiency of area cleaning.

[0053] Furthermore, when the photovoltaic cleaning robot detects that the distance to its neighboring photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold, it uses a swarm control method to adjust the movement speed of the photovoltaic cleaning robot and its neighboring photovoltaic cleaning robots. After the two adjacent photovoltaic cleaning robots reach a common boundary line, they move synchronously in a direction parallel to the common boundary line while maintaining a preset distance. When an obstacle is detected during the movement, the robot adjusts its turning angle to avoid the obstacle and simultaneously sends a trajectory adjustment signal to the neighboring photovoltaic cleaning robot to maintain the cooperative distance between the two adjacent photovoltaic cleaning robots within the preset distance range. This application, by adjusting the movement speed through swarm control, can avoid collisions between adjacent robots due to excessive proximity, laying a safe foundation for subsequent cooperative actions. Moving synchronously in a parallel direction with a preset distance after reaching the common boundary line allows the robots to maintain a neat formation, ensuring an orderly cleaning path, reducing overlap or omissions in the cleaning area, and improving cleaning coverage. The synchronous movement mode can ensure that the cleaning rhythm of adjacent robots in the same area is consistent, avoiding a decrease in cleaning efficiency due to disjointed actions (such as repeatedly cleaning a certain area or creating cleaning blank areas). When encountering obstacles, the robot not only flexibly avoids them but also synchronizes with adjacent robots via trajectory adjustment signals. This ensures that a preset cooperative distance is maintained during obstacle avoidance, preventing individual robots from disrupting the overall formation, guaranteeing the continuity of cleaning work, and reducing cleaning interruptions or missed areas caused by obstacles. From distance control and formation maintenance to cooperative obstacle avoidance, the entire collaborative mechanism reduces coordination losses between robots, enabling multi-robot teams to efficiently cover photovoltaic areas and improve overall cleaning efficiency.

[0054] These or other aspects of this application will become more apparent from the description of the following embodiments. Attached Figure Description

[0055] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 A flowchart illustrating a method for multi-robot task scheduling and collaborative operation of a photovoltaic cleaning robot, provided in an embodiment of this application;

[0057] Figure 2 A scenario diagram illustrating a method for multi-robot task scheduling and collaborative operation of a photovoltaic cleaning robot, provided in an embodiment of this application;

[0058] Figure 3 This is a schematic diagram illustrating a specific implementation of a photovoltaic cleaning robot multi-machine task scheduling and collaborative operation system provided in this application embodiment;

[0059] Figure 4 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application. Detailed Implementation

[0060] To address the poor performance of multi-robot task scheduling and collaborative operation in existing technologies, this application provides a method for multi-robot task scheduling and collaborative operation of photovoltaic cleaning robots. The method's design concept is as follows: by integrating the position and speed information of each photovoltaic cleaning robot, the positional relationships and motion states between robots are clearly presented, providing a basic basis for subsequent task allocation and collaborative operation, facilitating rapid identification of robot combinations capable of collaboration; employing a time-division multiple access protocol with the help of ad hoc network communication relay nodes to achieve information exchange, effectively avoiding information transmission conflicts, controlling communication delays within a preset range, and ensuring timely information transmission; simultaneously broadcasting obstacle information in real time. This system allows each robot to be aware of obstacles in advance, providing a reference for path planning. By combining a dynamic adjacency graph model, obstacle information, and the robot's remaining battery power and cleaned area, the area is rationally divided and task sub-regions are allocated. This achieves dynamic allocation of task load according to robot model adaptability, ensuring that the workload of each robot matches its capabilities and improving overall cleaning efficiency. Using a swarm control method that integrates artificial potential fields, synchronous cleaning of the area boundary is triggered when adjacent robots enter the preset cooperative distance range, ensuring the consistency of cleaning in the boundary area. By adjusting the travel trajectory through real-time updated task sub-region boundary information, it can flexibly respond to environmental changes and ensure continuous cleaning operations.

[0061] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0062] The core of this application is to provide a method for multi-machine task scheduling and collaborative operation of photovoltaic cleaning robots, and a flowchart of one specific implementation is shown below. Figure 1 As shown, the method includes:

[0063] S101. Obtain the position and speed information of each photovoltaic cleaning robot in the photovoltaic power station, and construct a dynamic adjacency graph model based on the position and speed information of each photovoltaic cleaning robot.

[0064] Optionally, the step S101, which involves constructing a dynamic adjacency graph model based on the position and speed information of each photovoltaic cleaning robot, includes:

[0065] Step 1011: Based on the movement speed and direction in the position information and speed information, calculate the straight-line distance between every two photovoltaic cleaning robots in the robot cluster.

[0066] Step 1012: Using the position point corresponding to the coordinate information of each photovoltaic cleaning robot as a node, when the straight-line distance between the two photovoltaic cleaning robots is within the preset adjacency determination range and the angle between the movement direction and the boundary line of their respective current task sub-regions meets the preset angle condition, a connecting edge is established between the position points of the two photovoltaic cleaning robots, the edge weight is determined based on the movement speed, and a dynamic adjacency graph model is constructed.

[0067] In the above scheme, location information refers to the coordinate data collected by the photovoltaic cleaning robot through its onboard positioning module, which identifies its specific location in the photovoltaic power station, including horizontal and vertical coordinates; speed information refers to the motion state data collected by the robot through motion detection components, including movement speed and movement direction; the dynamic adjacency graph model is a graphical model used to describe the positional associations of each robot in the robot cluster; the preset adjacency determination range is a pre-set distance threshold used to determine whether two robots are within a spatial range where an association can be established; the boundary line of the current task sub-region refers to the edge boundary of the cleaning area assigned to each robot; and the preset angle condition is a pre-set angle threshold used to determine whether the robot's movement direction meets the requirement of forming an association with the boundary line of its own task sub-region.

[0068] In this application example, step 1011 first calculates the straight-line distance between every two photovoltaic cleaning robots in the robot cluster based on the movement speed and direction information from the position and speed information. Specifically, the coordinates of the two robots are extracted from the position information collected by each robot and denoted as follows: and The formula for calculating straight-line distance is used. The calculation is performed, where d is the straight-line distance between the two robots. Let x be the horizontal and vertical coordinates of the first robot. Let be the horizontal and vertical coordinates of the second robot; simultaneously, record the movement speeds of both robots (denoted as ). and ) and the angle between the direction of movement and the boundary line of each task sub-region (denoted as ) and ).

[0069] Next, in step 1012, the straight-line distance d is compared with the preset adjacency determination range. If d is less than or equal to the preset adjacency determination range, the included angle between the two robots is further checked. and Whether both angles are less than or equal to a preset angle condition, if both conditions are met simultaneously, a connection edge is established between the two robot positions (nodes), based on the movement speed. and Determine the weight of the connection edge, for example, by taking the average of the two rates. As edge weights, by repeating the above operation on all pairwise combinations in the robot cluster, a dynamic adjacency graph model containing nodes, connecting edges, and edge weights is finally formed.

