A control system and control method for a weeding robot

Abstract

This invention relates to the field of robotics, specifically to a weeding robot control system and method. The system includes a weeding robot, a storage chamber, and a backend terminal communicating with the weeding robot. The weeding robot is equipped with a positioning module, a scanning module, a processing module, and a storage module. A charging device is located within the storage chamber. The storage chamber has an opening for the weeding robot to enter and exit. It also includes a magnetic conductor that starts from the charging device, runs along the interior boundary of the storage chamber and the boundary of the working area, and returns to the charging device, forming a closed area. The weeding robot is also equipped with an electromagnetic sensor. The backend terminal stores a working map, which is a topological map formed based on a baseline map, an interior map of the storage chamber, and an exterior map of the working area. The baseline map is the closed area formed by the magnetic conductor. This application avoids the need for manual secondary trimming and enables the weeding robot to move intelligently throughout the entire process.

Description

Technical Field

[0001] This invention relates to the field of robotics, specifically to a weeding robot control system and control method. Background Technology

[0002] Weeding robots, also known as intelligent weeding robots, are typically used for weeding operations in orchards, farms, and other similar locations. Compared to manual weeding, intelligent weeding robots offer advantages such as higher efficiency and better stability.

[0003] Because lawnmowers operate in open, unfenced outdoor environments, unlike indoor robotic vacuum cleaners, the indoor navigation and obstacle avoidance systems used in indoor vacuum cleaners cannot be directly applied to lawnmowers. Therefore, during operation, lawnmowers typically follow pre-planned paths (such as GPS-based path planning) and return to their initial position after completing their work, where they are then transported back to their storage location by staff. Additionally, when the lawnmower's battery is low, it will usually issue a notification, allowing staff to easily return it to its charging station. For ease of operation, some staff build storage rooms around the lawnmower's work area and install charging devices within these rooms. This allows staff to conveniently transport the robot to the storage room for storage or charging when it finishes its work or runs out of power.

[0004] While the above technologies can achieve intelligent weed control to some extent, they still have the following problems:

[0005] First, the working path of the weeding robot is based on GPS, but the GPS accuracy / signal in many areas (such as many orchards) is neither good nor stable. The navigation route of the weeding robot often deviates significantly. After the weeding robot works according to the path with a large deviation, there may be a situation where manual secondary pruning is required.

[0006] Secondly, when the weeding robot finishes its work or runs out of power, it still requires manual assistance to move it to the storage room, which remains relatively cumbersome. Some staff have tried to enable the weeding robot to move autonomously indoors and outdoors, but the indoor GPS positioning signal is too weak to be used for path planning. If indoor positioning technology is used, there are navigation problems at the connecting sections between indoor and outdoor areas. In addition, the accuracy of outdoor GPS signals is also a problem. As a result, there has been no breakthrough in the fully autonomous indoor and outdoor movement of the weeding robot. In other words, with the current technology, it is difficult to achieve the technology of fully autonomous indoor and outdoor movement of the weeding robot. Therefore, those skilled in the art can only improve and optimize the convenience of manual assistance in moving the robot.

[0007] In summary, how to avoid the need for secondary manual trimming while enabling fully intelligent movement of weeding robots to further reduce the workload of workers has become an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0008] In view of the shortcomings of the prior art, the technical problem to be solved by the present invention is: how to avoid the need for secondary manual trimming, while enabling the weeding robot to move intelligently throughout the process, thereby further reducing the workload of workers.

[0009] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0010] A weeding robot control system includes a weeding robot, a storage chamber, and a back-end terminal for communicating with the weeding robot; the storage chamber has an opening for the weeding robot to enter and exit; a charging device is installed inside the storage chamber; it also includes a magnetic wire, which starts from the charging device, runs along the indoor boundary of the storage chamber and the outdoor working area boundary, and returns to the charging device to form a closed area; the magnetic wire includes multiple long straight segments.

