Offline map-based rapid update method, device, equipment and storage medium

Abstract

The present application provides a method, apparatus, device and storage medium for rapid updating based on offline maps. By combining a preset number of frames of first lidar data collected for the current scene on the basis of the initial offline map, the initial offline map can be rapidly updated, thereby more efficiently identifying environmental changes that have occurred in the current scene relative to the initial offline map, thereby facilitating subsequent cleaning path planning and obstacle avoidance movements of the sweeper.

Description

Technical Field

[0001] The present application relates to the technical field of map updating, and in particular to a method, apparatus, device and storage medium for quickly updating an offline map. Background Art

[0002] In the application of laser SLAM in sweepers, the created map is usually reused. Each time the sweeper starts a new cleaning, it repositions directly on the created map, eliminating the need to re-create the map and saving time.

[0003] However, the actual scene environment may differ from the pre-built map during use. For example, if new furniture is added to the scene or its placement is changed, the robot vacuum will not be able to avoid obstacles if it plans the cleaning path directly according to the pre-built map. If the robot vacuum cleaner re-builds the map at the beginning of each cleaning, it will take a lot of time and reduce cleaning efficiency. Summary of the Invention

[0004] The main purpose of this application is to provide a method, device, equipment and storage medium for rapid updating based on offline maps, aiming to solve the problem that a sweeper cannot quickly update the current map when there is a difference between the existing actual scene and the pre-built map.

[0005] To achieve the above objectives, the present application provides a rapid update method based on an offline map, which is applied to a sweeper. The rapid update method includes:

[0006] Loading an initial offline map, and relocating the sweeping robot to obtain a first current position of the sweeping robot on the initial offline map;

[0007] Collecting a preset number of frames of first laser radar data, and constructing a map based on the first current position and an initial probability grid map of the initial offline map to obtain a first current probability grid map;

[0008] Comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determining whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map;

[0009] If the grid coordinate area with the reduced grid probability value exists in the first current probability grid map, setting the center grid coordinate of the grid coordinate area to a blank position;

[0010] If the grid coordinate area with the increased grid probability value exists in the first current probability grid map, the center grid coordinates of the grid coordinate area are set as the obstacle position.

[0011] The present application also provides a rapid update device based on an offline map, which is applied to a sweeper. The rapid update device includes:

[0012] a repositioning module, configured to load an initial offline map and reposition the sweeping robot to obtain a first current position of the sweeping robot on the initial offline map;

[0013] A first mapping module is configured to collect a preset number of frames of first lidar data, and to perform mapping based on the first current position in combination with an initial probability grid map of the initial offline map to obtain a first current probability grid map;

[0014] A first judgment module is configured to compare the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and to judge whether there is a grid coordinate region with a changed grid probability value in the first current probability grid map;

[0015] a first setting module, configured to set the center grid coordinates of the grid coordinate area to a blank position if the grid coordinate area with a reduced grid probability value exists in the first current probability grid map;

[0016] The second setting module is configured to set the center grid coordinates of the grid coordinate area as the obstacle position if the grid coordinate area with the increased grid probability value exists in the first current probability grid map.

[0017] The present application also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any of the above methods when executing the computer program.

[0018] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0019] The present application provides a method, device, equipment and storage medium for rapid updating based on offline maps. When a sweeper starts cleaning, it first loads the initial offline map and repositions the sweeper to obtain the first current position of the sweeper in the initial offline map. Then, a preset number of frames of first lidar data are collected, and a map is constructed based on the first current position combined with the initial probability grid map of the initial offline map to obtain a first current probability grid map. The first current probability grid map and the initial probability grid map are then compared one by one according to the corresponding grid coordinates to determine whether there is a grid coordinate area with a reduced grid probability value in the first current probability grid map. If there is a grid coordinate area with a reduced grid probability value in the first current probability grid map, the center grid coordinate of the grid coordinate area is set to a blank position. If there is a grid coordinate area with an increased grid probability value in the first current probability grid map, the center grid coordinate of the grid coordinate area is set to the obstacle position. This application achieves rapid updating of the initial offline map by combining a preset number of frames of first lidar data collected for the current scene on the basis of the initial offline map, thereby quickly identifying environmental changes that have occurred in the current scene relative to the initial offline map, facilitating subsequent cleaning path planning and obstacle avoidance movements of the sweeper. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a schematic diagram of the steps of a fast update method based on an offline map in an embodiment of the present application;

[0021] Figure 2 This is a block diagram of the overall structure of a fast update device based on an offline map in an embodiment of the present application;

[0022] Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present application.

[0023] The realization of the objectives, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0025] Reference Figure 1 In one embodiment of the present application, a rapid update method based on an offline map is provided, which is applied to a sweeper. The rapid update method includes:

[0026] S1: loading an initial offline map and relocating the sweeper to obtain a first current position of the sweeper in the initial offline map;

[0027] S2: Collect a preset number of frames of first laser radar data, and build a map based on the first current position in combination with the initial probability grid map of the initial offline map to obtain a first current probability grid map;

[0028] S3: comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determining whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map;

[0029] S4: if the grid coordinate area with the reduced grid probability value exists in the first current probability grid map, setting the center grid coordinate of the grid coordinate area to a blank position;

[0030] S5: If the grid coordinate area with the increased grid probability value exists in the first current probability grid map, the center grid coordinates of the grid coordinate area are set as the obstacle position.

[0031] In this embodiment, the rapid update method is applied to a robot vacuum as an example for specific explanation. The robot vacuum's control system first loads a pre-built initial offline map, which is obtained by the robot vacuuming through mapping the current scene. After loading the initial offline map, the control system repositions the robot vacuuming based on reference objects around its current location or using GPS positioning technology to obtain the robot vacuuming's first current position in the initial offline map, i.e., its position in the current scene. The robot vacuuming is equipped with a radar laser device. The control system uses the radar laser device to collect a preset number of frames of first lidar data (the preset number of frames is preferably 5 frames, but can be reduced to 2 or 3 frames depending on actual needs, but the preset number of frames is far less than the number of laser data frames required for remapping). Using the robot vacuuming's first current position as a position reference, the control system determines the area of ​​the initial offline map that needs to be updated. The preset number of frames of lidar data are then inserted into the initial grid map of the initial offline map for mapping, thereby obtaining a first current probabilistic grid map that matches the actual environment of the current scene.

