Cleaning robot control method and cleaning robot

Abstract

The invention relates to a cleaning robot control method and a cleaning robot. The method comprises the steps that under the condition that the cleaning robot is in a dormant state, task planning processing is conducted on the cleaning robot, a task execution strategy is obtained, a task execution instruction is monitored, and under the condition that the task execution instruction is received, the task execution strategy is executed; and controlling a robot body of the cleaning robot to execute the cleaning task according to the task execution strategy according to the task execution instruction. According to the method, when the cleaning robot is in the dormant state, it is represented that the cleaning robot is in the computing power idle time stage, task planning processing is conducted in the computing power idle time stage, the idle time computing power of the cleaning robot can be fully utilized, computing power idle waste is reduced, and the computing power utilization rate is increased.

Description

Technical Field

[0001] This application relates to the field of robotics, and in particular to a cleaning robot control method and a cleaning robot. Background Technology

[0002] With the advancement of science and technology, robots are being used more and more widely in people's daily lives and production.

[0003] Taking cleaning robots as an example, in related technologies, the computing power of cleaning robots is mainly used for online mapping and localization during the robot's movement, and is idle when it returns to the charging base to recharge.

[0004] However, the related technologies have resulted in a waste of computing power during off-peak hours and reduced the utilization rate of machine computing power. Summary of the Invention

[0005] Therefore, it is necessary to provide a cleaning robot control method and a cleaning robot to address the aforementioned technical problems.

[0006] In a first aspect, this application provides a cleaning robot control method, including:

[0007] While the cleaning robot itself is in a dormant state, task planning is performed on the cleaning robot to obtain a task execution strategy;

[0008] Listen for task execution instructions, and upon receiving such instructions, control the robot to perform cleaning tasks according to the task execution strategy.

[0009] In one embodiment, the task planning process includes path planning for at least one preset cleaning task; performing task planning processing on the cleaning robot to obtain a task execution strategy includes:

[0010] Obtain an offline scene map of the cleaning robot's driving scenario;

[0011] For each preset cleaning task, a path is planned based on the offline scene map to obtain the planned path, which serves as the task execution strategy.

[0012] In one embodiment, obtaining an offline scene map of the cleaning robot's driving scenario includes:

[0013] Acquire image data collected by the robot body in the driving scenario;

[0014] An offline scene map is obtained by performing 3D reconstruction based on the image data.

[0015] In one embodiment, path planning for a preset cleaning task is performed based on an offline scene map to obtain a planned path, including:

[0016] In the offline scene map, identify the area to be cleaned corresponding to the preset cleaning task;

[0017] Based on the offline scene map, a planned path is generated from the charging base to the area to be cleaned corresponding to the preset cleaning task, the cleaning operation is performed, and the path returns to the charging base.

[0018] In one embodiment, a planned path is generated based on an offline scene map, from the charging base to the area to be cleaned corresponding to a preset cleaning task, to perform the cleaning operation and return to the charging base, including:

[0019] Generate at least two simulated paths in the offline scene map for the robot body to start from the charging base, perform cleaning operations in the area to be cleaned, and return to the charging base;

[0020] The planned path is determined based on each simulated path.

[0021] In one embodiment, determining the planned path based on each simulated path includes:

[0022] Obtain the cleaning time of the robot body performing cleaning operations according to each simulated path;

[0023] The simulation path with the minimum cleaning time is determined from all simulation paths and used as the planned path.

[0024] In one embodiment, a planned path is generated based on an offline scene map, from the charging base to the area to be cleaned corresponding to a preset cleaning task, to perform the cleaning operation and return to the charging base, including:

[0025] Mark the areas to be cleaned and the charging base in the offline scene map;

[0026] The offline scene map, which marks the area to be cleaned and the charging base, is input into the path simulation model to obtain the planned path; the path simulation model is a pre-trained machine learning model.

[0027] In one embodiment, the method further includes:

[0028] Obtain the execution information of the target cleaning task indicated by the task execution command of the robot body;

[0029] If the execution information does not meet the preset requirements, the offline scene map of the cleaning robot's driving scenario will be updated.

[0030] In one embodiment, updating the offline scene map of the cleaning robot's driving scenario includes:

[0031] Determine the target cleaning area corresponding to the target cleaning task in the offline scene map;

[0032] As the robot moves to the target cleaning area, the image acquisition frequency is increased to obtain backup image data.

[0033] The offline scene map is updated based on the backup image data to obtain the updated offline scene map.

[0034] In one embodiment, before performing task planning on the cleaning robot to obtain a task execution strategy, the method further includes:

[0035] The available computing power resources of the cleaning robot are divided into a first computing power resource and a second computing power resource according to a preset allocation ratio; the second computing power resource is less than the first computing power resource.