[0070] In practical applications, the robot cluster in a photovoltaic power station consists of three robots: R1, R2, and R3. Robot R1 is located at coordinates (20, 30), moves at a speed of 0.4 m / s, and its direction of movement forms a 25-degree angle with the boundary of its task sub-region. Robot R2 is located at coordinates (23, 32), moves at a speed of 0.5 m / s, and its direction of movement forms a 28-degree angle with the boundary of its task sub-region. Robot R3 is located at coordinates (30, 35), moves at a speed of 0.3 m / s, and its direction of movement forms a 35-degree angle with the boundary of its task sub-region. The preset adjacency determination range is 6 meters, and the preset angle condition is 30 degrees. First, the straight-line distance between any two robots is calculated: the distance between robots R1 and R2, substituted into the formula... Meters (less than 6 meters); the distance between robots R1 and R3 Meters (greater than 6 meters); the distance between robots R2 and R3 Meters (greater than 6 meters). Next, the angle conditions are checked: the included angle of robot R1 (25 degrees) and the included angle of robot R2 (28 degrees) are both less than 30 degrees, thus satisfying the condition; among robots R1 and R3, and R2 and R3, at least one of the included angles does not satisfy the condition (e.g., R3's 35 degrees is greater than 30 degrees). Therefore, only connections are established between the corresponding nodes of R1 and R2, with the edge weight being the average of their speeds, i.e. The final dynamic adjacency graph model contains three nodes: R1, R2, and R3. There is only one connecting edge with a weight of 0.45 between R1 and R2.

[0071] The aforementioned S101 overall solution, by accurately collecting the robot's position and speed information, combined with quantitative distance calculation, angle judgment, and weight setting, constructs a dynamic adjacency graph model that can reflect the spatial association of the robot cluster in real time. The accurate calculation of straight-line distance ensures the accuracy of adjacency range judgment, the introduction of angle conditions avoids associations without cooperative significance, and the setting of edge weights provides a quantitative basis for subsequent task allocation. Overall, it improves the accuracy of the model's description of the robot cluster state and lays a reliable foundation for multi-machine scheduling and collaborative operation.

[0072] S102. Control each photovoltaic cleaning robot to interact with each other through the self-organizing network communication relay node using the time division multiple access protocol, so that the communication delay between each photovoltaic cleaning robot in the robot cluster is within the preset delay range, and at the same time, broadcast obstacle information in the robot cluster in real time through the self-organizing network communication relay node.

[0073] Optionally, in S102, controlling each photovoltaic cleaning robot to interact with information via a time-division multiple access protocol through an ad hoc network communication relay node, so that the communication delay between the photovoltaic cleaning robots in the robot cluster is within a preset delay range, and simultaneously broadcasting obstacle information in real time within the robot cluster through the ad hoc network communication relay node, including:

[0074] Step 1021: According to the preset time segment division rules, assign a dedicated information transmission time segment to each photovoltaic cleaning robot in the robot cluster. The duration of the information transmission time segment is set according to the amount of data of location information, speed information and task status information required for a single information interaction. The information to be interacted includes obstacle information, coordinate information of the collection location and time identifier of the collection time.

[0075] Step 1022: Control each photovoltaic cleaning robot to send the information to be interacted to the self-organizing network communication relay node within its own dedicated information sending time segment, so that the self-organizing network communication relay node adds a corresponding forwarding sequence number to each of the information to be interacted according to the receiving order, and broadcasts the information to be interacted to all other photovoltaic cleaning robots.

[0076] Step 1023: After receiving a delay anomaly signal sent by any other photovoltaic cleaning robot under preset conditions, readjust the exclusive time segments corresponding to the two photovoltaic cleaning robots corresponding to the delay anomaly, and shorten the interval between the exclusive time segments. The preset condition is that the total time difference exceeds the preset delay range. The total time difference is the time difference between when the information to be interacted is sent from itself to when it is received by the target photovoltaic cleaning robot.

[0077] In the above scheme, the self-organizing network communication relay node refers to a device that enables multiple photovoltaic cleaning robots to connect to each other to form a network and forward information, thereby realizing information transmission between robots; the time division multiple access protocol refers to the rule of dividing time into segments, allowing different robots to send information within their respective exclusive time segments, avoiding information transmission conflicts; communication delay refers to the time it takes for information to be sent from one robot to another; the preset delay range refers to the pre-set maximum acceptable communication delay; obstacle information refers to the relevant data of objects detected by the robot that obstruct its movement; task status information refers to data such as the progress of the robot's current cleaning task; the forwarding sequence number refers to the number added sequentially by the self-organizing network communication relay node to the received information, used to identify the order in which the information is received; the delay anomaly signal refers to the prompt signal issued by the robot when the communication delay exceeds the preset delay range.

[0078] In this application example, firstly, step 1021 allocates a dedicated information transmission time segment for each robot. Based on a preset time segment division rule, the amount of data (position information, speed information, and task status information) required for a single information interaction is calculated. It is assumed that the time required per unit data volume is... The duration of a single time segment It can be done through the formula The calculation is performed, where k is the redundancy coefficient (ranging from 1.1 to 1.3), and D in the formula is the amount of data in a single information exchange. Based on the calculated duration of each time segment, the total time is sequentially allocated to each photovoltaic cleaning robot in the robot cluster, forming their own exclusive information transmission time segments.

[0079] Next, in step 1022, information is sent and broadcast. Each robot, within its own designated information sending time slot, sends the information to be interacted with (including obstacle information, coordinates of the collection location, and a timestamp of the collection time) to the ad hoc network communication relay node. The ad hoc network communication relay node assigns a forwarding sequence number to each piece of information according to the order of receipt, starting from 1 and incrementing sequentially. The relay node then broadcasts the information with the forwarding sequence number to all other photovoltaic cleaning robots, ensuring that each robot receives the complete information during the broadcast process.

[0080] Finally, by adjusting the time segment in step 1023 to control communication delay, each robot records the time of information transmission after receiving information from other robots. and receiving time Through formula Calculate the time difference over the entire journey. .like Exceeding the preset delay range The robot then sends a delay anomaly signal to the information aggregation node. This signal contains the identifiers of the sender and receiver, as well as... The specific value. After receiving the signal, the information aggregation node, according to... and The difference Calculate the time interval that needs to be shortened. , Values The interval between the two robots is reduced to 1.2-1.5 times the original value, and the dedicated time segments for the two robots are readjusted accordingly to reduce the time difference between them. To reduce communication latency.

[0081] In practical applications, the number of relay nodes can be one, two, or other values. For example... Figure 2 As shown, this application sets up two relay nodes: Relay Node 1 and Relay Node 2. The photovoltaic power station has three photovoltaic cleaning robots, R1, R2, and R3. The self-organizing network communication relay nodes are responsible for information forwarding, with a preset delay range of 0.5 seconds and a time required per unit data volume. =0.001 seconds / unit of data, redundancy coefficient k=1.2, data volume of a single information exchange D=100 units of data, through the formula Each time segment is calculated to be 0.12 seconds. Based on this duration, R1 is allocated 0-0.12 seconds, R2 0.12-0.24 seconds, and R3 0.24-0.36 seconds as dedicated information transmission time segments. Within 0-0.12 seconds, robot R1 sends detected obstacle information (e.g., an obstacle at coordinates (10, 20), collected at 10:00:00) to relay node 1. Relay node 1 receives the information, adds a forwarding sequence number 1, and can directly broadcast it to R2, and can also broadcast it to R3 through relay node 2. Within 0.12-0.24 seconds, robot R2 sends its own position information, which is broadcast by relay node 1 with a forwarding sequence number 2. Robot R3 sends information similarly, and relay node 2 adds a forwarding sequence number 3. When robot R2 receives information from R1, it records the sending time as 10:00:00 and the receiving time as 10:00:00.6, and uses the formula... The time difference was calculated to be 0.6 seconds, exceeding the preset delay range of 0.5 seconds. Seconds, take If the delay exceeds 1 second, R2 sends a delay error signal. After receiving the signal, the time segments of R21 and R2 are adjusted. R1 is set to 0-0.12 seconds, and R2 is set to 0.12-0.13=-0.01 (adjusted to 0.01)-0.01+0.12=0.13 seconds, thus shortening the interval between the two to reduce communication delay.