[0032] The control system will use the initial probability grid map as the comparison benchmark, and compare the grid probability values ​​of the first current probability grid map and the initial probability grid map one by one according to the corresponding grid coordinates (the probability grid map has multiple grid coordinates, and a single grid coordinate corresponds to one grid probability value), so as to determine whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map, wherein the layout of the grid coordinate area is in a nine-square grid style. Specifically, the control system compares the grid probability value corresponding to each grid coordinate of the first current probability grid map and the initial probability grid map according to the corresponding grid coordinates (for example, grid coordinate A of the first current probability grid map corresponds to grid probability value A, and grid coordinate B corresponds to grid probability value B; grid coordinate a of the initial probability grid map corresponds to grid probability value a, and grid coordinate b corresponds to grid probability value b; grid coordinate A corresponds to grid coordinate a, then grid probability value A is compared with grid probability value a; grid coordinate B corresponds to grid coordinate b, then grid probability value B is compared with grid probability value b), so as to determine whether there is a first grid coordinate in the first current probability grid map whose first grid probability value is reduced (that is, for the same first grid coordinate, the grid probability value in the first current probability grid map is lower than the grid probability value in the initial probability grid map). If there are one or more grid coordinates in the first current probability grid map where the first grid probability value is reduced, namely the first grid coordinates, then a determination is made based on the initial probability grid map whether the second grid probability values ​​corresponding to all (i.e., eight) second grid coordinates adjacent to the single first grid coordinate in the first current probability grid map are reduced. If the second grid probability values ​​corresponding to the eight second grid coordinates adjacent to the first grid coordinate in the first current probability grid map are all reduced relative to the grid probability values ​​of the same grid coordinates in the initial probability grid map, then it is determined that a grid coordinate region with reduced grid probability values ​​exists in the first current probability grid map.

[0033] In another embodiment, if one or more first grid coordinates have decreased first grid probability values ​​in the first current probability grid map, a determination is made based on the initial probability grid map whether at least one of the second grid probability values ​​corresponding to the eight second grid coordinates adjacent to the single first grid coordinate in the first current probability grid map has decreased. If at least one of the second grid probability values ​​corresponding to the eight second grid coordinates adjacent to the first grid coordinate in the first current probability grid map has decreased relative to the grid probability value of the same grid coordinate in the initial probability grid map, then a grid coordinate region with decreased grid probability values ​​is determined to exist in the first current probability grid map.

[0034] In another embodiment, the control system compares the first current probability grid map and the initial probability grid map one by one according to the corresponding grid coordinates, and determines whether a third grid coordinate corresponding to a third grid probability value having a preset variation form exists in the first current probability grid map, where the preset variation form indicates that the grid probability value increases from zero to non-zero. If the third grid coordinate corresponding to the third grid probability value having the preset variation form exists in the first current probability grid map, then, based on the initial probability grid map, a determination is made as to whether the fourth grid probability values ​​corresponding to each of the eight fourth grid coordinates adjacent to the first grid coordinate all have the preset variation form. If the fourth grid probability values ​​corresponding to each of the eight fourth grid coordinates adjacent to the first grid coordinate all have the preset variation form, then a determination is made that a grid coordinate region with increasing grid probability values ​​exists in the first current probability grid map.

[0035] If there is a grid coordinate area with a reduced grid probability value in the first current probability grid map, it means that in the initial offline map of the first mapping, the position corresponding to the central grid coordinate of the grid coordinate area may be placed with an obstacle, and after the current mapping is performed based on the first lidar data of a preset number of frames, that is, in the real-time map corresponding to the first current probability grid map, the position corresponding to the central grid coordinate is a blank position (the original obstacle is removed). Therefore, the control system sets the central grid coordinate of the grid coordinate area (the central grid coordinate represents a grid coordinate located at the center of the grid coordinate area) to a blank position in the first current probability grid map, and the subsequent sweeper can pass through this blank position when performing path planning (that is, the sweeper does not need to avoid it).

[0036] If a grid coordinate region with an elevated grid probability value exists in the first current probability map, this indicates that the location corresponding to the center grid coordinate of the grid coordinate region was blank in the initial offline map constructed during initial mapping. However, after the current map is constructed based on the preset number of frames of first lidar data, that is, in the real-time map corresponding to the first current probability grid map, the location corresponding to the center grid coordinate has an obstacle. Therefore, the control system sets the center grid coordinate of the grid coordinate region in the first current probability grid map as the obstacle location, and the sweeper must avoid this obstacle during subsequent path planning.

[0037] This embodiment achieves rapid updating of the initial offline map by combining a preset number of frames of first lidar data collected for the current scene on the basis of the initial offline map, thereby more efficiently identifying environmental changes that have occurred in the current scene relative to the initial offline map, thereby facilitating subsequent cleaning path planning and obstacle avoidance movements of the sweeper.

[0038] Furthermore, the step of comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates to determine whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map includes:

[0039] S301: Compare the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determine whether there is a first grid coordinate with a reduced first grid probability value in the first current probability grid map;

[0040] S302: If the first grid coordinate having a reduced first grid probability value exists in the first current probability grid map, determining, based on the initial probability grid map, whether the second grid probability values ​​corresponding to all second grid coordinates adjacent to the first grid coordinate in the first current probability grid map are reduced;

[0041] S303: If the second grid probability values ​​corresponding to all the second grid coordinates adjacent to the first grid coordinate are reduced, it is determined that the grid coordinate area with the reduced grid probability value exists in the first current probability grid map.

