Map partitioning method and apparatus, autonomous mobile device, and storage medium

Abstract

The present disclosure provides a map partitioning method, device, autonomous mobile device and storage medium. The method comprises: obtaining a map of a target area; the map comprising reachable boundaries of the autonomous mobile device in the target area; for a plurality of internal points within the reachable boundaries in the map, determining a representative distance value of the internal points according to the distance between the internal points and each reachable boundary; determining the main structure of each candidate room in the map according to the representative distance values of the plurality of internal points within the reachable boundaries in the map; and expanding the main structure of each candidate room in the map to obtain each room in the map. Based on the rule that the distance from the boundary to the center is getting larger and larger, the distance value of each point is determined, and according to the characteristics of the distribution of the distance values of the points, the main structure of each candidate room is determined, and then through the expansion of the main structure of each candidate room, the boundary of each room is further determined, the partitioning of the map is completed, and the partitioning efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to data processing technology, and in particular, to a map partitioning method and device, an autonomous mobile device, and a storage medium. BACKGROUND

[0002] With the progress of science and technology and the improvement of living standards, autonomous mobile devices with different functions, such as cleaning robots, companion mobile robots, etc., have entered more and more families, making people's lives more comfortable and convenient.

[0003] An autonomous mobile device refers to an intelligent device that autonomously performs a preset task in a set working area. Currently, autonomous mobile devices generally include, but are not limited to, cleaning robots (such as intelligent sweeping robots, intelligent mopping robots, window-cleaning robots), companion mobile robots (such as intelligent electronic pets, nanny robots), service mobile robots (such as reception robots in hotels, inns, meeting places), industrial inspection intelligent devices (such as power inspection robots, intelligent forklifts, etc.), security robots (such as household or commercial intelligent security robots), etc.

[0004] Autonomous mobile devices usually move autonomously in a limited space, such as cleaning robots and companion mobile robots, which usually operate indoors, and service mobile robots, which usually operate in specific limited spaces such as inns and meeting places. This limited space can be referred to as the working area of the autonomous mobile device.

[0005] Many times, the working area of the autonomous mobile device is not an open space, but multiple indoor areas, such as multiple rooms, divided by walls, doors, or other partitions. In some scenarios, the user wants the autonomous mobile device to accurately identify the functional areas of the working area map, such as the living room, bedroom, kitchen, corridor, etc., and display them on the human-machine interaction interface (UI) to the user, so that the user can select one or more selected rooms to instruct the autonomous mobile device to perform a specific task, which requires the autonomous mobile device to intelligently and correctly distinguish the functional areas in the working area map.

[0006] In the prior art, the working area map is generally automatically divided into areas by identifying the position of the "door". However, due to the difference in design style of the rooms in the working area, there may be no special "door" as a partition mark between some rooms, resulting in the autonomous mobile device often failing to accurately identify the door. Therefore, this area division method has poor applicability and high misjudgment rate. SUMMARY

[0007] The present disclosure provides a map partitioning method and device, an autonomous mobile device, and a storage medium, which are based on area characteristics for partitioning, thereby improving the efficiency of map partitioning.

[0008] In a first aspect, the disclosure provides a map partitioning method, comprising:

[0009] obtaining a map of a target area; the map comprising reachable boundaries of an autonomous mobile device in the target area;

[0010] for a plurality of interior points within the reachable boundaries in the map, determining a representative distance value of the interior points according to distances of the interior points to each reachable boundary;

[0011] determining a main structure of each map candidate room according to the representative distance values of the plurality of interior points within the reachable boundaries in the map;

[0012] expanding the main structure of each map candidate room to obtain a map room.

[0013] Optionally, the determining a main structure of each map candidate room according to the representative distance values of the plurality of interior points within the reachable boundaries in the map comprises:

[0014] taking the shortest distance of the distances of the interior points to each reachable boundary as the representative distance value;

[0015] repeating the following steps until there is no seed in the updated map:

[0016] determining an interior point with the largest representative distance value in the map;

[0017] determining a first threshold according to the largest representative distance value in the map; the first threshold is smaller than the largest representative distance value;

[0018] determining at least one seed region in which the interior points with the representative distance values greater than the first threshold and smaller than the largest representative distance value in the map are located;

[0019] for each seed region, if the interior point with the largest representative distance value is located in the seed region, taking the seed region as a seed of a map candidate room;

[0020] inflating the seed of the map candidate room to obtain a main structure of the map candidate room;

[0021] deleting the main structure of the map candidate room from the map to obtain an updated map.