[0036] The task planning process for the cleaning robot is performed using the first computing power resource to obtain the task execution strategy steps.

[0037] The steps for executing listening tasks and instructions using a second computing resource.

[0038] In one embodiment, before performing task planning on the cleaning robot to obtain a task execution strategy, the method further includes:

[0039] Obtain the remaining battery power of the robot body;

[0040] When the remaining battery power reaches full charge, the task planning process for the cleaning robot is executed to obtain the task execution strategy.

[0041] Secondly, this application also provides a cleaning robot, including a robot body and a charging base that communicate with each other. The robot body includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of any of the above methods.

[0042] In the aforementioned cleaning robot control method and cleaning robot, task planning is performed on the cleaning robot while it is in a dormant state to obtain a task execution strategy. Task execution commands are then monitored, and upon receiving a command, the robot body is controlled to execute the cleaning task according to the task execution strategy. In this method, the cleaning robot being in a dormant state indicates that it is in a period of idle computing power. Performing task planning during this idle period fully utilizes the robot's idle computing power, reduces wasted computing power, and improves computing power utilization. Attached Figure Description

[0043] Figure 1This is a schematic diagram of the cleaning robot in one embodiment;

[0044] Figure 2 This is a flowchart illustrating a cleaning robot control method in one embodiment;

[0045] Figure 3 This is a flowchart illustrating the process of obtaining a task execution strategy in one embodiment;

[0046] Figure 4 This is a schematic diagram illustrating the process of obtaining an offline scene map in one embodiment;

[0047] Figure 5 This is a flowchart illustrating the process of obtaining the planned path in one embodiment;

[0048] Figure 6 This is a flowchart illustrating the process of obtaining the planned path in another embodiment;

[0049] Figure 7 This is a flowchart illustrating the process of obtaining the planned path in another embodiment;

[0050] Figure 8 This is a flowchart illustrating the process of obtaining the planned path in another embodiment;

[0051] Figure 9 This is a schematic diagram illustrating the process of updating an offline scene map in one embodiment;

[0052] Figure 10 This is a schematic diagram illustrating the process of updating an offline scene map in another embodiment;

[0053] Figure 11 This is a flowchart illustrating the cleaning robot control method in another embodiment;

[0054] Figure 12 This is a flowchart illustrating the cleaning robot control method in another embodiment;

[0055] Figure 13 This is a flowchart illustrating the cleaning robot control method in another embodiment;

[0056] Figure 14 This is a structural block diagram of a cleaning robot control device in one embodiment. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0058] The cleaning robot control method provided in this application embodiment can be applied to, for example... Figure 1 The cleaning robot shown is a robot body 101 and a charging base 102 that communicate with each other. The robot body 101 includes a memory and a processor. The robot body 101 performs cleaning tasks, while the charging base 102 charges the robot body 101 and provides other auxiliary functions such as washing, drying, recycling, and water replenishment. The processor provides computing and control capabilities, and the memory stores data required and generated by the robot during charging. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The robot body 101 and the charging base 102 communicate via a communication interface in a wired or wireless manner. Wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), infrared, or other technologies. When the computer program is executed by the processor, it implements a cleaning robot control method.

[0059] It should be noted that the cleaning robots provided in this application embodiment are cleaning robots used for cleaning home and non-home scenarios, including but not limited to various cleaning robots such as sweeping robots, mopping robots, and automatic floor scrubbers.

[0060] Those skilled in the art will understand that Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0061] In one embodiment, such as Figure 2 As shown, a cleaning robot control method is provided, which is applied to... Figure 1 Taking a cleaning robot as an example, the processor used in the cleaning robot specifically includes the following steps:

[0062] S210. When the cleaning robot is in a dormant state, perform task planning processing on the cleaning robot to obtain a task execution strategy.

[0063] In this context, the cleaning robot is in a dormant state, indicating that the robot is back at its charging base and charging, and its processor is in a period of idle computing power. Task planning processing represents the planning and processing of the cleaning tasks to be performed by the cleaning robot. For example, this task planning processing can be sequential planning or path planning for multiple cleaning tasks, or path planning for a single cleaning task.

[0064] Optionally, the cleaning robot can pre-store multiple cleaning tasks input by the user. When the cleaning robot detects that it is in a dormant state, it can sequentially plan all the pre-executed cleaning tasks to obtain a task execution strategy that includes the execution order of each cleaning task. Alternatively, it can perform overall path planning for all the pre-executed cleaning tasks to obtain an overall planned path that includes the execution of all the cleaning tasks, which serves as the task execution strategy. The cleaning robot can also pre-store multiple preset cleaning tasks configured by the user. When it detects that it is in a dormant state, the cleaning robot can perform path planning for each preset cleaning task to obtain a planned path for each preset cleaning task, which serves as the task execution strategy.