[0011] The weeding robot includes a positioning module, a processing module, and an electromagnetic sensor. The positioning module is used for real-time positioning, the electromagnetic sensor is used to collect magnetic data and send it to the processing module, and the processing module is used to generate a working map and plan the path of the weeding robot based on the positioning information and magnetic data. The weeding robot has two modes: a map generation mode and a weeding working mode.

[0012] The backend stores a working map, which is a topological map formed by a base map, an indoor map of the storage room, and an outdoor map of the working area. The base map is a closed area formed by magnetic wires.

[0013] Accordingly, the present invention also provides a method for generating the working map of the aforementioned weeding robot. To this end, the present invention adopts the following technical solution:

[0014] A method for generating a working map for a weeding robot, characterized in that it is applied to the aforementioned weeding robot control system, comprising:

[0015] S1. After the weeding robot starts the map generation mode, control the weeding robot to circle around the magnetic wire, generate a virtual boundary based on the positioning information of the weeding robot, and obtain a reference map. The reference map is used to describe the closed area formed by the magnetic wire as the boundary.

[0016] S2. Control the weeding robot to move in the storage room according to the preset working route, generate an indoor map based on the SLAM algorithm, and analyze the position of the weeding robot in the reference map in real time based on the data of the magnetic sensor during the movement.

[0017] S3. Control the weeding robot to move in the work area according to the preset work route, and build an outdoor map by combining laser SLAM outdoor mapping technology; during the movement, the position of the weeding robot in the reference map is analyzed in real time based on the data of the magnetic sensor.

[0018] S4. Using the position of the weeding robot in the baseline map as a reference, merge the baseline map, indoor map and outdoor map to obtain a topology map, and store it as the working map.

[0019] In the above-mentioned method for generating a working map for a weeding robot, as a preferred embodiment, in step S1,

[0020] The process of generating the virtual boundary includes using the positioning data obtained by the weeding robot as it circles the magnetic conductor once as the reference positioning data of the magnetic conductor, and using this reference positioning data as the virtual boundary of the closed area formed by the magnetic conductor.

[0021] In the above-mentioned method for generating a working map for a weeding robot, as a preferred embodiment, step S2, which involves analyzing the position of the weeding robot on the reference map in real time based on the data from the magnetic sensor, includes analyzing the positional relationship between the weeding robot and each long straight segment of the magnetic wire based on the data from the magnetic sensor and the current data of the magnetic wire, obtaining the position of the weeding robot within the closed area formed by the magnetic wire, and then analyzing the position of the weeding robot on the reference map.

[0022] In the above-mentioned method for generating a working map for a weeding robot, as a preferred embodiment, after executing S4, the method further includes:

[0023] S5. When the weeding robot is working in weeding mode, the indoor and outdoor maps are updated in real time.

[0024] S6. A gate mechanism is set at the intersection of the baseline map, indoor map and outdoor map. The gate mechanism is the update trigger mechanism for the working map. When the robot moves to the gate mechanism, the working map is updated according to the latest indoor map and outdoor map.

[0025] In the above-mentioned method for generating a working map for a weeding robot, as a preferred embodiment, step S4 includes:

[0026] S41. After rasterizing the reference map, extract the corresponding grids on the reference map for the driving trajectory of the weeding robot in S2 and S3, and mark them as candidate topology node sets.

[0027] S42. Obtain the environmental information of each candidate topology node in the indoor or outdoor map, and determine whether the environmental information corresponding to each candidate topology node meets the preset topology conditions. If it meets the conditions, the corresponding grid area is taken as a topology node; otherwise, the corresponding grid area is taken as an edge.

[0028] S43. Connect the topology nodes according to the preset connection method to obtain the topology map.

[0029] Furthermore, the present invention also provides a working control method for the aforementioned weeding robot. To this end, the present invention adopts the following technical solution:

[0030] A method for controlling a weeding robot, using the aforementioned weeding robot control system, includes:

[0031] Step 1: After the weeding robot starts its weeding mode, the current position of the weeding robot is obtained through the electromagnetic sensor.

[0032] Step 2: If the current location is inside the storage room, proceed to Step 3; if the current location is outside the storage room, proceed to Step 4.