[0042] In this embodiment, the control system uses the initial probability grid map as a reference and the same grid coordinates as the corresponding basis to compare the grid probability value of each grid coordinate in the first current probability grid map with the grid probability value of each grid coordinate in the initial probability grid map one by one, thereby determining whether there are one or more first grid coordinates in the first current probability grid map whose first grid probability value is reduced (i.e., the first grid probability value of the first grid coordinate in the first current probability grid map is less than the grid probability value of the first grid coordinate in the initial probability grid map). If one or more first grid coordinates in the first current probability grid map have a reduced first grid probability value, then the eight second grid coordinates adjacent to the first grid coordinate are obtained (the first grid coordinate and the eight second grid coordinates constitute a grid coordinate region, and the first grid coordinate is located at the center of the grid coordinate region). The second grid probability values ​​corresponding to each of the eight second grid coordinates in the first current probability grid map and the grid probability values ​​corresponding to each of the eight second grid coordinates in the initial probability grid map are obtained. Using the same second grid coordinate as a comparison benchmark, it is determined whether each second grid probability value is less than the corresponding grid probability value in the initial probability grid map. If each second grid probability value is less than the corresponding grid probability value in the initial probability grid map, it is determined that the second grid probability values ​​corresponding to each of the eight second grid coordinates adjacent to the first grid coordinate have all decreased, and further, it is determined that a grid coordinate region with reduced grid probability values ​​exists in the first current probability grid map.

[0043] Furthermore, the step of comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates to determine whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map further includes:

[0044] S304: Comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determining whether there is a third grid coordinate corresponding to a third grid probability value having a preset change form in the first current probability grid map, where the preset change form indicates that the grid probability value increases from zero to non-zero;

[0045] S305: If the third grid coordinate corresponding to the third grid probability value having the preset variation exists in the first current probability grid map, determining, based on the initial probability grid map, whether the fourth grid probability values ​​corresponding to all fourth grid coordinates adjacent to the first grid coordinate all have the preset variation;

[0046] S306: If the fourth grid probability values ​​corresponding to all the fourth grid coordinates adjacent to the first grid coordinate have the preset variation form, it is determined that there is a grid coordinate area with an increased grid probability value in the first current probability grid map.

[0047] In this embodiment, the control system uses the initial probability grid map as a comparison benchmark and the same grid coordinates as a corresponding basis to compare the grid probability values ​​of each grid coordinate in the first current probability grid map with the grid probability values ​​of each grid coordinate in the initial probability grid map, thereby determining whether there is a third grid coordinate in the first current probability grid map corresponding to a third grid probability value having a preset variation. The preset variation indicates that the grid probability value increases from zero to non-zero, that is, the grid probability of the third grid coordinate in the initial probability grid map is 0, while the grid probability value of the third grid coordinate in the first current probability grid map is non-zero (e.g., 0.5). If one or more third grid coordinates corresponding to first grid probability values ​​having a preset variation form exist in the first current probability grid map, then eight fourth grid coordinates adjacent to the single third grid coordinate are obtained (the third grid coordinate and the eight fourth grid coordinates constitute a grid coordinate region, and the third grid coordinate is located at the center of the grid coordinate region). Furthermore, the fourth grid probability values ​​corresponding to each of the eight fourth grid coordinates in the first current probability grid map and the grid probability values ​​corresponding to each of the eight fourth grid coordinates in the initial probability grid map are obtained. Using the same fourth grid coordinate as a comparison benchmark, it is determined whether each fourth grid probability value has a preset variation form (i.e., the grid probability value corresponding to the fourth grid coordinate in the initial probability grid map is 0, while the fourth grid probability value is non-zero). If all fourth grid probability values ​​have the preset variation form, then it is determined that a grid coordinate region with increased grid probability values ​​exists in the first current probability grid map.

[0048] Furthermore, if the grid coordinate area with the increased grid probability value exists in the first current probability grid map, after the step of setting the center grid coordinates of the grid coordinate area as the obstacle position, the method further includes:

[0049] S6: updating the initial real-time map corresponding to the first current probability grid map according to the central grid coordinates corresponding to the obstacle position to obtain a current real-time map;

[0050] S7: Plan the cleaning path of the sweeper according to the current real-time map to avoid the position of the obstacle.

[0051] In this embodiment, after identifying a grid coordinate region with an elevated grid probability value in the first current probability grid map, the control system sets the center grid coordinate of the grid coordinate region as the obstacle location. Furthermore, the initial instantaneous map corresponding to the first current probability grid map is updated based on the center grid coordinate corresponding to the obstacle location, and the obstacle location is displayed on the initial instantaneous map, thereby obtaining an updated current instantaneous map that matches the actual environment of the current scene. The control system plans the sweeper's cleaning path based on the updated current instantaneous map so that the sweeper can avoid the obstacle location during subsequent movements, preventing collisions with obstacles due to untimely updates.

[0052] Preferably, if the control system identifies that there is a grid coordinate area with a reduced grid probability value in the first current probability grid map, the grid probability value corresponding to the central grid coordinate of the grid coordinate area is set to 0, thereby setting the central grid coordinate of the grid coordinate area to a blank position.

[0053] Furthermore, if the grid coordinate area with the increased grid probability value exists in the first current probability grid map, after the step of setting the center grid coordinates of the grid coordinate area as the obstacle position, the method further includes:

[0054] S8: Control the sweeping robot to move a preset distance in a preset direction, recollect the preset number of frames of second laser radar data, and reposition the sweeping robot to obtain a second current position of the sweeping robot in the initial offline map;

[0055] S9: Building a map based on the second current position and the initial probability grid map according to each of the second laser radar data to obtain a second current probability grid map;

[0056] S10: searching for a grid coordinate region to be compared in the second current probability grid map based on a position change between the first current position and the second current position, the grid coordinate region to be compared being a grid coordinate region in the first current probability grid map where the grid probability value increases;

[0057] S11: Determine whether each fifth grid probability value of the grid coordinate area to be compared is in one-to-one correspondence with each first grid probability value of the grid coordinate area with increased grid probability value in the first current probability grid map;

[0058] S12: If the fifth grid probability values ​​of the grid coordinate area to be compared are in one-to-one correspondence with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map, it is determined that the obstacle position is set correctly.