[0065] S220: Listen for task execution instructions. Upon receiving a task execution instruction, control the cleaning robot body to perform the cleaning task according to the task execution strategy.

[0066] The task execution instruction is used to instruct the robot body within the cleaning robot to perform cleaning tasks. For example, this execution instruction can be a real-time instruction triggered by the user, or a timed instruction pre-set by the user.

[0067] Optionally, the cleaning robot can synchronously listen for task execution instructions. Upon receiving a task execution instruction, the robot body can control itself to execute cleaning tasks according to a planned task execution strategy. For example, if the task execution strategy includes an execution order / overall planned path for multiple pre-executed cleaning tasks, the robot body responds to the listened-for task execution instructions and executes each pre-executed cleaning task according to the execution order / overall planned path in the task execution strategy. If the task execution strategy includes a planned path for multiple preset cleaning tasks, the robot body can determine the target cleaning task to be executed based on the task execution instructions and execute the cleaning task according to the planned path corresponding to that target cleaning task in the task execution strategy.

[0068] In this embodiment, when the cleaning robot is in a dormant state, task planning is performed on the cleaning robot to obtain a task execution strategy, and task execution instructions are monitored. Upon receiving a task execution instruction, the robot body is controlled to execute the cleaning task according to the task execution strategy. In the above method, the cleaning robot being in a dormant state indicates that the cleaning robot is in a computing idle period. Performing task planning during this computing idle period can make full use of the cleaning robot's idle computing power, reduce computing power waste, and improve computing power utilization.

[0069] Task planning processing includes path planning for at least one preset cleaning task. Based on this, in one embodiment, such as... Figure 3 As shown, in S210 above, the task planning process for the cleaning robot is performed to obtain the task execution strategy, including:

[0070] S310. Obtain an offline scene map of the cleaning robot's driving scenario.

[0071] The offline scene map of the cleaning robot's driving scenario can be pre-stored in the robot itself, or it can be reconstructed when the cleaning robot is in a dormant state.

[0072] Optionally, the cleaning robot can directly obtain a pre-stored offline scene map of the driving scenario, or it can reconstruct the map based on the image data collected by the robot itself in the driving scenario to obtain an offline scene map of the driving scenario.

[0073] S320. For each preset cleaning task, the path planning for the preset cleaning task is performed based on the offline scene map to obtain the planned path, which is used as the task execution strategy.

[0074] The preset cleaning tasks are cleaning tasks pre-configured by the user. Each preset cleaning task corresponds to a cleaning area in the driving scenario. For example, the cleaning areas in the driving scenario include the master bedroom, living room, and kitchen. The user can configure three preset cleaning tasks for the cleaning robot: Task 1 – Clean the master bedroom; Task 2 – Clean the living room; Task 3 – Clean the living room and kitchen. The division of cleaning areas in the driving scenario and the configuration of preset cleaning tasks can be set by the user according to actual needs. The planned path for a preset cleaning task can be the robot's journey from the charging base to each cleaning area corresponding to the preset cleaning task, and the return path from the last cleaning area to the charging base.

[0075] Optionally, the cleaning robot can read a pre-stored offline scene map and receive user marking operations on the charging base and each area to be cleaned in the offline scene map, thus clarifying the location of the charging base and each area to be cleaned in the offline scene map. The cleaning robot can read various preset cleaning tasks, determine the location of the area to be cleaned corresponding to each preset cleaning task in the offline scene map, and plan the path corresponding to each preset cleaning task in the offline scene map.

[0076] For example, for each preset cleaning task, the cleaning robot can generate a planned path in an offline scene map from the charging base to the area to be cleaned corresponding to the preset cleaning task, and from the last area to be cleaned back to the charging base. This planned path can be the shortest travel path, the travel path with the shortest travel time, or the travel path with the fewest turns.

[0077] In this embodiment, when the task planning process includes path planning for at least one preset cleaning task, an offline scene map of the cleaning robot's driving scenario is obtained. For each preset cleaning task, path planning is performed based on the offline scene map to obtain a planned path, which serves as the task execution strategy. In this method, path planning is performed during the cleaning robot's idle computing time, reducing wasted computing power and improving computing power utilization.

[0078] In an alternative embodiment, such as Figure 4 As shown, S310 above, obtaining the offline scene map of the cleaning robot's driving scenario, includes:

[0079] S410. Acquire image data collected by the robot body in the driving scene.

[0080] Among them, the image data collected by the robot body in the driving scenario is the image data collected by the cleaning robot in the driving state (i.e., non-dormant state) while driving in the driving scenario.

[0081] Optionally, the robot body is equipped with an image acquisition device for collecting image data. While the cleaning robot is in motion, it can acquire image data through this device, and after the robot returns to its charging base and enters a dormant state, it can retrieve the image data collected by the image acquisition device during the driving scenario.