[0033] Step 3: Combining the working map and current location information, control the weeding robot to move to the passage opening of the storage room, update the working map, and then proceed to step 4.

[0034] Step 4: Generate a working path by combining the working map and the current location information, and control the weeding robot to work according to the working path.

[0035] In the above-mentioned weeding robot control method, as a preferred solution, in step one, after controlling the weeding robot to start the weeding working mode, the current battery level of the weeding robot is also obtained; if the current battery level is lower than the preset minimum battery level, then proceed to step five, otherwise proceed to step two.

[0036] Step 5: After generating the charging path based on the work map and current location information, control the weeding robot to travel along the charging path to the charging device to charge; and continuously obtain the current charging power of the weeding robot. If the current charging power is greater than the preset minimum charging power, proceed to step 3.

[0037] In the above-mentioned weeding robot control method, as a preferred embodiment, in step four, the working path includes multiple target positions arranged in sequence; when controlling the weeding robot to work according to the working path, the optimal path planning between adjacent target positions is performed through DWA path planning.

[0038] In the above-mentioned weeding robot control method, as a preferred solution, in step four, when controlling the weeding robot to work according to the working path, obstacle avoidance is also performed in real time based on the detection data of the lidar.

[0039] Compared with the prior art, this application has the following advantages:

[0040] 1. The working path of the weeding robot in this application is generated based on a topology map, which in turn is generated based on a reference map obtained from magnetic conductors. Since the position of the magnetic conductors is essentially fixed, the position of each location on the topology map is also fixed on the reference map. In essence, this application obtains a relative position system through the magnetic conductors. Based on this relative position system, the weeding robot can accurately determine its precise indoor / outdoor location during operation. Arranging the specific locations the weeding robot needs to traverse in a certain order yields the working path. The weeding robot operates according to this working path, ensuring the accuracy of its forward position and preventing situations where excessive deviations in the navigation route necessitate manual corrections.

[0041] 2. This application uses magnetic wires to form a grid map, then merges the indoor and outdoor maps with the grid map to obtain a topological map. This topological map contains both indoor and outdoor location information, achieving seamless global navigation both indoors and outdoors. When using the weeding robot, staff only need to power it on; subsequent operations (such as automatically driving to the storage location or the charging location) can be completed automatically without human assistance.

[0042] 3. The presence of magnetic wires can also serve as a boundary limit, preventing the weeding robot from traveling to areas outside the baseline map during operation and causing damage such as bumps, collisions, or falls.

[0043] 4. Each time the weeding robot works, it updates the current indoor and outdoor maps in a timely manner. The next time it works, it will generate a topology map based on the updated information as the updated working map, which can ensure the effectiveness of the working map.

[0044] 5. The weeding robot in this application will perform a power check after power-on and before starting work. If the power is lower than the preset power, it will be controlled to charge before starting work, which can prevent insufficient power during operation and affect the overall efficiency. This ensures the stability of the overall efficiency. Attached Figure Description

[0045] To make the objectives, technical solutions, and advantages of the invention clearer, the invention will now be described in further detail with reference to the accompanying drawings, wherein:

[0046] Figure 1 This is a logic block diagram of the weeding robot control system in the embodiment;

[0047] Figure 2 This is a flowchart illustrating the method for generating a working map for the weeding robot in this embodiment;

[0048] Figure 3 This is a flowchart of the weeding robot control method in the embodiment. Detailed Implementation

[0049] The invention will now be described in further detail with reference to the accompanying drawings. Example

[0050] To address the shortcomings of existing technologies, this invention provides a control system and method for a weeding robot. It creatively introduces magnetic guide wires and, through their rational arrangement, effectively integrates indoor and outdoor map information. Furthermore, it generates a relative position system based on the baseline map obtained from the magnetic guide wires, ensuring the accuracy of the weeding robot's autonomous movement and avoiding the need for manual secondary trimming. This also enables the weeding robot to move intelligently throughout the entire process, further reducing the workload of workers. Additionally, it improves the safety of the weeding robot during operation.