[0059] In this embodiment, after the control system determines that an obstacle appears in the original blank position in the current scene, in order to avoid misjudgment of the grid value due to light and shadow (camera type), glass (laser type), etc., it adjusts the collection angle and position of the lidar data to make a re-judgment, thereby avoiding the problem of misjudgment.

[0060] Specifically, a laser radar is deployed on the sweeping machine, and the control system controls the sweeping machine to move a preset distance in a preset direction, and then re-collects a preset number of frames of second laser radar data through the laser radar, and repositions the sweeping machine based on the surrounding environmental reference objects, thereby obtaining the second current position of the sweeping machine in the initial offline map. The control system uses the second current position as the position reference, inserts each frame of the second laser radar data into the initial probability grid map for mapping, and thus obtains the second current probability grid map. Then, the control system searches for the grid coordinate area to be compared in the second current probability grid map based on the position change between the first current position and the second current position (specifically, refer to the above-mentioned preset direction and preset distance). The grid coordinate area to be compared is the grid coordinate area where the grid probability value in the first current probability grid map increases. The control system compares each fifth grid probability value in the grid coordinate area to be compared with each first grid probability value in the grid coordinate area with increased grid probability values ​​in the first current probability grid map, and compares the fifth grid probability values ​​and first grid probability values ​​corresponding to the same position one by one, thereby determining whether each fifth grid probability value and each first grid probability value at the same position are consistent in a one-to-one correspondence. If each fifth grid probability value in the grid coordinate area to be compared is consistent in a one-to-one correspondence with each first grid probability value at the same position in the grid coordinate area with increased grid probability values ​​in the first current probability grid map, then it is determined that the obstacle position setting is correct.

[0061] Furthermore, the step of controlling the sweeping robot to move a preset distance in a preset direction and recollecting the preset number of frames of second laser radar data includes:

[0062] S801: Retrieve historical position information, where the historical position information includes the position information of each reference object in the current scene and the length of time the position has been fixed;

[0063] S802: Filtering fixed reference objects that meet preset conditions from the historical location information, the preset conditions including a time condition and a distance condition, the time condition being that the position of the reference object remains fixed for a period greater than a time threshold, and the distance condition being that the distance between the position of the reference object and the first current location is less than a distance threshold and is the minimum;

[0064] S803: Determine, based on the initial offline map, a relative direction and a relative distance between the first current position and a reference position of the fixed reference object;

[0065] S804: Using the relative direction as the preset direction and the relative distance as the preset distance, control the sweeper to move to the reference object position, and collect the preset number of frames of second laser radar data.

[0066] In this embodiment, the control system retrieves pre-recorded historical position information, which includes the position information of each reference object in the current scene (such as the position coordinate A of the dressing table placed in the room) and the length of time the position is fixed (such as the length of time the dressing table remains unchanged at position coordinate A). The control system uses the first current position of the sweeper in the current scene as the position reference, and filters out fixed reference objects that meet the preset conditions from the historical position information; wherein, the preset conditions include time conditions and distance conditions, the time condition is that the length of time the position of the selected reference object is fixed is greater than the time threshold (such as 5 days), thereby ensuring the position stability of the selected reference object; and the distance condition means that the distance between the position of the selected reference object and the first current position of the sweeper is less than the distance threshold and the distance is minimum, thereby ensuring that the sweeper does not need to move the distance, reducing the difficulty of comparison. The control system is based on the initial offline map to determine the relative direction and relative distance between the first current position and the reference position of the fixed reference object. Finally, the relative direction is set as the preset direction of movement of the sweeping machine, the relative distance is set as the preset distance of movement of the sweeping machine, and the sweeping machine is controlled to move the preset distance in the preset direction to move to the reference object position, thereby performing data collection and obtaining a preset number of frames of second lidar data.

[0067] Furthermore, the sweeping robot is equipped with a camera, and after the step of determining whether the second grid probability values ​​of the grid coordinate area to be compared are in one-to-one correspondence with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map, the method includes:

[0068] S13: If the fifth grid probability values ​​of the grid coordinate area to be compared do not correspond one-to-one with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map, collecting an environmental image of the obstacle position through the camera;

[0069] S14: Determine whether there is an obstacle at the obstacle position according to the environment image;

[0070] S15: If the obstacle exists at the obstacle position, determining that the obstacle position is set correctly;

[0071] S16: If the obstacle does not exist at the obstacle position, it is determined that the obstacle position is set incorrectly.

[0072] In this embodiment, a camera is deployed on the sweeper. If the fifth grid probability values ​​of the grid coordinate area to be compared do not correspond one-to-one with the first grid probability values ​​corresponding to the same position in the grid coordinate area where the grid probability value in the first current probability grid map increases, the control system collects the environmental image of the obstacle position through the camera. Then, the control system processes the environmental image through AI recognition technology, and judges whether there is an obstacle at the obstacle position through the environmental image (for example, whether there is furniture placed at the obstacle position in the environmental image). If there is an obstacle at the obstacle position, it is determined that the current obstacle position is set correctly. If there is no obstacle at the obstacle position, it is determined that the obstacle position is set incorrectly.