[0082] The above-mentioned S102 overall solution effectively avoids information transmission conflicts and ensures the orderly transmission of information by allocating dedicated information transmission time segments to each robot and using a time-division multiple access protocol for information interaction. Information is forwarded and broadcast through self-organizing network communication relay nodes, enabling each robot to obtain the necessary information, including obstacle information, in a timely manner. When communication delays occur abnormally, the communication delay can be controlled within a preset range by adjusting the time segment interval, ensuring the timeliness and stability of robot cluster information interaction and providing a solid communication foundation for multi-robot task scheduling and collaborative operations.

[0083] S103. Based on the dynamic adjacency graph model and the obstacle information, and combined with the remaining power and cleaned area of ​​each photovoltaic cleaning robot, the area to be cleaned in the photovoltaic power station is divided into sections. The task sub-region of each photovoltaic cleaning robot is determined by the robot cluster to realize the dynamic allocation of task load according to the model adaptability.

[0084] Optionally, in S103, based on the dynamic adjacency graph model and the obstacle information, and combining the remaining power and cleaned area of ​​each photovoltaic cleaning robot, the area to be cleaned in the photovoltaic power station is divided into sections, and the task sub-region of each photovoltaic cleaning robot is determined by the robot cluster, including:

[0085] Step 1031: Obtain the remaining power, cleaned area, and energy consumption per unit area of ​​each photovoltaic cleaning robot. Based on the connection relationship of each node in the dynamic adjacency graph model, group the photovoltaic cleaning robots with a distance value less than a preset distance threshold into the same cooperative group.

[0086] Step 1032: Calculate the remaining cleaning time based on the remaining power of each photovoltaic cleaning robot in the same collaborative group, and calculate the cleanable area of ​​each photovoltaic cleaning robot in the same collaborative group based on the energy consumption per unit area. Determine the proportion of incomplete cleaning based on the cleaned area.

[0087] Step 1033: Divide the remaining area in the photovoltaic power station area to be cleaned, excluding the obstacle information, into several consecutive initial area blocks, wherein the area of ​​each initial area block is not less than the maximum area that a single photovoltaic cleaning robot can clean in one go, and allocate at least two initial area blocks to each of the same collaborative groups. Based on the cleanable area and the proportion of incomplete cleaning, allocate sub-areas in the initial area blocks to each photovoltaic cleaning robot in the same collaborative group to form a preliminary task plan.

[0088] Step 1033 may specifically include the following process: Dividing the remaining area into several continuous strips along the arrangement direction parallel to the photovoltaic panels in the photovoltaic power station; dividing each continuous strip into several initial area blocks along the arrangement direction perpendicular to the photovoltaic panels; the area of ​​each initial area block is not less than the maximum cleaning area that a single photovoltaic cleaning robot can clean in one pass, and the initial area block contains an integer number of photovoltaic panels; calculating the straight-line distance between each cooperative group and the center of each initial area block based on the position information of all photovoltaic cleaning robots in each cooperative group; and prioritizing the allocation rule that the straight-line distance between the cooperative group and the corresponding initial area block does not exceed the maximum distance between any two photovoltaic cleaning robots in the cooperative group, assigning at least two initial area blocks that are closer together to the corresponding cooperative group; and statistically analyzing each... The total cleanable area of ​​all photovoltaic cleaning robots within the collaborative group is used to calculate the ratio of the area of ​​the initial region block to the total cleanable area. The initial region block is then divided into sub-regions equal to the number of photovoltaic cleaning robots within the collaborative group, according to this ratio. The area of ​​each sub-region is determined based on the cleanable area of ​​each photovoltaic cleaning robot, ensuring that the ratio of the sub-region's area to the corresponding photovoltaic cleaning robot's cleanable area is consistent. Considering the incomplete cleaning ratio of each photovoltaic cleaning robot, and based on the principle that the straight-line distance between the edge of the sub-region corresponding to a photovoltaic cleaning robot with a higher incomplete cleaning ratio and the edge of the cleaned area is smaller, the sub-regions are adjusted. The adjusted sub-regions are then associated with their corresponding photovoltaic cleaning robots to form a preliminary task plan.

[0089] Step 1034: Send the preliminary task plan to each photovoltaic cleaning robot in the same collaborative group, so that each photovoltaic cleaning robot can respond with an acceptance or adjustment request based on its current position and the straight-line distance between itself and the sub-area.

[0090] Step 1035: If any of the photovoltaic cleaning robots submits an adjustment request, the sub-region is reallocated based on the acceptance or adjustment requests submitted by all photovoltaic cleaning robots. The sub-region is preferentially assigned to the photovoltaic cleaning robot that is closer to the sub-region, until all photovoltaic cleaning robots in the same cooperative group accept the allocation result, thus forming a task sub-region.

[0091] In the above scheme, the dynamic adjacency graph model refers to a graphical model used to reflect the positional association of each photovoltaic cleaning robot in the robot cluster; obstacle information refers to the data such as the position and size of objects that the robot detects that obstruct its movement; model adaptability refers to assigning suitable tasks according to the robot's performance parameters; a cooperative group is a group of robots that are close to each other, which facilitates collaborative operation; the preset distance threshold is the distance standard for judging whether a robot can be classified into the same cooperative group; the continuous strip is a long strip-shaped area divided along the direction parallel to the arrangement of photovoltaic panels; the sub-region is the area assigned to a single robot after the initial area block is further divided; and the task sub-region is the cleaning area finally determined and assigned to each robot.

[0092] In this application example, firstly, a collaborative group is formed through step 1031 to obtain the remaining power, cleaned area, and energy consumption per unit area of ​​each photovoltaic cleaning robot. Based on the connection relationship of each node in the dynamic adjacency graph model, the position coordinates of any two robots are extracted, and then the formula is used to... (where d is the distance between the two robots,) , Calculate the distance between the two robots (their position coordinates are given), and group the robots whose distance values ​​are less than a preset distance threshold into the same collaborative group to form a collaborative group list.

[0093] Next, step 1032 calculates the cleanable area and the percentage of incomplete cleaning. For each robot in the same collaborative group, based on the remaining battery power E, energy consumption per unit area e, and the area cleaned per unit time s, the formula is used to calculate the area cleaned. Calculate the remaining cleaning time, where t is the remaining cleaning time; then use the formula... The cleanable area is obtained, where A is the cleanable area. Simultaneously, based on the already cleaned area 'a' and the total task area... (The total area of ​​the area the collaborative group is responsible for), through the formula Determine the percentage of areas not cleaned, and record the corresponding cleanable area and percentage of areas not cleaned for each robot, where r is the percentage of areas not cleaned.