[0051] The weeding robot control system of this invention, such as Figure 1 As shown, it includes a weeding robot, a storage chamber, magnetic wires, and a back-end terminal that communicates with the weeding robot.

[0052] In this embodiment, the storage chamber is a hollow cuboid, and a charging device is installed inside. Specifically, in this embodiment, the charging device is a charging pile. The storage chamber also has an opening for the weeding robot to enter and exit; specifically, the opening of the opening faces the working area, and the wall of the storage chamber where the opening is located is adjacent to the boundary of the working area. For ease of explanation, the vertical projection of the working area in this embodiment is rectangular. The bottom part of the working area overlaps with the bottom part of the wall where the storage chamber's opening is located. The magnetic wire starts from the charging device, runs around the interior boundary of the storage chamber and the boundary of the working area, and returns to the charging device, forming a closed area. The magnetic wire includes multiple long straight segments. The specific number and length of the long straight segments can be specifically set by those skilled in the art according to the size and shape of the storage chamber and the working area.

[0053] The weeding robot is equipped with a positioning module, a scanning module, a communication module, a storage module, an obstacle avoidance module, and electromagnetic sensors. In this embodiment, the positioning module is a GPS / BeiDou positioning module; the communication module is a 5G module; and both the obstacle avoidance module and the scanning module are LiDAR modules. The positioning module is used for real-time positioning, and the electromagnetic sensors are used to collect magnetic data and send it to the processing module. The processing module is used to generate a working map and plan the path for the weeding robot based on the positioning information and magnetic data. The weeding robot has two modes: a map generation mode and a weeding working mode.

[0054] The backend is a server. The backend stores a working map, which is a topological map formed based on a baseline map, an indoor map of the storage room, and an outdoor map of the working area. The baseline map is a closed area formed by magnetic conductors.

[0055] like Figure 2 As shown, this application also provides a method for generating a working map for a weeding robot, applied to the aforementioned weeding robot control system, comprising:

[0056] S1. After controlling the weeding robot to start the map generation mode, control the weeding robot to circle around the magnetic conductor once. Generate a virtual boundary based on the positioning information of the weeding robot and obtain a reference map. The reference map is used to describe the closed area formed by the magnetic conductor as the boundary. Specifically, the virtual boundary generation process includes using the positioning data obtained when the weeding robot circles around the magnetic conductor once as the reference positioning data of the magnetic conductor, and using the reference positioning data as the virtual boundary of the closed area formed by the magnetic conductor.

[0057] Although GPS signals in the working environment of the weeding robot are unstable and their accuracy is difficult to guarantee, maintaining consistent GPS signal strength over a short period is still possible. When generating the reference map, virtual boundaries are created using positioning information. This positioning information is merely reference data (i.e., reference positioning data) used to describe the boundaries of the reference map; its specific accuracy is not important because this application does not rely on GPS positioning information for navigation, but rather on the free position system of the reference map. In subsequent operations, if the GPS accuracy changes (e.g., from a 10-meter eastward offset to a 15-meter eastward offset), it is only necessary to perform a general translation of the received GPS values ​​based on the reference map.

[0058] S2. Following a preset work route, control the weeding robot to move within the storage room and generate an indoor map based on the SLAM algorithm. During movement, analyze the robot's position on the baseline map in real time based on data from the magnetic sensor. SLAM (Simultaneous Localization and Mapping) is an algorithm for simultaneous localization and mapping, or concurrent mapping and localization. Specifically, when building the indoor map, a basic SLAM algorithm can be implemented using the ROS platform on a Linux system. The weeding robot can then build an indoor environment map, perform intelligent obstacle avoidance navigation, and achieve indoor localization. Its underlying architecture primarily uses STM32 and Raspberry Pi as its core, employing LiDAR to acquire indoor map information. Corresponding scripts are executed on ROS, calling Gmapping and ACML algorithms to achieve map building and localization.