[0073] Reference Figure 2 In one embodiment of the present application, a rapid update device based on an offline map is provided, which is applied to a sweeper. The rapid update device includes:

[0074] A repositioning module 1 is configured to load an initial offline map and reposition the sweeper to obtain a first current position of the sweeper on the initial offline map;

[0075] A first mapping module 2 is configured to collect a preset number of frames of first lidar data, and to construct a map based on the first current position in combination with an initial probability grid map of the initial offline map to obtain a first current probability grid map;

[0076] A first judgment module 3 is configured to compare the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determine whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map;

[0077] A first setting module 4 is configured to set the center grid coordinates of the grid coordinate area to a blank position if the grid coordinate area with a reduced grid probability value exists in the first current probability grid map;

[0078] The second setting module 5 is configured to set the center grid coordinates of the grid coordinate area as the obstacle position if the grid coordinate area with the increased grid probability value exists in the first current probability grid map.

[0079] Furthermore, the first judgment module 3 includes:

[0080] a first determining unit, configured to compare the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determine whether there is a first grid coordinate in the first current probability grid map where the first grid probability value is reduced;

[0081] a second determining unit configured to determine, based on the initial probability grid map, whether the second grid probability values ​​corresponding to all second grid coordinates adjacent to the first grid coordinate in the first current probability grid map are reduced, if the first grid coordinate having a reduced first grid probability value exists in the first current probability grid map;

[0082] The first determining unit is configured to determine that a grid coordinate region with reduced grid probability values ​​exists in the first current probability grid map if the second grid probability values ​​corresponding to all the second grid coordinates adjacent to the first grid coordinates are reduced.

[0083] Furthermore, the first judgment module 3 further includes:

[0084] a third determining unit, configured to compare the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determine whether there is a third grid coordinate corresponding to a third grid probability value having a preset change form in the first current probability grid map, where the preset change form indicates that the grid probability value increases from zero to non-zero;

[0085] a fourth determining unit configured to determine, based on the initial probability grid map, whether the fourth grid probability values ​​corresponding to all fourth grid coordinates adjacent to the first grid coordinate all have the preset variation form, if the third grid coordinate corresponding to the third grid probability value having the preset variation form exists in the first current probability grid map;

[0086] The second determining unit is configured to determine that there is a grid coordinate area with an increased grid probability value in the first current probability grid map if the fourth grid probability values ​​corresponding to all the fourth grid coordinates adjacent to the first grid coordinate have the preset change form.

[0087] Furthermore, the rapid update device further includes:

[0088] An updating module 6 is configured to update an initial real-time map corresponding to the first current probability grid map according to the central grid coordinates corresponding to the obstacle position to obtain a current real-time map;

[0089] The planning module 7 is used to plan the cleaning path of the sweeper according to the current real-time map to avoid the position of the obstacle.

[0090] Furthermore, the rapid update device further includes:

[0091] a control module 8, configured to control the sweeping machine to move a preset distance in a preset direction, recollect the preset number of frames of second laser radar data, and reposition the sweeping machine to obtain a second current position of the sweeping machine in the initial offline map;

[0092] A second mapping module 9 is configured to construct a map based on each of the second laser radar data, the second current position and the initial probability grid map, to obtain a second current probability grid map;

[0093] a search module 10 configured to search, in the second current probability grid map, for a grid coordinate region to be compared based on a position change between the first current position and the second current position, wherein the grid coordinate region to be compared is a grid coordinate region in the first current probability grid map where the grid probability value increases;

[0094] A second judgment module 11 is configured to judge, based on the same position, whether the fifth grid probability values ​​of the grid coordinate area to be compared are in one-to-one correspondence with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map;

[0095] The first determination module 12 is configured to determine that the obstacle position is correctly set if the fifth grid probability values ​​of the grid coordinate area to be compared are in one-to-one correspondence with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map.

[0096] Furthermore, the control module 8 includes:

[0097] A retrieving unit, configured to retrieve historical position information, wherein the historical position information includes position information of each reference object in the current scene and a fixed position retention time;

[0098] a screening unit, configured to screen out fixed reference objects that meet preset conditions from the historical position information, the preset conditions including a time condition and a distance condition, the time condition being that the position of the reference object is fixed for a period greater than a time threshold, and the distance condition being that the distance between the position of the reference object and the first current position is less than a distance threshold and is minimum;

[0099] a determining unit, configured to determine a relative direction and a relative distance between the first current position and a reference object position of the fixed reference object based on the initial offline map;

[0100] A control unit is used to use the relative direction as the preset direction, the relative distance as the preset distance, control the sweeper to move to the reference object position, and collect the preset number of frames of second laser radar data.

[0101] Furthermore, the sweeping robot is equipped with a camera, and the rapid update device further includes:

[0102] The acquisition module 13 is configured to acquire an environmental image of the obstacle position through the camera if the fifth grid probability values ​​of the grid coordinate area to be compared do not correspond one-to-one with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map;

[0103] A third judgment module 14 is used to judge whether there is an obstacle at the obstacle position according to the environment image;

[0104] A second determination module 15 is configured to determine whether the obstacle position is set correctly if the obstacle exists at the obstacle position;

[0105] The third determining module 16 is configured to determine that the obstacle position is incorrectly set if the obstacle does not exist at the obstacle position.

[0106] In this embodiment, each module and unit in the offline map-based rapid update device is used to correspondingly execute each step in the above-mentioned offline map-based rapid update method, and its specific implementation process will not be described in detail here.

[0107] This embodiment provides a fast update device based on an offline map. When the sweeper starts cleaning, it first loads the initial offline map and repositions the sweeper to obtain the first current position of the sweeper in the initial offline map. Then, a preset number of frames of first lidar data are collected, and a map is constructed based on the first current position combined with the initial probability grid map of the initial offline map to obtain a first current probability grid map. The first current probability grid map and the initial probability grid map are then compared one by one according to the corresponding grid coordinates to determine whether there is a grid coordinate area with a reduced grid probability value in the first current probability grid map. If there is a grid coordinate area with a reduced grid probability value in the first current probability grid map, the center grid coordinate of the grid coordinate area is set to a blank position. If there is a grid coordinate area with an increased grid probability value in the first current probability grid map, the center grid coordinate of the grid coordinate area is set to the obstacle position. This application achieves rapid updating of the initial offline map by combining a preset number of frames of first lidar data collected for the current scene on the basis of the initial offline map, thereby more efficiently identifying the environmental changes that have occurred in the current scene relative to the initial offline map, thereby facilitating the subsequent cleaning path planning and obstacle avoidance movement of the sweeper.