[0094] Then, in step 1033, the area is divided and a preliminary task plan is formed. First, the remaining area to be cleaned, after removing obstacle information, is divided into continuous strips along the direction parallel to the arrangement of the photovoltaic panels. The strip width is determined based on the width of the photovoltaic panels and the robot's cleaning width. Then, the strips are divided into initial area blocks along the vertical direction. The area of ​​each initial area block is determined by the formula. (in Let n be the initial area of ​​the region block, and n be the number of photovoltaic panels contained therein. Calculate the area of ​​a single photovoltaic panel, ensuring that n is an integer and The total cleanable area of ​​all robots in a single sweep is calculated, ensuring it is not less than the maximum cleanable area of ​​a single robot. The straight-line distance between the center of each collaborative group and the center of each initial area block is determined. Based on the rule that the closest distance does not exceed the maximum distance between any two robots within the collaborative group, at least two initial area blocks are allocated to each collaborative group. The total cleanable area of ​​all robots within the collaborative group is then calculated. ,(in (where i is the cleanable area of ​​the i-th robot in the collaborative group), calculate the area of ​​each initial region block. and proportion The initial block is divided into sub-regions proportional to the number of robots, with each sub-region having an area of... ,make Towards consistency; based on this, combined with the proportion of incomplete cleaning. Following the rule that "the straight-line distance between the edge of the sub-area corresponding to the photovoltaic cleaning robot with a higher proportion of incomplete cleaning and the edge of the cleaned area is smaller," the results were compared among the various robots. Value, if a certain robot's The value is greater than that of at least one other robot in the group. If the value is high, then the robot has a high percentage of incomplete cleaning. The straight-line distance between the edge of the sub-area and the edge of the cleaned area is calculated. ,in Let be the distance to the sub-region corresponding to the i-th robot. Based on the distance, Let i be the percentage of incomplete cleaning by the i-th robot, based on The larger the robot, the corresponding Based on the principle that the smaller the value, the smaller the straight-line distance between the edge of the sub-region and the edge of the cleaned area, the position of each sub-region is adjusted. After adjustment, the sub-region is associated with the corresponding robot to form a preliminary task plan.

[0095] Next, the task plan acceptance status is fed back through step 1034. The preliminary task plan is sent to each robot in the collaboration group. Each robot extracts its current position coordinates and the center coordinates of the assigned sub-region. The straight-line distance is calculated using the distance formula in step 1031. If the distance is less than or equal to the preset acceptable distance, the robot is accepted. Otherwise, the robot is asked to adjust the distance value.

[0096] Finally, the final task sub-region is determined through step 1035. Feedback from all robots is collected. If there is an adjustment request, the straight-line distance between each robot and each sub-region is recalculated based on the distance value attached to the request. Sub-regions are assigned to robots that are closer in order of increasing distance. The association between sub-regions and robots is updated. The adjusted plan is sent to each robot again. "Closer" means that within the same collaborative group, the straight-line distance between each photovoltaic cleaning robot and a certain sub-region is compared laterally. The straight-line length is calculated using coordinates. If the distance value of a certain robot is less than the distance value of at least one other robot in the group, the sub-region is assigned to that robot first. Steps 1034 and 1035 are repeated until all photovoltaic cleaning robots have information such as the coordinate range, area size, and edge position of their corresponding task sub-regions after reassignment, forming the task sub-regions.

[0097] In practical applications, the number of robots in a photovoltaic power station can be 2, 3, 4, or other numbers. For example, a photovoltaic power station may have robots R1, R2, R3, and R4. The dynamic adjacency graph model shows their position coordinates as (10,20), (15,25), (20,30), and (25,35), respectively, with a preset distance threshold of 10 meters. The distance between R1 and R2 can be calculated using a formula. Meters, the distance between R2 and R3 Meters, the distance between R3 and R4 All four are less than 10 meters long, therefore they are grouped into the same collaborative group. Remaining battery power for each robot: R1 150 units, R2 200 units, R3 180 units, R4 120 units; energy consumption per unit area (e) is 3 units / square meter for all four, and cleaning area per unit time (s) is 4 square meters / minute for all four. Therefore, the remaining cleaning time for R1 is... Minutes, area that can be cleaned square meters; R2: Minutes, area that can be cleaned square meters; R3: Minutes, area that can be cleaned square meters; R4: Minutes, area that can be cleaned Based on the above cleanable area, the total task area for the collaborative group is determined to be 50 + 66.68 + 60 + 40 = 216.68 square meters; then, the percentage of incomplete cleaning is calculated: R1 has cleaned 30 square meters, the percentage of incomplete cleaning is [not specified], and the percentage of incomplete cleaning is [not specified]. R2 has cleaned an area of ​​40 square meters. R3 has cleaned an area of ​​25 square meters. R4 has cleaned an area of ​​15 square meters. After removing obstacles, the remaining area to be cleaned is 216.68 square meters, and the area of ​​a single photovoltaic panel is... The area is divided into two continuous strips along the direction parallel to the photovoltaic panels, and each strip is further divided vertically into two initial region blocks. Each initial region block contains... photovoltaic panels ( (The area is greater than the maximum cleaning area of ​​40 square meters for a single unit). Calculate the distance between the center of the collaborative group (17.5, 27.5) and the centers of each initial area block, and assign the four closest initial area blocks to the collaborative group. Based on the total cleaning area of ​​the collaborative group... square meters, initial area of ​​the region =60 square meters, proportion Then the area of ​​the subregion square meters, square meters, square meters, Square meters. At this point, adjustments are made to the sub-areas based on the proportion of areas not yet cleaned, and the basic distance is... Based on the principle that "the straight-line distance between the edge of the sub-area corresponding to the robot with a higher proportion of incomplete cleaning and the edge of the cleaned area is smaller," the distance between the edges of the sub-areas of each robot is calculated. rice, rice, rice, The initial task plan is formed by assigning a minimum distance of 15 meters between each robot and its assigned sub-region. This distance is then adjusted so that the edge of R4's sub-region is closest to the already cleaned area. R3's distance is next, and so on. After the initial task plan is sent, R4 reports a distance of 15 meters from its assigned sub-region (greater than the preset acceptable distance of 10 meters). The distance between each robot and its sub-region is recalculated, and R3 is found to be 8 meters away. This sub-region is then assigned to R3, and R4 is reassigned a sub-region with a distance of 9 meters. R4 accepts this assignment, thus forming the final task sub-regions.

[0098] The aforementioned overall solution (103) scientifically divides the area to be cleaned into initial blocks and sub-regions by forming collaborative groups and combining parameters such as the robot's remaining battery power and the area already cleaned to calculate the cleaning capacity and the proportion of unfinished work. A feedback adjustment mechanism determines the final task sub-regions, achieving dynamic allocation of task load based on robot model adaptability. This process fully considers the actual capabilities and positional relationships of the robots, ensuring the rationality and efficiency of task allocation, facilitating collaborative operation among robots, and improving the cleaning efficiency and coverage of photovoltaic power stations.

[0099] S104. When the photovoltaic cleaning robot detects that the distance between itself and the adjacent photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold, it adjusts the movement speed of the photovoltaic cleaning robot and the adjacent photovoltaic cleaning robot using a swarm control method. After the two adjacent photovoltaic cleaning robots reach the common boundary line, when an obstacle is detected on the trajectory, the robot adjusts the turning angle to bypass the obstacle and sends a trajectory adjustment signal to the adjacent photovoltaic cleaning robot so that the adjacent photovoltaic cleaning robot adjusts its corresponding trajectory based on the trajectory adjustment signal.

[0100] Optionally, in step S104, when the photovoltaic cleaning robot detects that the distance to its neighboring photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold, a swarm control method is used to adjust the movement speed of the photovoltaic cleaning robot and the neighboring photovoltaic cleaning robot. After the two neighboring photovoltaic cleaning robots reach a common boundary line, when an obstacle is detected on the trajectory, the turning angle is adjusted to bypass the obstacle, and a trajectory adjustment signal is sent to the neighboring photovoltaic cleaning robot so that the neighboring photovoltaic cleaning robot adjusts its corresponding trajectory based on the trajectory adjustment signal, including:

[0101] Step 1041: When the photovoltaic cleaning robot detects that the distance to the adjacent photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold, it sends a boundary synchronization request to the adjacent photovoltaic cleaning robot. Based on the boundary synchronization request, the two adjacent photovoltaic cleaning robots exchange their current travel direction and distance information to the common boundary line, and use a swarm control method to adjust their respective movement speed so that the time difference between the two adjacent photovoltaic cleaning robots reaching the common boundary line is within a preset range.