[0082] S420. Perform 3D reconstruction based on image data to obtain an offline scene map.

[0083] Optionally, after acquiring the image data, the cleaning robot can use computer vision algorithms to perform 3D reconstruction based on the acquired image data to obtain a 3D map of the driving scene, which can then serve as an offline scene map for the cleaning robot's driving scenario.

[0084] For example, the robot body can collect image data of the driving scene while performing cleaning tasks in the driving scene, and when the cleaning robot is in a dormant state, the collected image data is supplied to the processor of the cleaning robot for three-dimensional reconstruction.

[0085] It should be noted that, in order to improve the coverage of the offline scene map for the driving scene, the cleaning robot can reconstruct the offline scene map based on the image data collected by the robot body multiple times in the driving scene, until the robot body has driven through all areas of the driving scene, thus obtaining a complete offline scene map of the driving scene.

[0086] In this embodiment, image data collected by the robot body in the driving scene is acquired, and 3D reconstruction is performed based on the image data to obtain an offline scene map. In the above method, map reconstruction and path planning are performed simultaneously during the idle period of the cleaning robot's computing power, which improves the utilization of idle computing power and further improves the computing power utilization rate of the cleaning robot.

[0087] Different preset cleaning tasks correspond to different areas to be cleaned. In one embodiment, such as... Figure 5 As shown, in step S320 above, path planning for the preset cleaning task is performed based on the offline scene map to obtain the planned path, including:

[0088] S510. Determine the area to be cleaned corresponding to the preset cleaning task in the offline scene map.

[0089] Optionally, for each preset cleaning task, the cleaning robot can determine the area to be cleaned corresponding to the preset cleaning task based on the correspondence between the task and the area, and determine the location of the area to be cleaned corresponding to the preset cleaning task based on the area marking information in the offline scene map.

[0090] S520: Generate a planned path from the charging base to the area to be cleaned corresponding to the preset cleaning task based on the offline scene map, perform the cleaning operation, and return to the charging base.

[0091] Optionally, after identifying the area to be cleaned corresponding to the preset cleaning task in the offline scene map, the cleaning robot can generate a planned path from the charging base to the area to be cleaned corresponding to the preset cleaning task and return to the charging base based on the location of the charging base and the location of the area to be cleaned corresponding to the preset cleaning task in the offline scene map.

[0092] In this embodiment, the area to be cleaned corresponding to the preset cleaning task is determined in the offline scene map, and a planned path is generated based on the offline scene map to go from the charging base to the area to be cleaned corresponding to the preset cleaning task, perform the cleaning operation, and return to the charging base. In the above method, the complete path planning of the robot body from the charging base to the area to be cleaned corresponding to the preset cleaning task, perform the cleaning operation, and return to the charging base is realized based on the offline scene map, which improves the comprehensiveness of the path planning.

[0093] The planned path for the preset cleaning task is obtained by selecting from multiple simulated paths. Based on this, in one embodiment, such as Figure 6 As shown, the above-mentioned S520, which generates a planned path from the charging base to the area to be cleaned corresponding to the preset cleaning task based on the offline scene map, performs the cleaning operation, and returns to the charging base, includes:

[0094] S610. Generate at least two simulated paths in the offline scene map for the robot body to start from the charging base, go to the area to be cleaned to perform cleaning operations, and return to the charging base.

[0095] Optionally, after the cleaning robot has clearly identified the location of the area to be cleaned and the charging base corresponding to the preset cleaning task in the offline scene map, it can generate multiple simulated paths from the charging base to the area to be cleaned to perform the cleaning operation and return to the charging base.

[0096] Each simulated path can be divided into three sub-paths: a first sub-path from the charging base to the area to be cleaned, a second sub-path performing the cleaning operation in the area to be cleaned, and a third sub-path returning from the last area to be cleaned to the charging base. For multiple simulated paths of the same preset cleaning task, at least one of the first, second, and third sub-paths can be different.

[0097] It should be noted that there are no specific limitations on how the cleaning robot generates multiple simulated paths from the charging base to the area to be cleaned, performs the cleaning operation, and returns to the charging base. The paths can be generated randomly or according to a preset requirement algorithm. The preset requirements can be the fewest turns, the longest straight-line length, etc. For each preset cleaning task, at least two simulated paths need to be obtained.

[0098] S620. Determine the planned path based on each simulated path.

[0099] Optionally, after obtaining at least two simulated paths for each preset cleaning task, the cleaning robot can select the planned path for the corresponding preset cleaning task from these at least two simulated paths.

[0100] Cleaning time is a primary criterion for screening. Therefore, in one embodiment, such as Figure 7 As shown, S620 above, determining the planned path based on each simulated path, includes:

[0101] S710. Obtain the cleaning time of the robot body performing cleaning operations according to each simulated path.