[0059] Meanwhile, according to the Biot-Saffar law, when a stable direct current is passed through a long straight conductor, an electromagnetic field will be induced around the conductor. The weeding robot can detect the electromagnetic field through electromagnetic sensors and thus identify the position of the magnetic conductor in the environment. In other words, based on the data from the magnetic sensor and the current data in the magnetic conductor, the positional relationship between the weeding robot and each long straight segment of the magnetic conductor can be analyzed, thereby understanding its position within the closed area formed by the magnetic conductor, and thus obtaining its position on the reference map.

[0060] The preset working route can be set by those skilled in the art according to the specific work content of the weeding robot in the specific use environment.

[0061] S3. Following a preset work route, control the weeding robot to move within the work area, and construct an outdoor map using the robot's onboard positioning system and gyroscope, combined with laser SLAM outdoor mapping technology; during movement, analyze the weeding robot's position on the baseline map in real time based on data from the magnetic sensor. The construction of the outdoor map is largely the same as that of the indoor map, and will not be elaborated further here.

[0062] S4. Using the position of the weeding robot in the base map as a correlation, the base map, indoor map, and outdoor map are fused to obtain a topology map, which is then stored as the working map. Specifically, S4 includes:

[0063] S41. After rasterizing the reference map, extract the corresponding grids on the reference map for the weeding robot's travel trajectory in S2 and S3, and mark them as candidate topological node sets. In this embodiment, when rasterizing the reference map, the width of the grid is the same as the weeding width when the robot moves forward.

[0064] S42. Obtain the environmental information of each candidate topology node in the indoor or outdoor map, and determine whether the environmental information corresponding to each candidate topology node meets the preset topology conditions. If it does, the corresponding grid area is taken as a topology node; otherwise, the corresponding grid area is taken as an edge. The specific topology conditions for the topology nodes can be set by those skilled in the art based on the specific working environment of the weeding robot. In this embodiment, the specific content of the topology conditions is as follows: if, based on the current position information of the mobile robot, it is determined that the current area can be divided into mapping sub-areas and a local grid map is created and saved in this area, and there are no sub-areas that have not yet been created in the current area, then the current position meets the topology conditions. The purpose of this processing is to determine whether there is a new working area in the indoor or outdoor map environment (i.e., to determine whether the indoor or outdoor map environment area has changed). If so, the map is updated accordingly through topology.

[0065] S43. Connect the topology nodes according to the preset connection method to obtain the topology map.

[0066] S5. When the weeding robot is working in weeding mode, the indoor and outdoor maps are updated in real time.

[0067] S6. A gate mechanism is set at the intersection of the base map, indoor map, and outdoor map. The gate mechanism is a trigger mechanism for updating the working map. When the robot moves to the gate mechanism, the working map is updated according to the latest indoor and outdoor maps. In this embodiment, the passage opening of the storage room is one of the gate mechanism locations.

[0068] like Figure 3 As shown, the present invention also provides a weeding robot control method, which uses the above-mentioned weeding robot control system and includes:

[0069] Step 1: After the weeding robot starts its weeding working mode, the current position and current battery level of the weeding robot are obtained through the electromagnetic sensor. If the current battery level is lower than the preset minimum battery level, proceed to step 5; otherwise, proceed to step 2. In this embodiment, the preset minimum battery level is 45%. In other embodiments, those skilled in the art can set the minimum battery level according to the workload of the weeding robot in a single operation, which will not be elaborated here.

[0070] Step 2: If the current location is inside the storage room, proceed to Step 3; if the current location is outside the storage room, proceed to Step 4.

[0071] Step three: Combining the working map and current location information, control the weeding robot to move to the entrance of the storage chamber's passageway, update the working map, and then proceed to step four. Since the environment inside the storage chamber and the orchard environment may change, if the working area... Figure 1If the initial state is maintained, the effectiveness of the working map will gradually decrease over time. With this setting, every time the weeding robot moves to the opening of the passage, the current indoor and outdoor maps will be updated in a timely manner. The next time it works, a topology map will be generated based on the updated information to serve as the updated working map, thus ensuring the effectiveness of the working map.