[0102] Step 1042: After the two adjacent photovoltaic cleaning robots reach the common boundary line, they move synchronously in a direction parallel to the common boundary line, maintaining a preset distance during the movement. When an obstacle is detected on the movement trajectory, the robot adjusts its turning angle to bypass the obstacle based on the obstacle information and the straight-line distance between itself and the boundary of the task sub-area. At the same time, a trajectory adjustment signal is sent to the adjacent photovoltaic cleaning robot so that the adjacent photovoltaic cleaning robot adjusts its own movement trajectory accordingly based on the trajectory adjustment signal, thereby maintaining the cooperative distance between the two adjacent photovoltaic cleaning robots within a preset distance range. Here, the boundary of the task sub-area refers to the edge line of the cleaning area that a single photovoltaic cleaning robot is responsible for, belonging exclusively to that robot and used to define its independent working range. The common boundary refers to the common edge line between two adjacent task sub-areas, which is jointly responsible by the two adjacent photovoltaic cleaning robots and is the boundary line between their working ranges. For example, the task sub-region boundary of robot A includes four edges a, b, c, and d, while the task sub-region boundary of robot B includes four edges c, e, f, and g. Edge c is the common boundary of both robots, and only edge c is shared by both robots. The other edges are the exclusive task sub-region boundaries of each robot.

[0103] Step 1042 may specifically include the following processes: collecting the vertical distance information between the two adjacent photovoltaic cleaning robots and the common boundary line; exchanging the vertical distance information through a self-organizing network communication relay node; and adjusting the travel direction of the two adjacent photovoltaic cleaning robots so that the travel direction is parallel to the common boundary line and the distance difference between them is within a preset safe distance; calculating the straight-line distance between the two adjacent photovoltaic cleaning robots; when the straight-line distance between the two adjacent photovoltaic cleaning robots is greater than a preset spacing, increasing the movement speed of the photovoltaic cleaning robot closer to the inside of the common boundary and decreasing the movement speed of the photovoltaic cleaning robot on the outside, so that the straight-line distance is within the preset spacing; when the detection of... When there are obstacles in the travel paths of two adjacent photovoltaic cleaning robots, the coordinate information and lateral width of the obstacles are recorded. Based on the coordinate information of the obstacles, the straight-line distance between the photovoltaic cleaning robot and the boundary of the task sub-area is calculated. When the straight-line distance between the photovoltaic cleaning robot and the boundary of the task sub-area is greater than the lateral width of the obstacle, the turning angle is adjusted away from the common boundary line based on the principle that the turning angle is proportional to the lateral width. Alternatively, when the straight-line distance between the photovoltaic cleaning robot and the boundary of the task sub-area is less than or equal to the lateral width, the turning angle is adjusted towards the common boundary line, and a new travel path is generated. The minimum distance between the new travel path and the common boundary line is not less than a preset safety distance.

[0104] Step 1043: After the adjacent photovoltaic cleaning robots avoid obstacles, based on the boundary information of the task sub-area updated in real time, they restore the original cleaning path or replan the travel route to the next boundary segment until the cleaning operation of the common boundary line is completed.

[0105] In the above scheme, the swarm control method integrating artificial potential field refers to a control method that combines simulated "potential force" and group cooperative motion rules to coordinate the movement of multiple robots; area boundary synchronous cleaning refers to adjacent robots cooperating to clean a common boundary; the boundary information of the task sub-region refers to the edge position data of the area to be cleaned by each robot; the preset cooperative distance threshold is the critical distance value that triggers the boundary synchronization request; the boundary synchronization request is a signal sent by a robot to adjacent robots to coordinate the cleaning of the boundary; the trajectory adjustment signal is a signal sent by a robot to adjacent robots to adjust its motion trajectory when it avoids an obstacle; the cooperative distance is the distance that should be maintained between adjacent robots; the vertical distance information is the distance of the robot to the common boundary line in the vertical direction; the lateral width of the obstacle is the width of the obstacle along the direction perpendicular to the robot's travel direction; the new travel trajectory is the new motion path planned by the robot when it avoids the obstacle; the original cleaning path is the cleaning route of the robot before encountering the obstacle.

[0106] In this application example, firstly, boundary synchronization is triggered through step 1041, and each robot detects its distance to neighboring robots in real time, using the distance formula... ,) Calculate the distance value, where d is the distance. These are the coordinates of the robot itself and its neighboring robots. When the distance is less than or equal to a preset cooperative distance threshold, a boundary synchronization request is sent to the neighboring robots. After receiving the request, the neighboring robots exchange their respective travel directions (e.g., the angle with the direction of the photovoltaic panels) and distance information to the common boundary line (e.g., vertical distance). A swarm control method is then used, based on the distance between each robot and the common boundary line. and current movement speed Through formula Calculate the arrival time difference; if the time difference exceeds a preset range, adjust the speed (e.g., increase the speed of robots at longer distances) to ensure... It is within the preset range.

[0107] Secondly, through step 1042, synchronized movement and obstacle avoidance adjustments are achieved. After adjacent robots reach the common boundary line, their vertical distances to the common boundary line are collected. , After exchanging information, they adjusted their direction of travel to ensure that all directions were parallel to the common boundary line, and If the distance is less than the preset safety distance, the straight-line distance d between the two robots is calculated in real time during movement. When d is greater than the preset distance, the robot closer to the inside increases its speed. The rate of decrease on the outer side , The distance can be determined based on the difference between d and the preset distance, keeping d within the preset distance. If an obstacle is detected ahead, its coordinates and lateral width W are recorded. The straight-line distance S between the user and the boundary of the task sub-region is calculated. When S > W, the steering angle is adjusted accordingly. (k is a proportionality coefficient) Adjust in the direction away from the common boundary line; when S≤W, adjust in the direction closer to the common boundary line. This ensures that the minimum distance between the new trajectory and the common boundary line is not less than the preset safety distance. At the same time, a trajectory adjustment signal (including the new direction and adjustment range) is sent to the adjacent robot. The adjacent robot adjusts its trajectory accordingly to maintain the cooperative distance within the preset range.

[0108] Finally, by restoring or replanning the path through step 1043 and avoiding obstacles, the robot obtains real-time updated boundary information of the task sub-area (such as changes in boundary line coordinates). If the original cleaning path is not affected, it restores to the original path. If the original path is blocked, it replans the route to the next boundary segment based on the new boundary information (such as along the extension direction of the new boundary line) and continues synchronous cleaning until the cleaning operation of the common boundary line is completed.

[0109] In practical applications, robots R1 and R2 in a photovoltaic power station are adjacent robots with a preset cooperative distance threshold of 5 meters, a preset time difference range of 2 seconds, a preset spacing of 2 meters, a preset safety distance of 0.5 meters, and a scaling factor k = 0.1 radians / meter. Robot R1's coordinates are (10, 20), its direction of travel is parallel to the common boundary line, and its perpendicular distance from the common boundary line is... ,rate m / s, robot R2 coordinates (14,23), calculated distance Meters, triggering a boundary synchronization request. Perpendicular distance between R2 and the common boundary line. meters, speed meters per second, time difference of arrival Seconds (within the range). After reaching the common boundary line, The distance is d = 3 meters (within safe distance) during synchronized movement (greater than the preset spacing of 2 meters). R1 ​​(inner side) increases its speed to 0.6 m / s, and R2 decreases it to 0.4 m / s, restoring d to 2 meters. Then, when an obstacle is detected during movement, based on the lateral width W = 1 meter and the distance S = 1.5 meters between R1 and the boundary of the task sub-region (S > W), the turning angle θ = 0.1 × 1 = 0.1 radians is adjusted away from the boundary, and a signal is sent to R2. R2 adjusts synchronously, maintaining the cooperative distance. After avoiding the obstacle, R1 and R2 resume their original path based on the updated boundary information and continue clearing the common boundary line until completion.