[0102] Optionally, for each simulated path, the cleaning robot can perform path analysis to determine the cleaning time required for the robot body to perform cleaning operations according to each simulated path.

[0103] For example, the cleaning robot can determine the total length of the simulated path and the number of turns in the simulated path. Based on the robot's cleaning speed and the time taken for a single turn, it determines the travel time for each simulated path and uses this as the cleaning time for the robot to perform cleaning operations along the simulated path. For instance, for a simulated path with a total length of s, a number of turns of m, a movement speed of v, and a time taken for a single turn of t, the travel time for that simulated path, i.e., the cleaning time, is calculated as T = s / v + m*t.

[0104] S720. Determine the simulation path with the minimum cleaning time among all simulation paths, and use it as the planned path.

[0105] Optionally, after obtaining the cleaning time of each simulated path corresponding to the preset cleaning task, the cleaning robot can compare the cleaning time of each simulated path and determine the simulated path with the minimum cleaning time as the planning path for the corresponding preset cleaning task.

[0106] In this embodiment, at least two simulated paths are generated in the offline scene map, showing the robot body starting from the charging base, performing cleaning operations in the area to be cleaned, and returning to the charging base. A planned path is then determined based on each simulated path. Specifically, the cleaning time of the robot body performing cleaning operations according to each simulated path is obtained, and the simulated path with the shortest cleaning time is determined from all simulated paths as the planned path. In the above method, for each preset cleaning task, a planned path for the corresponding preset cleaning task is obtained by filtering from multiple corresponding simulated paths. Specifically, selecting the simulated path with the shortest cleaning time as the planned path can reduce subsequent cleaning time and correspondingly improve the execution efficiency of the cleaning task.

[0107] There are several ways to generate a planned path. In one alternative embodiment, such as... Figure 8 As shown, the above-mentioned S520, which generates a planned path from the charging base to the area to be cleaned corresponding to the preset cleaning task based on the offline scene map, performs the cleaning operation, and returns to the charging base, includes:

[0108] S810 marks the areas to be cleaned and the charging base in the offline scene map.

[0109] Optionally, the cleaning robot marks the areas to be cleaned and the locations of the charging bases corresponding to preset cleaning tasks on an offline map. To distinguish between the areas to be cleaned and the charging bases, the cleaning robot can use different markers to mark their locations on the offline scene map. For example, the cleaning robot can label the location of the charging base as "S" and the area to be cleaned as "O". Different areas to be cleaned can use different markers, such as O1 for the master bedroom, O2 for the living room, O3 for the kitchen, and so on.

[0110] S820: Input the offline scene map marking the area to be cleaned and the charging base into the path simulation model to obtain the planned path; the path simulation model is a pre-trained machine learning model.

[0111] The path simulation model can be used to simulate the robot's travel path, generating a planned path that meets design requirements based on the locations of the areas to be cleaned and the charging base marked in the offline scene map. For example, this path simulation model can be a neural network model trained using a large number of scene maps marked with the locations of the areas to be cleaned and the charging base as training samples.

[0112] Optionally, the cleaning robot is equipped with a pre-trained path simulation model. After obtaining an offline scene map that marks the area to be cleaned and the charging base corresponding to the preset cleaning task, the cleaning robot can input the marked offline scene map into the path simulation model, and the path simulation model can output the planned path corresponding to the preset cleaning task.

[0113] In this embodiment, the area to be cleaned and the charging base are marked on an offline scene map. The offline scene map with the marked areas to be cleaned and the charging base is then input into a path simulation model to obtain a planned path. The path simulation model is a pre-trained machine learning model. In the above method, the path simulation model is trained based on the user's actual needs, which improves the matching degree between the obtained planning results and the user's actual needs, thus improving the accuracy and reliability of the planning results.

[0114] The above method may also include an update process for the offline scene map, in one embodiment, such as Figure 9 As shown, the above method also includes:

[0115] S910: Obtain the execution information of the target cleaning task indicated by the task execution command of the robot body.

[0116] The execution information of the target cleaning task may include the execution result of the robot body on the target cleaning task, such as successful execution or failure. In the case of successful execution, it may also include the execution time of the robot body on the target cleaning task.

[0117] Optionally, when the robot returns to its charging base, the cleaning robot can obtain execution information of the target cleaning task indicated by the task execution command. For example, the robot can record the execution result of the target cleaning task, such as success or failure. In the case of success, it can simultaneously record the total time from receiving the execution command for the target cleaning task to completing the task and returning to the charging base, as the execution time, and generate execution information for the processor to access.

[0118] S920. If the execution information does not meet the preset requirements, update the offline scene map of the cleaning robot's driving scenario.