[0022] Optionally, the determining a main structure of each map candidate room according to the representative distance values of the plurality of interior points within the reachable boundaries in the map comprises:

[0023] taking the longest distance of the distances of the interior points to each reachable boundary as the representative distance value;

[0024] Repeat the following steps until there are no seeds in the updated map:

[0025] Identify the interior point in the map that represents the smallest distance value;

[0026] A second threshold is determined based on the smallest representative distance value in the map, and the second threshold is greater than the smallest representative distance value;

[0027] Determine at least one seed region in the map containing internal points whose representative distance values ​​are less than a second threshold and greater than a minimum representative distance value;

[0028] For each seed region, if the interior point with the smallest representative distance value is located in the seed region, then the seed region is used as the seed for the candidate room in the graph;

[0029] The seeds of the candidate rooms in the figure are expanded to obtain the main structure of the candidate rooms in the figure;

[0030] The main structure of the candidate rooms in the map is deleted to obtain an updated map.

[0031] Optionally, expanding the main structure of the candidate rooms in each of the diagrams to obtain the rooms in each diagram includes:

[0032] The main structure of each candidate room in each diagram expands simultaneously at the same preset rate, becoming the expanding main structure.

[0033] If the main expansion structure of a candidate room in the above figure meets the main expansion structure of a candidate room in other figures, creating a main structure boundary, or if the main expansion structure of a candidate room in the above figure meets the reachable boundary of the map, then the expansion stops, and the rooms in each figure are obtained.

[0034] Optionally, after expanding the main structure of the candidate rooms in each of the figures to obtain the rooms in each figure, the method further includes:

[0035] The rooms in each image are merged to obtain the rooms in the merged image.

[0036] Optionally, the process of merging candidate rooms in each image to obtain rooms in the merged image includes:

[0037] Obtain the area of ​​each room in the diagram and the main structural boundaries between the rooms;

[0038] Determine the adjacency relationship between rooms in the diagram based on the main structural boundaries between them, and thus identify adjacent rooms.

[0039] comparing the area of the room in the graph with the size of the area threshold value, and determining that the room in the graph is a target room if the area of the room in the graph is less than the area threshold value;

[0040] fusing the target room and one adjacent room of the target room into a room in a fused graph.

[0041] Optionally, the method further includes:

[0042] obtaining an original map of the target area;

[0043] preprocessing the original map to obtain the map of the target area.

[0044] In a second aspect, the present disclosure provides a map partitioning device, including:

[0045] a obtaining module configured to obtain a map of a target area; the map including reachable boundaries of an autonomous mobile device in the target area;

[0046] a representative distance value determining module configured to determine, for a plurality of internal points within the reachable boundaries in the map, a representative distance value of each internal point according to a distance between the internal point and each reachable boundary;

[0047] a main structure determining module configured to determine a main structure of each graph room candidate according to the representative distance values of the plurality of internal points within the reachable boundaries in the map;

[0048] a main structure expanding module configured to expand the main structure of each graph room candidate to obtain each room in the graph.

[0049] Optionally, the main structure determining module is specifically configured to:

[0050] determine the shortest distance between the internal point and each reachable boundary as the representative distance value;

[0051] repeating the following steps until there is no seed in the updated map:

[0052] determining an internal point with the largest representative distance value in the map;

[0053] determining a first threshold value according to the largest representative distance value in the map; the first threshold value is less than the largest representative distance value;

[0054] determining at least one seed region in which the internal points with the representative distance values greater than the first threshold value and less than the largest representative distance value are located in the map;

[0055] for each seed region, if the internal point with the largest representative distance value is located in the seed region, regarding the seed region as a seed of a graph room candidate.

[0056] inflating the seeds of the candidate room in the graph to obtain a main structure of the candidate room in the graph;

[0057] deleting the main structure of the candidate room in the graph from the map to obtain an updated map.

[0058] Optionally, the main structure determining module is specifically configured to:

[0059] taking the longest distance among distances from the interior point to each reachable boundary as a representative distance value;

[0060] repeating the following steps until there is no seed in the updated map:

[0061] determining an interior point with the minimum representative distance value in the map;

[0062] determining a second threshold according to the minimum representative distance value in the map, the second threshold being greater than the minimum representative distance value;

[0063] determining at least one seed region in which an interior point with a representative distance value less than the second threshold and greater than the minimum representative distance value is located;

[0064] for each seed region, if the interior point with the minimum representative distance value is located in the seed region, taking the seed region as a seed of the candidate room in the graph;

[0065] inflating the seeds of the candidate room in the graph to obtain a main structure of the candidate room in the graph;

[0066] deleting the main structure of the candidate room in the graph from the map to obtain an updated map.

[0067] Optionally, the main structure expanding module is specifically configured to:

[0068] the main structures of each candidate room in the graph expand at the same preset rate simultaneously to become expanded main structures;

[0069] if the expanded main structure of the candidate room in the graph meets the expanded main structure of another candidate room in the graph to generate a main structure boundary, or if the expanded main structure of the candidate room in the graph meets a reachable boundary of the map to generate a main structure boundary, the expansion is stopped to obtain a room in the graph.