[0108] Reference Figure 3In the embodiment of the present application, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as initial offline maps. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a fast update method based on offline maps is implemented, which is applied to a sweeper.

[0109] The processor executes the following steps of the offline map-based fast update method:

[0110] S1: loading an initial offline map and relocating the sweeper to obtain a first current position of the sweeper in the initial offline map;

[0111] S2: Collect a preset number of frames of first laser radar data, and build a map based on the first current position in combination with the initial probability grid map of the initial offline map to obtain a first current probability grid map;

[0112] S3: comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determining whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map;

[0113] S4: if the grid coordinate area with the reduced grid probability value exists in the first current probability grid map, setting the center grid coordinate of the grid coordinate area to a blank position;

[0114] S5: If the grid coordinate area with the increased grid probability value exists in the first current probability grid map, the center grid coordinates of the grid coordinate area are set as the obstacle position.

[0115] Furthermore, the step of comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates to determine whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map includes:

[0116] S301: Compare the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determine whether there is a first grid coordinate with a reduced first grid probability value in the first current probability grid map;

[0117] S302: If the first grid coordinate having a reduced first grid probability value exists in the first current probability grid map, determining, based on the initial probability grid map, whether the second grid probability values ​​corresponding to all second grid coordinates adjacent to the first grid coordinate in the first current probability grid map are reduced;

[0118] S303: If the second grid probability values ​​corresponding to all the second grid coordinates adjacent to the first grid coordinate are reduced, it is determined that the grid coordinate area with the reduced grid probability value exists in the first current probability grid map.

[0119] Furthermore, the step of comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates to determine whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map further includes:

[0120] S304: Comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determining whether there is a third grid coordinate corresponding to a third grid probability value having a preset change form in the first current probability grid map, where the preset change form indicates that the grid probability value increases from zero to non-zero;

[0121] S305: If the third grid coordinate corresponding to the third grid probability value having the preset variation exists in the first current probability grid map, determining, based on the initial probability grid map, whether the fourth grid probability values ​​corresponding to all fourth grid coordinates adjacent to the first grid coordinate all have the preset variation;

[0122] S306: If the fourth grid probability values ​​corresponding to all the fourth grid coordinates adjacent to the first grid coordinate have the preset variation form, it is determined that there is a grid coordinate area with an increased grid probability value in the first current probability grid map.

[0123] Furthermore, if the grid coordinate area with the increased grid probability value exists in the first current probability grid map, after the step of setting the center grid coordinates of the grid coordinate area as the obstacle position, the method further includes:

[0124] S6: updating the initial real-time map corresponding to the first current probability grid map according to the central grid coordinates corresponding to the obstacle position to obtain a current real-time map;

[0125] S7: Plan the cleaning path of the sweeper according to the current real-time map to avoid the position of the obstacle.

[0126] Furthermore, if the grid coordinate area with the increased grid probability value exists in the first current probability grid map, after the step of setting the center grid coordinates of the grid coordinate area as the obstacle position, the method further includes:

[0127] S8: Control the sweeping robot to move a preset distance in a preset direction, recollect the preset number of frames of second laser radar data, and reposition the sweeping robot to obtain a second current position of the sweeping robot in the initial offline map;

[0128] S9: Building a map based on the second current position and the initial probability grid map according to each of the second laser radar data to obtain a second current probability grid map;

[0129] S10: searching for a grid coordinate region to be compared in the second current probability grid map based on a position change between the first current position and the second current position, the grid coordinate region to be compared being a grid coordinate region in the first current probability grid map where the grid probability value increases;

[0130] S11: Determine whether each fifth grid probability value of the grid coordinate area to be compared is in one-to-one correspondence with each first grid probability value of the grid coordinate area with increased grid probability value in the first current probability grid map;

[0131] S12: If the fifth grid probability values ​​of the grid coordinate area to be compared are in one-to-one correspondence with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map, it is determined that the obstacle position is set correctly.

[0132] Furthermore, the step of controlling the sweeping robot to move a preset distance in a preset direction and recollecting the preset number of frames of second laser radar data includes:

[0133] S801: Retrieve historical position information, where the historical position information includes the position information of each reference object in the current scene and the length of time the position has been fixed;

[0134] S802: Filtering fixed reference objects that meet preset conditions from the historical location information, the preset conditions including a time condition and a distance condition, the time condition being that the position of the reference object remains fixed for a period greater than a time threshold, and the distance condition being that the distance between the position of the reference object and the first current location is less than a distance threshold and is the minimum;

[0135] S803: Determine, based on the initial offline map, a relative direction and a relative distance between the first current position and a reference position of the fixed reference object;

[0136] S804: Using the relative direction as the preset direction and the relative distance as the preset distance, control the sweeper to move to the reference object position, and collect the preset number of frames of second laser radar data.

[0137] Furthermore, the sweeping robot is equipped with a camera, and after the step of determining whether the second grid probability values ​​of the grid coordinate area to be compared are in one-to-one correspondence with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map, the method includes:

[0138] S13: If the fifth grid probability values ​​of the grid coordinate area to be compared do not correspond one-to-one with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map, collecting an environmental image of the obstacle position through the camera;

[0139] S14: Determine whether there is an obstacle at the obstacle position according to the environment image;

[0140] S15: If the obstacle exists at the obstacle position, determining that the obstacle position is set correctly;

[0141] S16: If the obstacle does not exist at the obstacle position, it is determined that the obstacle position is set incorrectly.