[0110] The aforementioned 104-part overall solution, through a swarm control method incorporating artificial potential fields, achieves synchronous boundary cleaning by adjacent robots within a preset cooperative distance, ensuring stable distance during synchronized movement. When encountering obstacles, it can flexibly adjust its turning angle based on its distance from the boundary, and simultaneously achieve cooperative obstacle avoidance through trajectory adjustment signals, preventing collisions and maintaining cooperative relationships. After avoiding obstacles, it promptly resumes or replans its path, ensuring the continuity and integrity of the cleaning operation and improving the efficiency and safety of robot swarm collaborative operations.

[0111] The following is a complete example for steps S101 to S104. Firstly, at the photovoltaic power station, there are three photovoltaic cleaning robots, A, B, and C, which can be respectively... Figure 2 The coordinates of R1, R2, and R3 are given. Position A is at (5, 10), moving at a speed of 0.3 m / s eastward; position B is at (8, 12), moving at a speed of 0.4 m / s eastward; position C is at (15, 18), moving at a speed of 0.35 m / s eastward. Using the formula... Calculate the distance, where d is the straight-line distance between the two robots. Given the coordinates of two robots, the distances between robots A and B are approximately 3.61m, A and C are approximately 10.44m, and B and C are approximately 7.81m. Using the positions of these three robots as nodes, a dynamic adjacency graph model is constructed. In the model, nodes A and B, and B and C are connected by edges, while A and C are not connected. This distance data and connectivity will be used for subsequent division of collaborative groups.

[0112] Next, based on the positional relationships of robots A, B, and C in the photovoltaic power station, the information exchange is pre-programmed to use a time-division multiple access protocol. The information aggregation node divides 1 second into 3 dedicated time segments (the preset duration of a single segment is set to 0.3 seconds based on the amount of information data): A's segment is 0-0.3 seconds, B's is 0.3-0.6 seconds, and C's is 0.6-1 seconds. At 0.1 seconds of the 0-0.3 second segment, A sends its own position (5,10) to the relay node. The relay node forwards it to B and C. B receives it at 0.15 seconds, and the time difference is calculated as 0.15-0.1=0.05 seconds, which is within the preset delay range of 0.2 seconds. When B detects an obstacle at coordinates (10,12) during its movement, it sends it to the relay node with a timestamp of 10:00:00. The relay node adds a forwarding sequence number 1 according to the receiving order and broadcasts it to A and C. This obstacle information will be used for the subsequent division of the area to be cleaned.

[0113] Then, based on the dynamic adjacency graph model, with a preset distance threshold of 10m and a preset cleaning area per unit time of 2㎡ / s, A, B, and C are grouped into the same collaborative group because their distances are all less than this value. Considering the information interaction status without delay anomalies, the preset energy consumption per unit area is a fixed value. A has 120 units of remaining power, and with an energy consumption of 2 units / ㎡, the calculated remaining cleaning time is: 120 ÷ 2 ÷ 2㎡ / s = 30s, and the area that can be cleaned is: 30 × 2 = 60㎡. 30㎡ has been cleaned, the total task area is 200㎡, and the percentage of incomplete cleaning is: (200 - 30) ÷ 200 = 0.85. B has 150 units of remaining power, and the area that can be cleaned is: (150 ÷ ​​2 ÷ 2) × 2 = 75㎡. 40㎡ has been cleaned, and the percentage of incomplete cleaning is: (200 - 40) ÷ 200 = 0.8. C has 100 units of remaining battery power, and can clean an area of ​​(100 ÷ 2 ÷ 2) × 2 = 50㎡. 25㎡ has been cleaned, leaving an uncompleted percentage of (200 - 25) ÷ 200 = 0.875. Referring to the obstacle (10, 12) broadcast in step 2, the area to be cleaned is divided into four initial blocks of 50㎡ each after removing the obstacle (the preset area of ​​a single initial block is no less than the maximum cleanable area of ​​a single unit is 40㎡), and these blocks are assigned to the collaborative group. The sub-areas are divided according to the cleanable area ratio: A: 60 ÷ (60 + 75 + 50) × 50 ≈ 15.38㎡, B: 75 ÷ 215 × 50 ≈ 19.23㎡, C: 50 ÷ 215 × 50 ≈ 15.38㎡. After minor adjustments based on the uncompleted percentage, a preliminary task plan is formed. A's sub-area is closer to the already cleaned area. This task plan will be used for subsequent cleaning operations.

[0114] Finally, based on the task sub-region determined in step 3, with a preset collaborative distance range of 5m, and sub-regions A and B being adjacent, synchronous cleaning is triggered when the distance between them reaches 5m. The vertical distance between A and the common boundary line is 1m, and for B it is 1.2m. After exchanging information, the direction is adjusted to be parallel to the boundary line, with a preset spacing of 2m. During the movement, the obstacle (10,12) broadcast in step 2 is detected, and its lateral width is measured to be 2m. The distance between A and the boundary of the task sub-region is 3m (greater than 2m), and the preset turning angle formula is used. (in (where 0.1 is the preset scaling factor and W is the lateral width). Calculate the turning angle: 0.1 × 2 = 0.2 radians. Adjust the direction away from the boundary and send a trajectory adjustment signal to B. B adjusts synchronously to maintain the cooperative distance. After avoiding obstacles, restore the original path according to the updated task sub-region boundary information and continue synchronous cleaning until completion.

[0115] Figure 3 This is a schematic diagram illustrating a specific implementation of a photovoltaic cleaning robot multi-machine task scheduling and collaborative operation system provided in this application embodiment. (Refer to...) Figure 3 The system may include:

[0116] Module 31 is used to obtain the position and speed information of each photovoltaic cleaning robot in the robot cluster in the photovoltaic power station, and to construct a dynamic adjacency graph model based on the position and speed information of each photovoltaic cleaning robot.

[0117] Control module 32 is used to control each photovoltaic cleaning robot to interact with each other through the self-organizing network communication relay node using the time division multiple access protocol, so that the communication delay between each photovoltaic cleaning robot in the robot cluster is within a preset delay range, and at the same time, obstacle information is broadcast in real time in the robot cluster through the self-organizing network communication relay node.

[0118] The partitioning module 33 is used to divide the area to be cleaned in the photovoltaic power station into sections based on the dynamic adjacency graph model and the obstacle information, combined with the remaining power and cleaned area of ​​each photovoltaic cleaning robot. The robot cluster determines the task sub-region of each photovoltaic cleaning robot to realize the dynamic allocation of task load according to the model adaptability.

[0119] The adjustment module 34 is used to adjust the movement speed of the photovoltaic cleaning robot and the adjacent photovoltaic cleaning robot by means of swarm control when the photovoltaic cleaning robot detects that the distance between itself and the adjacent photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold. After the two adjacent photovoltaic cleaning robots reach the common boundary line, when an obstacle is detected on the trajectory, the turning angle is adjusted to bypass the obstacle. At the same time, a trajectory adjustment signal is sent to the adjacent photovoltaic cleaning robot so that the adjacent photovoltaic cleaning robot adjusts its corresponding trajectory according to the trajectory adjustment signal.