[0119] Optionally, upon obtaining the execution information for the target cleaning task, the cleaning robot can compare the execution information with preset requirements to determine whether the execution information meets the preset requirements. If the preset requirements are not met, the offline scene map of the cleaning robot's driving scenario can be updated.

[0120] For example, the preset requirements include the required execution result, such as successful execution, and may also include an execution time threshold for successful execution. If the execution result in the obtained execution information is execution failure, the cleaning robot can determine that the execution information does not meet the preset requirements. Alternatively, if the execution result in the obtained execution information is execution failure, but the execution time is greater than the execution time threshold, the cleaning robot can determine that the execution information does not meet the preset requirements. Conversely, if the execution result in the obtained execution information is execution success, but the execution time is less than or equal to the execution time threshold, the cleaning robot can determine that the execution information meets the preset requirements.

[0121] It should be noted that, in the case of path planning, if the execution information does not meet the preset requirements, indicating that the planned path determined by the system did not achieve the expected results, the reason lies in the discrepancy between the constructed offline scene map and the actual driving scenario. This could be due to inaccurate mapping or changes in the scene. Therefore, the offline scene map needs to be updated.

[0122] Offline maps can be updated based on newly acquired image data. In one embodiment, such as... Figure 10 As shown, the offline scene map update for the cleaning robot's driving scenario in S920 includes:

[0123] S1010. Determine the target cleaning area corresponding to the target cleaning task in the offline scene map.

[0124] Optionally, if the robot's execution information for the target cleaning task does not meet preset requirements, the cleaning robot can determine the corresponding cleaning area for the target cleaning task in the offline scene map as the target cleaning area. For example, continuing the above example, if the target cleaning task is task 3 (cleaning the living room and kitchen), the cleaning robot will use the areas to be cleaned in the living room and kitchen in the offline scene map as the target cleaning area.

[0125] S1020. When the robot body subsequently moves to the target cleaning area, increase the image acquisition frequency to obtain backup image data.

[0126] Optionally, the cleaning robot can mark the target cleaning area in the offline scene map, and when the robot body needs to go to the target cleaning area to perform subsequent cleaning tasks, the image acquisition frequency can be increased to re-acquire more image data during the process of the robot body going from the charging base to the target cleaning area to perform cleaning operations and returning to the charging base, as backup image data.

[0127] S1030. Update the offline scene map based on the backup image data to obtain the updated offline scene map.

[0128] Optionally, after obtaining more abundant backup image data, the cleaning robot can perform 3D reconstruction based on the backup image data to obtain a local scene map, perform feature matching between the local scene map and the offline scene map, and then use the local scene map to replace the matching area in the offline scene region to obtain an updated offline scene map.

[0129] In this embodiment, by acquiring the execution information of the target cleaning task indicated by the robot's task execution command, the offline scene map of the cleaning robot's driving scenario is updated when the execution information does not meet preset requirements. Specifically, the target cleaning area corresponding to the target cleaning task can be determined in the offline scene map. When the robot subsequently moves to the target cleaning area, the image acquisition frequency is increased to obtain backup image data. The offline scene map is then updated based on the backup image data, resulting in an updated offline scene map. In the above method, the offline scene map is updated when the execution information of the target cleaning task does not achieve the expected results, identifying and addressing weak points in cleaning. It can also adapt to scene changes. Furthermore, updating the offline scene map using backup image data obtained by increasing the acquisition frequency further refines the offline scene map, thereby improving the fidelity of the offline 3D map to the actual driving scenario and contributing to improved subsequent cleaning effects and efficiency.

[0130] To synchronously monitor task execution instructions, in one embodiment, such as Figure 11 As shown, the above method also includes:

[0131] S1110. According to the preset allocation ratio, the available computing power resources of the cleaning robot are divided into the first computing power resource and the second computing power resource; the second computing power resource is less than the first computing power resource.

[0132] Optionally, before executing S210 above, the cleaning robot can pre-read the processor's available computing power resources and divide the available computing power resources according to a preset allocation ratio to obtain a larger first computing power resource and a smaller second computing power resource. For example, the preset allocation ratio can be preset based on user needs; for instance, the preset allocation ratio can be 4:1, and the available computing power resource C can be divided into a first computing power resource C1 = 4 / 5 * C and a second computing power resource C1 = 1 / 5 * C.

[0133] S1120, The step of performing task planning processing on the cleaning robot through the first computing power resource to obtain the task execution strategy.

[0134] Optionally, after allocating the first computing power resource and the second computing power resource, the cleaning robot can use the larger first computing power resource to perform task planning processing to obtain the task execution strategy.

[0135] S1130, the steps of executing the listening task execution instruction through the second computing power resource.

[0136] Optionally, after the first and second computing resources are allocated, the cleaning robot can use the smaller second computing resource to execute the listening task execution instructions.