[0142] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, a method for quickly updating an offline map is implemented, which is applied to a sweeping robot. The method for quickly updating an offline map is specifically as follows:

[0143] S1: loading an initial offline map and relocating the sweeper to obtain a first current position of the sweeper in the initial offline map;

[0144] S2: Collect a preset number of frames of first laser radar data, and build a map based on the first current position in combination with the initial probability grid map of the initial offline map to obtain a first current probability grid map;

[0145] S3: comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determining whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map;

[0146] S4: if the grid coordinate area with the reduced grid probability value exists in the first current probability grid map, setting the center grid coordinate of the grid coordinate area to a blank position;

[0147] S5: If the grid coordinate area with the increased grid probability value exists in the first current probability grid map, the center grid coordinates of the grid coordinate area are set as the obstacle position.

[0148] Furthermore, the step of comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates to determine whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map includes:

[0149] S301: Compare the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determine whether there is a first grid coordinate with a reduced first grid probability value in the first current probability grid map;

[0150] S302: If the first grid coordinate having a reduced first grid probability value exists in the first current probability grid map, determining, based on the initial probability grid map, whether the second grid probability values ​​corresponding to all second grid coordinates adjacent to the first grid coordinate in the first current probability grid map are reduced;

[0151] S303: If the second grid probability values ​​corresponding to all the second grid coordinates adjacent to the first grid coordinate are reduced, it is determined that the grid coordinate area with the reduced grid probability value exists in the first current probability grid map.

[0152] Furthermore, the step of comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates to determine whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map further includes:

[0153] S304: Comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determining whether there is a third grid coordinate corresponding to a third grid probability value having a preset change form in the first current probability grid map, where the preset change form indicates that the grid probability value increases from zero to non-zero;

[0154] S305: If the third grid coordinate corresponding to the third grid probability value having the preset variation exists in the first current probability grid map, determining, based on the initial probability grid map, whether the fourth grid probability values ​​corresponding to all fourth grid coordinates adjacent to the first grid coordinate all have the preset variation;

[0155] S306: If the fourth grid probability values ​​corresponding to all the fourth grid coordinates adjacent to the first grid coordinate have the preset variation form, it is determined that there is a grid coordinate area with an increased grid probability value in the first current probability grid map.

[0156] Furthermore, if the grid coordinate area with the increased grid probability value exists in the first current probability grid map, after the step of setting the center grid coordinates of the grid coordinate area as the obstacle position, the method further includes:

[0157] S6: updating the initial real-time map corresponding to the first current probability grid map according to the central grid coordinates corresponding to the obstacle position to obtain a current real-time map;

[0158] S7: Plan the cleaning path of the sweeper according to the current real-time map to avoid the position of the obstacle.

[0159] Furthermore, if the grid coordinate area with the increased grid probability value exists in the first current probability grid map, after the step of setting the center grid coordinates of the grid coordinate area as the obstacle position, the method further includes:

[0160] S8: Control the sweeping robot to move a preset distance in a preset direction, recollect the preset number of frames of second laser radar data, and reposition the sweeping robot to obtain a second current position of the sweeping robot in the initial offline map;

[0161] S9: Building a map based on the second current position and the initial probability grid map according to each of the second laser radar data to obtain a second current probability grid map;

[0162] S10: searching for a grid coordinate region to be compared in the second current probability grid map based on a position change between the first current position and the second current position, the grid coordinate region to be compared being a grid coordinate region in the first current probability grid map where the grid probability value increases;

[0163] S11: Determine whether each fifth grid probability value of the grid coordinate area to be compared is in one-to-one correspondence with each first grid probability value of the grid coordinate area with increased grid probability value in the first current probability grid map;

[0164] S12: If the fifth grid probability values ​​of the grid coordinate area to be compared are in one-to-one correspondence with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map, it is determined that the obstacle position is set correctly.

[0165] Furthermore, the step of controlling the sweeping robot to move a preset distance in a preset direction and recollecting the preset number of frames of second laser radar data includes:

[0166] S801: Retrieve historical position information, where the historical position information includes the position information of each reference object in the current scene and the length of time the position has been fixed;

[0167] S802: Filtering fixed reference objects that meet preset conditions from the historical location information, the preset conditions including a time condition and a distance condition, the time condition being that the position of the reference object remains fixed for a period greater than a time threshold, and the distance condition being that the distance between the position of the reference object and the first current location is less than a distance threshold and is the minimum;

[0168] S803: Determine, based on the initial offline map, a relative direction and a relative distance between the first current position and a reference position of the fixed reference object;

[0169] S804: Using the relative direction as the preset direction and the relative distance as the preset distance, control the sweeper to move to the reference object position, and collect the preset number of frames of second laser radar data.

[0170] Furthermore, the sweeping robot is equipped with a camera, and after the step of determining whether the second grid probability values ​​of the grid coordinate area to be compared are in one-to-one correspondence with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map, the method includes:

[0171] S13: If the fifth grid probability values ​​of the grid coordinate area to be compared do not correspond one-to-one with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map, collecting an environmental image of the obstacle position through the camera;

[0172] S14: Determine whether there is an obstacle at the obstacle position according to the environment image;

[0173] S15: If the obstacle exists at the obstacle position, determining that the obstacle position is set correctly;

[0174] S16: If the obstacle does not exist at the obstacle position, it is determined that the obstacle position is set incorrectly.

[0175] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0176] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, apparatus, first object, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, apparatus, first object, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, apparatus, first object, or method comprising the element.