[0072] Step 4: Generate a working path by combining the working map and the current location information, and control the weeding robot to work according to the working path. The working path includes multiple target locations arranged in sequence. When controlling the weeding robot according to the working path, optimal path planning between adjacent target locations is performed using Dynamic Window Approach (DWA).

[0073] To ensure the effectiveness of the work path, this embodiment plans the path based on the principle of completely covering the working environment. The planned path inflection points are used as nodes for robot navigation, and the optimal path selection between the current position and the next target point is performed using DWA path planning. While controlling the weeding robot to work according to the work path, real-time obstacle avoidance is also performed based on LiDAR detection data. Since the intelligent weeding robot operates in a hilly orchard environment, its working area includes complex orchard environments such as steps, small hills, pits, and ditches. The robot should perceive these environmental features and actively avoid them. To perceive cliff environments, in addition to LiDAR and GPS / BeiDou positioning obstacle avoidance sensors, infrared sensors can be mounted on the front and bottom vertical directions of the weeding robot. Based on the reflection ranging principle of these sensors, steps, hills, and mounds above the horizontal ground can be detected in front, and pits and ditches below the horizontal ground can be detected in the bottom vertical direction.

[0074] Step 5: After generating a charging path based on the work map and current location information, control the weeding robot to travel along the charging path to the charging device for charging; continuously acquire the current charging level of the weeding robot. If the current charging level is greater than the preset minimum charging level, proceed to Step 3. Controlling the weeding robot to charge before resuming work when the charging level is lower than the preset level prevents insufficient power during operation and ensures overall efficiency. In this embodiment, the preset minimum charging level is 80%.

[0075] In this application, the working path of the weeding robot is generated based on a topology map, which in turn is generated based on a reference map obtained from magnetic conductors. Since the position of the magnetic conductors is essentially fixed, the position of each location on the topology map is also fixed on the reference map. Essentially, this application obtains a relative position system through the magnetic conductors. Based on this relative position system, the weeding robot can accurately determine its indoor / outdoor location during operation. Arranging the specific locations the weeding robot needs to traverse in a certain order yields the working path. The weeding robot operates according to this working path, ensuring the accuracy of its forward position and preventing situations where excessive deviations in the navigation route necessitate manual corrections.

[0076] In addition, this application uses magnetic wires to form a grid map, then merges the indoor and outdoor maps with the grid map to obtain a topological map. This topological map includes both indoor and outdoor location information, achieving seamless global navigation both indoors and outdoors. When using the weeding robot, operators only need to power it on; subsequent operations (such as automatically driving to the storage location or charging location) are completed automatically without human assistance. Furthermore, the magnetic wires also serve as boundary demarcation, preventing the weeding robot from wandering outside the base map during operation and causing damage such as bumps, collisions, or falls.

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit the technical solutions. Those skilled in the art should understand that any modifications or equivalent substitutions to the technical solutions of the present invention without departing from the spirit and scope of the present invention should be covered within the scope of the claims of the present invention.