[0137] It should be noted that executing the planning path requires a large amount of computing power, while executing the listening task only requires a small amount of computing power. Therefore, a larger amount of primary computing power can be used to execute the planning path, while a smaller amount of secondary computing power can be used to execute the listening task, which improves the rationality and utilization of computing power allocation.

[0138] In this embodiment, the available computing power resources of the cleaning robot are divided into a first computing power resource and a second computing power resource according to a preset allocation ratio. The first computing power resource is used to perform task planning processing on the cleaning robot to obtain a task execution strategy, and the second computing power resource is used to listen for task execution instructions. The second computing power resource is smaller than the first computing power resource. In this method, the allocation of computing power resources enables the synchronous execution of task planning processing and listening for task execution instructions, improving both computing power utilization and response efficiency to task execution instructions, thereby increasing cleaning efficiency.

[0139] To improve the lifespan of the cleaning robot, in one embodiment, such as Figure 12 As shown, the above method also includes:

[0140] S1210: Obtain the remaining battery power of the robot body.

[0141] Optionally, before performing task planning for the cleaning robot, the cleaning robot can periodically acquire its remaining power (State of Charge, SOC). The charging base charges the robot while it is in sleep mode, and the robot's SOC increases with charging time until it reaches 100%, indicating a full charge.

[0142] S1220. When the remaining power reaches full charge, perform task planning processing on the cleaning robot to obtain the task execution strategy.

[0143] Optionally, for each SOC acquired, the cleaning robot can determine whether the SOC has reached full charge. If it has reached full charge, the robot performs task planning to obtain a task execution strategy, i.e., S210 above. Subsequently, the robot can continue to execute the following steps, i.e., S220.

[0144] It should be noted that the processor of the cleaning robot continuously generates heat, causing the robot body to heat up. The robot itself also generates heat during charging due to the charging current. If the processor runs simultaneously with the robot body charging, it will increase heat generation, leading to high temperatures that not only affect the robot's lifespan but also pose safety hazards. However, once the robot body is fully charged, the charging current decreases significantly, and heat generation decreases accordingly. Therefore, running the processor to execute steps S210-S220 after the robot body is fully charged helps to disperse heat, improving the cleaning robot's lifespan and safety.

[0145] In this embodiment, the remaining battery power of the robot is obtained, and when the remaining battery power reaches full charge, task planning is performed on the cleaning robot to obtain a task execution strategy. In this method, the high-power charging process is staggered from the processor operation, allowing heat generation from the robot body to be dispersed, thus improving the service life and safety of the cleaning robot.

[0146] To facilitate understanding by those skilled in the art, the cleaning robot control method provided in this application is described in detail below, such as... Figure 13 As shown, the method may include:

[0147] S1301. When the cleaning robot is in a dormant state, obtain the remaining power of the robot body;

[0148] S1302. When the remaining power reaches full charge, the available computing power resources of the cleaning robot are divided into first computing power resources and second computing power resources according to a preset allocation ratio; the second computing power resources are less than the first computing power resources.

[0149] S1303. Using the first computing power resource, perform three-dimensional reconstruction based on the image data collected by the cleaning robot in the driving scene to obtain an offline scene map of the driving scene;

[0150] S1304. For each preset cleaning task, determine the area to be cleaned corresponding to the preset cleaning task in the offline scene map;

[0151] S1305. Generate at least two simulated paths in the offline scene map for the robot body to start from the charging base, perform cleaning operations in the area to be cleaned, and return to the charging base.

[0152] S1306. Obtain the cleaning time of the robot body performing cleaning operations according to each simulated path;

[0153] S1307. Determine the simulation path with the minimum cleaning time among all simulation paths and use it as the planned path;

[0154] S1308. Listen to the task execution instructions through the second computing power resources, and when the task execution instructions are received, control the robot body to execute the target cleaning task according to the planned path of the target cleaning task indicated by the corresponding task execution instructions.

[0155] S1309. Obtain the execution information of the robot body performing the target cleaning task;

[0156] S1310. If the execution information does not meet the preset requirements, determine the target cleaning area corresponding to the target cleaning task in the offline scene map.

[0157] S1311. When the robot body subsequently moves to the target cleaning area, increase the image acquisition frequency to obtain backup image data;

[0158] S1312. Update the offline scene map based on the backup image data to obtain the updated offline scene map.

[0159] It should be noted that the descriptions in S1301-S1312 above can be found in the relevant descriptions in the above embodiments, and their effects are similar, so they will not be repeated here.

[0160] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0161] Based on the same inventive concept, this application also provides a cleaning robot control device for implementing the cleaning robot control method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more charging device embodiments for cleaning robots provided below can be found in the limitations of the cleaning robot control method described above, and will not be repeated here.