[0120] This application provides a photovoltaic cleaning robot multi-machine task scheduling and collaborative operation system to implement the aforementioned photovoltaic cleaning robot multi-machine task scheduling and collaborative operation method. Therefore, the specific implementation of the photovoltaic cleaning robot multi-machine task scheduling and collaborative operation system can be found in the embodiment section of the photovoltaic cleaning robot multi-machine task scheduling and collaborative operation method above. The specific implementation can be referred to the description of the corresponding embodiment, and will not be repeated here.

[0121] like Figure 4 As shown, this application also provides an electronic device, including: a memory 41 for storing a computer program; and a processor 42 for executing the computer program to implement the steps of the above-described method for multi-machine task scheduling and collaborative operation of a photovoltaic cleaning robot.

[0122] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for multi-machine task scheduling and collaborative operation of a photovoltaic cleaning robot.

[0123] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0124] The embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the embodiments of the photovoltaic cleaning robot multi-machine task scheduling and collaborative operation method.

[0125] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0126] The foregoing has provided a detailed description of a multi-machine task scheduling and collaborative operation method, system, electronic device, and storage medium for a photovoltaic cleaning robot provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A method for multi-machine task scheduling and collaborative operation of photovoltaic cleaning robots, characterized in that, include: Obtain the position and speed information of each photovoltaic cleaning robot in the robot cluster in the photovoltaic power station, and construct a dynamic adjacency graph model based on the position and speed information of each photovoltaic cleaning robot; The photovoltaic cleaning robots are controlled to interact with each other through the self-organizing network communication relay node using the time division multiple access protocol, so that the communication delay between the photovoltaic cleaning robots in the robot cluster is within a preset delay range. At the same time, obstacle information is broadcast in real time in the robot cluster through the self-organizing network communication relay node. Based on the dynamic adjacency graph model and the obstacle information, combined with the remaining power and cleaned area of ​​each photovoltaic cleaning robot, the area to be cleaned in the photovoltaic power station is divided into sections. The task sub-area of ​​each photovoltaic cleaning robot is determined by the robot cluster, so as to realize the dynamic allocation of task load according to the model adaptability. When the photovoltaic cleaning robot detects that the distance to its neighboring photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold, it uses a swarm control method to adjust the movement speed of the photovoltaic cleaning robot and the neighboring photovoltaic cleaning robot. After the two neighboring photovoltaic cleaning robots reach a common boundary line, when an obstacle is detected on the trajectory, the robot adjusts its turning angle to bypass the obstacle and sends a trajectory adjustment signal to the neighboring photovoltaic cleaning robot so that the neighboring photovoltaic cleaning robot adjusts its corresponding trajectory based on the trajectory adjustment signal. Based on the dynamic adjacency graph model and the obstacle information, combined with the remaining power and cleaned area of ​​each photovoltaic cleaning robot, the area to be cleaned in the photovoltaic power station is divided into sections. The task sub-region of each photovoltaic cleaning robot is determined by the robot cluster, including the following steps: The remaining power, cleaned area, and energy consumption per unit area of ​​each photovoltaic cleaning robot are obtained. Based on the connection relationship of each node in the dynamic adjacency graph model, photovoltaic cleaning robots with a distance value less than a preset distance threshold are grouped into the same cooperative group. The remaining cleaning time is calculated based on the remaining power of each photovoltaic cleaning robot in the same collaborative group, and the cleanable area of ​​each photovoltaic cleaning robot in the same collaborative group is calculated based on the energy consumption per unit area. The proportion of incomplete cleaning is determined based on the cleaned area. The remaining area in the photovoltaic power station area to be cleaned, excluding the obstacle information, is divided into several consecutive initial area blocks, wherein the area of ​​each initial area block is not less than the maximum area that a single photovoltaic cleaning robot can clean in one go, and at least two initial area blocks are allocated to each of the same collaborative groups. Based on the cleanable area and the proportion of incomplete cleaning, sub-areas in the initial area blocks are allocated to each photovoltaic cleaning robot in the same collaborative group to form a preliminary task plan. The preliminary task plan is sent to each photovoltaic cleaning robot in the same collaborative group, so that each photovoltaic cleaning robot can respond with an acceptance or adjustment request based on its current position and the straight-line distance between itself and the sub-area. If any of the photovoltaic cleaning robots requests an adjustment, the sub-region is reassigned based on the acceptance or adjustment requests from all photovoltaic cleaning robots. The sub-region is preferentially assigned to the photovoltaic cleaning robot that is closer to the sub-region, until all photovoltaic cleaning robots in the same cooperative group accept the assignment result, thus forming a task sub-region.

2. The method according to claim 1, characterized in that, The remaining area within the photovoltaic power station's cleaning area, excluding the obstacle information, is divided into several consecutive initial area blocks. The area of ​​each initial area block is not less than the maximum cleaning area that a single photovoltaic cleaning robot can clean in a single operation. At least two initial area blocks are allocated to each collaborative group. Based on the cleanable area and the proportion of uncompleted cleaning, sub-areas within the initial area blocks are allocated to each photovoltaic cleaning robot within the same collaborative group, forming a preliminary task plan, including the following steps: Along the arrangement direction parallel to the photovoltaic panels in the photovoltaic power station, the remaining area is divided into several continuous strips. Along the arrangement direction perpendicular to the photovoltaic panels, each continuous strip is divided into several initial area blocks. The area of ​​each initial area block is not less than the maximum cleaning area that a single photovoltaic cleaning robot can clean in one go, and the initial area block contains an integer number of photovoltaic panels. Based on the location information of all photovoltaic cleaning robots in each collaborative group, the straight-line distance between each collaborative group and the center of each initial area block is calculated. Based on the allocation rule that the straight-line distance between the collaborative group and the corresponding initial area block does not exceed the maximum distance between any two photovoltaic cleaning robots in the collaborative group, at least two initial area blocks that are closer to each other are preferentially allocated to the corresponding collaborative group. The total cleanable area of ​​all photovoltaic cleaning robots in each collaborative group is counted, the ratio of the area of ​​the initial area block to the total cleanable area is calculated, and the initial area block is divided into sub-areas with the same number of photovoltaic cleaning robots in the collaborative group according to the ratio. Based on the cleanable area of ​​each photovoltaic cleaning robot, the area of ​​each sub-region is determined so that the ratio of the area of ​​the sub-region to the cleanable area of ​​the corresponding photovoltaic cleaning robot tends to be consistent. Based on the incomplete cleaning ratio of each photovoltaic cleaning robot, and adhering to the principle that the straight-line distance between the edge of the sub-region corresponding to the photovoltaic cleaning robot with a higher incomplete cleaning ratio and the edge of the cleaned region is smaller, the sub-region is adjusted. The formula for calculating the straight-line distance is as follows: , in the formula This represents the distance to the sub-region corresponding to the i-th photovoltaic cleaning robot. Indicates the base distance. This represents the uncompleted cleaning percentage of the i-th photovoltaic cleaning robot. The adjusted sub-region is then associated with the corresponding photovoltaic cleaning robot to form a preliminary task plan.