Claims

1. A method for generating a working map for a weeding robot, characterized in that: An application is made to a weeding robot control system, the weeding robot control system including a weeding robot, a storage chamber and a back-end terminal for communicating with the weeding robot; the storage chamber has an opening for the weeding robot to enter and exit; a charging device is provided in the storage chamber; it also includes a magnetic wire, the magnetic wire starting from the charging device, running along the indoor boundary of the storage chamber and the outdoor working area boundary and returning to the charging device to form a closed area; the magnetic wire includes multiple long straight segments; The weeding robot includes a positioning module, a processing module, and electromagnetic sensors. The positioning module is used for real-time positioning, the electromagnetic sensors are used to collect magnetic data and send it to the processing module, and the processing module is used to generate a working map and plan and control the path of the weeding robot based on the positioning information and magnetic data. The weeding robot has two modes: a map generation mode and a weeding working mode. The backend stores a working map, which is a topological map formed based on a baseline map, an indoor map of the storage room, and an outdoor map of the working area. The baseline map is a closed area formed by magnetic wires. The method for generating the working map of the weeding robot includes: S1. After the weeding robot starts the map generation mode, control the weeding robot to circle around the magnetic wire, generate a virtual boundary based on the positioning information of the weeding robot, and obtain a reference map. The reference map is used to describe the closed area formed by the magnetic wire as the boundary. S2. Control the weeding robot to move in the storage room according to the preset working route, generate an indoor map based on the SLAM algorithm, and analyze the position of the weeding robot in the reference map in real time based on the data of the magnetic sensor during the movement. S3. Control the weeding robot to move in the work area according to the preset work route, and build an outdoor map by combining laser SLAM outdoor mapping technology; during the movement, the position of the weeding robot in the reference map is analyzed in real time based on the data of the magnetic sensor. S4. Using the position of the weeding robot in the baseline map as a correlation, the baseline map, indoor map and outdoor map are merged to obtain a topology map, which is then stored as the working map. In step S1, the virtual boundary generation process includes using the positioning data obtained when the weeding robot circles the magnetic conductor once as the reference positioning data of the magnetic conductor, and using the reference positioning data as the virtual boundary of the closed area formed by the magnetic conductor; in step S2, the real-time analysis of the position of the weeding robot on the reference map based on the data of the magnetic sensor includes analyzing the positional relationship between the weeding robot and each long straight segment of the magnetic conductor based on the data of the magnetic sensor and the current data of the magnetic conductor, obtaining the position of the weeding robot within the closed area formed by the magnetic conductor, and then analyzing the position of the weeding robot on the reference map.

2. The method for generating a working map for a weeding robot as described in claim 1, characterized in that: After S4 is executed, the following is also included: S5. When the weeding robot is working in weeding mode, the indoor and outdoor maps are updated in real time. S6. A gate mechanism is set at the intersection of the baseline map, indoor map and outdoor map. The gate mechanism is the update trigger mechanism for the working map. When the robot moves to the gate mechanism, the working map is updated according to the latest indoor map and outdoor map.

3. The method for generating a working map for a weeding robot as described in claim 1, characterized in that: Step S4 includes: S41. After rasterizing the reference map, extract the corresponding grids on the reference map for the driving trajectory of the weeding robot in S2 and S3, and mark them as candidate topology node sets. S42. Obtain the environmental information of each candidate topology node in the indoor or outdoor map, and determine whether the environmental information corresponding to each candidate topology node meets the preset topology conditions. If it meets the conditions, the corresponding grid area is taken as a topology node; otherwise, the corresponding grid area is taken as an edge. S43. Connect the topology nodes according to the preset connection method to obtain the topology map.

4. A control method for a weeding robot, characterized in that, The method for generating a working map for a weeding robot according to claim 1 includes: Step 1: After the weeding robot starts its weeding mode, the current position of the weeding robot is obtained through the electromagnetic sensor. Step 2: If the current location is inside the storage room, proceed to Step 3; if the current location is outside the storage room, proceed to Step 4. Step 3: Combining the working map and current location information, control the weeding robot to move to the passage opening of the storage room, update the working map, and then proceed to step 4. Step 4: Generate a working path by combining the working map and the current location information, and control the weeding robot to work according to the working path.

5. The weeding robot control method as described in claim 4, characterized in that: In step one, after controlling the weeding robot to start the weeding working mode, the current battery level of the weeding robot is also obtained; if the current battery level is lower than the preset minimum battery level, proceed to step five, otherwise proceed to step two. Step 5: After generating the charging path based on the work map and current location information, control the weeding robot to travel along the charging path to the charging device to charge; and continuously obtain the current charging power of the weeding robot. If the current charging power is greater than the preset minimum charging power, proceed to step 3.

6. The weeding robot control method as described in claim 4, characterized in that: In step four, the working path includes multiple target locations arranged in sequence; when controlling the weeding robot to work according to the working path, the optimal path planning between adjacent target locations is performed through DWA path planning.

7. The weeding robot control method as described in claim 4, characterized in that: In step four, while controlling the weeding robot to work according to the work path, it also performs real-time obstacle avoidance based on the detection data of the lidar.

Images