[0177] The above description is only a preferred embodiment of the present application and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A fast update method based on offline maps, characterized in that: Applied to a sweeping machine, the method includes: Loading an initial offline map, and relocating the sweeping robot to obtain a first current position of the sweeping robot on the initial offline map; Collecting a preset number of frames of first laser radar data, and constructing a map based on the first current position and an initial probability grid map of the initial offline map to obtain a first current probability grid map; Comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determining whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map; If the grid coordinate area with the reduced grid probability value exists in the first current probability grid map, setting the center grid coordinate of the grid coordinate area to a blank position; If the grid coordinate area with the increased grid probability value exists in the first current probability grid map, setting the center grid coordinates of the grid coordinate area as the obstacle position; After the step of setting the center grid coordinates of the grid coordinate area as the obstacle position if the grid coordinate area with the increased grid probability value exists in the first current probability grid map, the method further includes: Controlling the sweeping robot to move a preset distance in a preset direction, recollecting the preset number of frames of second laser radar data, and repositioning the sweeping robot to obtain a second current position of the sweeping robot in the initial offline map; Building a map based on the second current position and the initial probability grid map according to each of the second laser radar data to obtain a second current probability grid map; Finding a grid coordinate region to be compared in the second current probability grid map based on a position change between the first current position and the second current position, the grid coordinate region to be compared being a grid coordinate region in the first current probability grid map where the grid probability value increases; Based on the same position, determining whether each fifth grid probability value of the grid coordinate area to be compared is consistent with each first grid probability value of the grid coordinate area with increased grid probability value in the first current probability grid map; If the fifth grid probability values ​​of the grid coordinate area to be compared are in one-to-one correspondence with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map, it is determined that the obstacle position setting is correct; The step of controlling the sweeping machine to move a preset distance in a preset direction and recollecting the preset number of frames of second laser radar data includes: Retrieving historical location information, which includes the location information of each reference object in the current scene and the length of time the location was fixed; Filtering a fixed reference object that meets preset conditions from the historical position information, the preset conditions including a time condition and a distance condition, the time condition being that the position of the reference object is fixed for a period greater than a time threshold, and the distance condition being that the distance between the position of the reference object and the first current position is less than a distance threshold and is the minimum; Determining a relative direction and a relative distance between the first current position and a reference object position of the fixed reference object based on the initial offline map; The relative direction is used as the preset direction, the relative distance is used as the preset distance, the sweeper is controlled to move to the reference object position, and the preset number of frames of second laser radar data are collected.

2. The offline map-based rapid update method according to claim 1, characterized in that: The step of comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determining whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map, includes: Comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determining whether there is a first grid coordinate in the first current probability grid map where the first grid probability value decreases; If the first grid coordinate having a reduced first grid probability value exists in the first current probability grid map, determining, based on the initial probability grid map, whether the second grid probability values ​​corresponding to all second grid coordinates adjacent to the first grid coordinate in the first current probability grid map are reduced; If the second grid probability values ​​corresponding to all the second grid coordinates adjacent to the first grid coordinate are reduced, it is determined that the grid coordinate area with the reduced grid probability value exists in the first current probability grid map.

3. The offline map-based rapid update method according to claim 2, characterized in that: The step of comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates to determine whether there is a grid coordinate area with a changed grid probability value in the first current probability grid map further includes: Comparing the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and determining whether there is a third grid coordinate corresponding to a third grid probability value having a preset change form in the first current probability grid map, where the preset change form indicates that the grid probability value increases from zero to non-zero; If the third grid coordinate corresponding to the third grid probability value having the preset variation form exists in the first current probability grid map, determining, based on the initial probability grid map, whether the fourth grid probability values ​​corresponding to all fourth grid coordinates adjacent to the first grid coordinate all have the preset variation form; If the fourth grid probability values ​​corresponding to all the fourth grid coordinates adjacent to the first grid coordinate have the preset variation form, it is determined that the grid coordinate region with increased grid probability values ​​exists in the first current probability grid map.

4. The offline map-based rapid update method according to claim 1, characterized in that: After the step of setting the center grid coordinates of the grid coordinate area as the obstacle position if the grid coordinate area with the increased grid probability value exists in the first current probability grid map, the method further includes: updating the initial real-time map corresponding to the first current probability grid map according to the central grid coordinates corresponding to the obstacle position to obtain a current real-time map; The cleaning path of the sweeper is planned according to the current real-time map to avoid the position of the obstacle.

5. The offline map-based rapid update method according to claim 1, characterized in that: The sweeping robot is equipped with a camera, and after the step of determining whether each fifth grid probability value of the grid coordinate area to be compared is in one-to-one correspondence with each first grid probability value of the grid coordinate area with increased grid probability value in the first current probability grid map, the method includes: If the fifth grid probability values ​​of the grid coordinate area to be compared do not correspond one-to-one with the first grid probability values ​​of the grid coordinate area with increased grid probability values ​​in the first current probability grid map, capturing an environmental image of the obstacle position through the camera; Determining whether there is an obstacle at the obstacle position according to the environment image; If the obstacle exists at the obstacle position, determining that the obstacle position is set correctly; If the obstacle does not exist at the obstacle position, it is determined that the obstacle position is set incorrectly.

6. A rapid update device based on an offline map, which implements the rapid update method based on an offline map according to any one of claims 1 to 5, characterized in that: Applied to a sweeper, the rapid update device includes: a repositioning module, configured to load an initial offline map and reposition the sweeping robot to obtain a first current position of the sweeping robot on the initial offline map; A first mapping module is configured to collect a preset number of frames of first lidar data, and to perform mapping based on the first current position in combination with an initial probability grid map of the initial offline map to obtain a first current probability grid map; A first judgment module is configured to compare the first current probability grid map and the initial probability grid map one by one according to corresponding grid coordinates, and to judge whether there is a grid coordinate region with a changed grid probability value in the first current probability grid map; a first setting module, configured to set the center grid coordinates of the grid coordinate area to a blank position if the grid coordinate area with a reduced grid probability value exists in the first current probability grid map; The second setting module is configured to set the center grid coordinates of the grid coordinate area as the obstacle position if the grid coordinate area with the increased grid probability value exists in the first current probability grid map.

7. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images