[0162] In one embodiment, such as Figure 14 As shown, a cleaning robot control device is provided, including: a task planning module 1401 and a strategy execution module 1402, wherein:

[0163] The task planning module 1401 is used to perform task planning for the cleaning robot when the cleaning robot is in a dormant state, and to obtain a task execution strategy.

[0164] The strategy execution module 1402 is used to listen for task execution instructions. Upon receiving a task execution instruction, it controls the robot body of the cleaning robot to perform the cleaning task according to the task execution strategy.

[0165] Each module in the aforementioned cleaning robot control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0166] In one embodiment, a cleaning robot is provided, including a robot body and a charging base that communicate with each other. The robot body includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of any of the above-described cleaning robot control methods.

[0167] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-described cleaning robot control methods.

[0168] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of any of the above-described cleaning robot control methods.

[0169] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0170] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0171] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A cleaning robot control method, characterized by, The method comprises: In the case that the cleaning robot is in a dormant state, performing task planning processing on the cleaning robot to obtain a task execution strategy; Listening to a task execution instruction, and in the case that the task execution instruction is received, controlling a robot body of the cleaning robot to execute a cleaning task according to the task execution strategy.

2. The method of claim 1, wherein, The task planning processing comprises path planning for at least one preset cleaning task; the task planning processing on the cleaning robot to obtain a task execution strategy comprises: Obtaining an offline scene map of a driving scene of the cleaning robot; For each of the preset cleaning tasks, performing path planning for the preset cleaning task according to the offline scene map to obtain a planned path as the task execution strategy.

3. The method of claim 2, wherein, The obtaining of the offline scene map of the driving scene of the cleaning robot comprises: Obtaining image data collected by the robot body in the driving scene; Performing three-dimensional reconstruction according to the image data to obtain the offline scene map.

4. The method of claim 2, wherein, The path planning for the preset cleaning task according to the offline scene map to obtain a planned path comprises: Determining a to-be-cleaned area corresponding to the preset cleaning task in the offline scene map; Generating, according to the offline scene map, a planned path from a charging base to the to-be-cleaned area corresponding to the preset cleaning task to execute a cleaning operation and back to the charging base.

5. The method of claim 4, wherein, The generating, according to the offline scene map, of a planned path from the charging base to the to-be-cleaned area corresponding to the preset cleaning task to execute a cleaning operation and back to the charging base comprises: Generating, in the offline scene map, at least two simulation paths of the robot body starting from the charging base, going to the to-be-cleaned area to execute a cleaning operation and returning to the charging base; Determining the planned path according to each of the simulation paths.

6. The method of claim 5, wherein, The determining of the planned path according to each of the simulation paths comprises: Obtaining cleaning time consumption of the robot body executing a cleaning operation according to each of the simulation paths; Determining, among all the simulation paths, a simulation path with minimum cleaning time consumption as the planned path.

7. The method of claim 4, wherein, The generating, according to the offline scene map, of a planned path from the charging base to the to-be-cleaned area corresponding to the preset cleaning task to execute a cleaning operation and back to the charging base comprises: Marking the to-be-cleaned area and the charging base in the offline scene map; Inputting the offline scene map marking the to-be-cleaned area and the charging base into a path simulation model to obtain the planned path; the path simulation model is a machine learning model obtained by pre-training.

8. The method according to any one of claims 1 to 7, characterized in that, The method further comprises: Obtaining execution information of a target cleaning task indicated by the task execution instruction executed by the robot body; In the case that the execution information does not meet a preset requirement, updating the offline scene map of the driving scene of the cleaning robot.

9. The method of claim 8, wherein, The updating of the offline scene map of the driving scene of the cleaning robot comprises: Determining a target cleaning area corresponding to the target cleaning task in the offline scene map; In a case where the robot body subsequently goes to the target cleaning area, the image acquisition frequency is increased to obtain backup image data; The offline scene map is updated according to the backup image data to obtain an updated offline scene map.

10. The method according to any one of claims 1-7, characterized in that, Before the task planning process of the cleaning robot is performed to obtain a task execution strategy, the method further comprises: According to a preset allocation ratio, available computing resource of the cleaning robot is divided into first computing resource and second computing resource; the second computing resource is less than the first computing resource; The task planning process of the cleaning robot is performed by using the first computing resource to obtain a task execution strategy; The step of listening to the task execution instruction is performed by using the second computing resource.

11. The method according to any one of claims 1-7, characterized in that, Before the task planning process of the cleaning robot is performed to obtain a task execution strategy, the method further comprises: The remaining power of the robot body is obtained; In a case where the remaining power reaches full power, the task planning process of the cleaning robot is performed to obtain a task execution strategy.

12. A cleaning robot comprising a robot body and a charging base in communication with each other, the robot body comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to implement the steps of the method in any one of claims 1 to 11.

Images