3. The method according to claim 1, characterized in that, When the photovoltaic cleaning robot detects that the distance to its neighboring photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold, it adjusts the movement speed of the photovoltaic cleaning robot and the neighboring photovoltaic cleaning robot using a swarm control method. After the two neighboring photovoltaic cleaning robots reach a common boundary line, when an obstacle is detected on their trajectory, the robot adjusts its turning angle to avoid the obstacle and simultaneously sends a trajectory adjustment signal to the neighboring photovoltaic cleaning robot, so that the neighboring photovoltaic cleaning robot adjusts its corresponding trajectory based on the trajectory adjustment signal. This includes the following steps: When the photovoltaic cleaning robot detects that the distance to its neighboring photovoltaic cleaning robot is less than or equal to a preset collaborative distance threshold, it sends a boundary synchronization request to the neighboring photovoltaic cleaning robot. Based on the boundary synchronization request, the two neighboring photovoltaic cleaning robots exchange their current direction of travel and distance information to the common boundary line, and use a swarm control method to adjust their respective movement speed so that the time difference between the two neighboring photovoltaic cleaning robots reaching the common boundary line is within a preset range. After the two adjacent photovoltaic cleaning robots reach the common boundary line, they move synchronously in a direction parallel to the common boundary line, maintaining a preset distance during the movement. When an obstacle is detected on the movement trajectory, the robot adjusts its turning angle to bypass the obstacle based on the obstacle information and the straight-line distance between itself and the boundary of the task sub-area. At the same time, it sends a trajectory adjustment signal to the adjacent photovoltaic cleaning robot, so that the adjacent photovoltaic cleaning robot adjusts its own movement trajectory accordingly based on the trajectory adjustment signal, thereby maintaining the cooperative distance between the two adjacent photovoltaic cleaning robots within a preset distance range. After the adjacent photovoltaic cleaning robots avoid obstacles, they resume the original cleaning path or replan their travel route to the next boundary segment based on the boundary information of the task sub-area updated in real time, until the cleaning operation of the common boundary line is completed.

4. The method according to claim 3, characterized in that, The robot moves synchronously in a direction parallel to the common boundary line, maintaining a preset distance during movement. When an obstacle is detected on the trajectory, the robot adjusts its turning angle to avoid the obstacle based on the obstacle information and the straight-line distance between the photovoltaic cleaning robot and the boundary of the task sub-area. This includes the following steps: The vertical distance information between the two adjacent photovoltaic cleaning robots and the common boundary line is collected. The vertical distance information is exchanged through the self-organizing network communication relay node. The travel direction of the two adjacent photovoltaic cleaning robots is adjusted so that the travel direction is parallel to the common boundary line and the distance difference between them and the common boundary line is within a preset safe distance. Calculate the straight-line distance between two adjacent photovoltaic cleaning robots. When the straight-line distance between two adjacent photovoltaic cleaning robots is greater than a preset distance, increase the movement speed of the photovoltaic cleaning robot closer to the inside of the common boundary and decrease the movement speed of the photovoltaic cleaning robot on the outside, so that the straight-line distance is within the preset distance. When an obstacle is detected in the travel trajectory in front of the two adjacent photovoltaic cleaning robots, the coordinate information and lateral width of the obstacle are recorded. Based on the coordinate information of the obstacle, the straight-line distance between the photovoltaic cleaning robot itself and the boundary of the task sub-area is calculated. When the straight-line distance between the photovoltaic cleaning robot and the boundary of the task sub-area is greater than the lateral width of the obstacle, the robot adjusts its turning angle away from the common boundary line based on the principle that the turning angle is proportional to the lateral width. Alternatively, when the straight-line distance between the photovoltaic cleaning robot and the boundary of the task sub-area is less than or equal to the lateral width, the robot adjusts its turning angle towards the common boundary line and generates a new travel trajectory. The minimum distance between the new travel trajectory and the common boundary line is not less than a preset safety distance.

5. The method according to claim 1, characterized in that, The process of constructing a dynamic adjacency graph model based on the position and speed information of each photovoltaic cleaning robot includes the following steps: Based on the movement speed and direction in the location and speed information, the straight-line distance between every two photovoltaic cleaning robots in the robot cluster is calculated. Using the location point corresponding to the coordinate information of each photovoltaic cleaning robot as a node, when the straight-line distance between the two photovoltaic cleaning robots is within the preset adjacency determination range and the angle between the movement direction and the boundary line of their respective current task sub-regions meets the preset angle condition, a connecting edge is established between the location points of the two photovoltaic cleaning robots, the edge weight is determined based on the movement speed, and a dynamic adjacency graph model is constructed.

6. The method according to claim 1, characterized in that, The control of each photovoltaic cleaning robot via a self-organizing network communication relay node using a time-division multiple access protocol ensures that the communication delay between the photovoltaic cleaning robots in the robot cluster is within a preset delay range. Simultaneously, obstacle information is broadcast in real-time throughout the robot cluster via the self-organizing network communication relay node. This includes the following steps: According to the preset time segment division rules, each photovoltaic cleaning robot in the robot cluster is assigned a dedicated information sending time segment. The duration of the information sending time segment is set according to the amount of data, including location information, speed information and task status information, required for a single information interaction. Each photovoltaic cleaning robot is controlled to send the information to be interacted to the self-organizing network communication relay node within its own dedicated information sending time segment. The self-organizing network communication relay node adds a corresponding forwarding sequence number to each of the information to be interacted according to the receiving order, and broadcasts the information to be interacted to all other photovoltaic cleaning robots. The information to be interacted includes obstacle information, coordinate information of the collection location, and time stamp of the collection time. After receiving a delay anomaly signal sent by any other photovoltaic cleaning robot under preset conditions, the dedicated time segments corresponding to the two photovoltaic cleaning robots with the delay anomaly are readjusted, and the interval between the dedicated time segments is shortened. The preset condition is that the total time difference exceeds the preset delay range. The total time difference is the time difference between when the information to be interacted is sent from itself to when it is received by the target photovoltaic cleaning robot.

7. A multi-machine task scheduling and collaborative operation system for photovoltaic cleaning robots, characterized in that, A method for scheduling and coordinating multi-machine tasks of a photovoltaic cleaning robot as described in claim 1 includes: The module is used to obtain the position and speed information of each photovoltaic cleaning robot in the robot cluster in the photovoltaic power station, and to construct a dynamic adjacency graph model based on the position and speed information of each photovoltaic cleaning robot. The control module is used to control each photovoltaic cleaning robot to interact with each other through the self-organizing network communication relay node using the time division multiple access protocol, so that the communication delay between each photovoltaic cleaning robot in the robot cluster is within a preset delay range, and at the same time, the obstacle information is broadcast in real time in the robot cluster through the self-organizing network communication relay node. The partitioning module is used to divide the area to be cleaned in the photovoltaic power station into sections based on the dynamic adjacency graph model and the obstacle information, combined with the remaining power and cleaned area of ​​each photovoltaic cleaning robot. The robot cluster determines the task sub-region of each photovoltaic cleaning robot so as to realize the dynamic allocation of task load according to the model adaptability. The adjustment module is used to adjust the movement speed of the photovoltaic cleaning robot and the adjacent photovoltaic cleaning robot by means of swarm control when the photovoltaic cleaning robot detects that the distance between itself and the adjacent photovoltaic cleaning robot is less than or equal to a preset cooperative distance threshold. After the two adjacent photovoltaic cleaning robots reach the common boundary line, when an obstacle is detected on the trajectory, the turning angle is adjusted to bypass the obstacle. At the same time, a trajectory adjustment signal is sent to the adjacent photovoltaic cleaning robot so that the adjacent photovoltaic cleaning robot adjusts its corresponding trajectory according to the trajectory adjustment signal.

8. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor is configured to execute the computer program to implement the steps of a multi-machine task scheduling and collaborative operation method for a photovoltaic cleaning robot as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables a method for multi-machine task scheduling and collaborative operation of a photovoltaic cleaning robot as described in any one of claims 1 to 6.

Images