[0070] Optionally, the apparatus further includes a fusion module configured to fuse the rooms in the graphs after the main structure expanding module expands the main structures of the candidate rooms in the graphs to obtain rooms in the graphs.

[0071] Optionally, when the fusion module fuses the candidate rooms in each graph to obtain the rooms in the fused graph, the fusion module is specifically configured to:

[0072] The area of each room in the graph and the main structure intersection between the rooms in the graph are obtained;

[0073] The adjacent relationship between the rooms in the graph is determined according to the main structure intersection between the rooms in the graph, so as to determine the adjacent rooms;

[0074] The area of the room in the graph is compared with the area threshold value, and if the area of the room in the graph is less than the area threshold value, the room in the graph is determined as the target room;

[0075] The target room and the adjacent rooms thereof are fused.

[0076] Optionally, the acquisition module is specifically configured to:

[0077] An original map of the target area is acquired;

[0078] The original map is preprocessed to obtain the map of the target area.

[0079] In a third aspect, the present disclosure provides an autonomous mobile device, comprising: a memory configured to store program instructions; and a processor configured to invoke and execute the program instructions in the memory, and execute the method in the first aspect.

[0080] In a fourth aspect, the present disclosure provides a computer-readable storage medium, wherein the storage medium stores a computer program, and the computer program is executed by a processor to implement the method in the first aspect.

[0081] In a fifth aspect, the present disclosure provides a computer program product, comprising a computer program, and the computer program is executed by a processor to implement the method in the first aspect.

[0082] The disclosure provides a map partitioning method, device, autonomous mobile device and storage medium. The method comprises: obtaining a map of a target area; the map comprising reachable boundaries of an autonomous mobile device in the target area; for a plurality of internal points within the reachable boundaries in the map, determining a representative distance value of the internal point according to the distance between the internal point and each reachable boundary; determining the main structure of each map candidate room according to the representative distance value of the plurality of internal points within the reachable boundaries in the map; and expanding the main structure of each map candidate room to obtain each map room. The scheme of the disclosure determines the representative distance value of each point in the map, and determines the main structure of each candidate room according to the characteristics of the distribution of the representative distance value of each point, and further determines the reachable boundary of each room by expanding the main structure of each candidate room, thereby completing the automatic partitioning of the map. It does not rely on actual walls, doors and other partitioning marks in the space, and can be applied to partitioning in more scenarios, and has higher partitioning accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0083] In order to more clearly illustrate the technical solutions in the disclosure or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the disclosure, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0084] Figure 1 A schematic diagram of an application scenario provided by the disclosure;

[0085] Figure 2 A flowchart of a map partitioning method provided by an embodiment of the disclosure;

[0086] Figure 3 A schematic diagram of a distance value of a point in a space provided by an embodiment of the disclosure;

[0087] Figure 4 A flowchart of determining the main structure of a map candidate room provided by an embodiment of the disclosure;

[0088] Figure 5 A schematic diagram of main structure expansion of a map candidate room provided by an embodiment of the disclosure;

[0089] Figure 6 A flowchart of fusing each map candidate room provided by an embodiment of the disclosure;

[0090] Figures 7a to 7n A schematic diagram of a map generated in the map partitioning process provided by an embodiment of the disclosure;

[0091] Figure 8A structural schematic diagram of a map partitioning device provided by an embodiment of the present disclosure is shown in FIG. 1.

[0092] Figure 9 A structural schematic diagram of an autonomous mobile device provided by an embodiment of the present disclosure is shown in FIG. 2. DETAILED DESCRIPTION

[0093] In order to make the objectives, technical solutions, and advantages of the present disclosure clearer, the technical solutions in the present disclosure will be described clearly and completely below with reference to the drawings in the present disclosure. Obviously, the described embodiments are only some, but not all, of the embodiments of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present disclosure.

[0094] According to the nature of the task and the partition of the working area, the autonomous mobile device can need to perform a task in a part of the specified area in the working area. For example, a smart cleaning robot performs a task of cleaning a living room area. This requires that the autonomous mobile device can automatically partition the working area.

[0095] The present disclosure provides a map partitioning method, device, autonomous mobile device, and storage medium, which partitions based on the characteristics of the area, and can improve the efficiency and accuracy of map partitioning.

[0096] Figure 1 A schematic diagram of an application scenario provided by the present disclosure is shown in FIG. 3. The autonomous mobile device in the present embodiment is a cleaning robot 101, which performs cleaning work in a working area in a room, as shown in FIG. 3. Figure 1 The cleaning robot 101 automatically partitions the working area in the room based on the current indoor map (i.e., the map of the working area) based on the map partitioning method of the present disclosure, determines a plurality of functional areas (i.e., rooms) contained in the room, obtains a functional area map, and then the user issues a cleaning task instruction to the autonomous mobile device through the UI of a terminal device such as a mobile phone or a computer. The cleaning task instruction includes a selected functional area (such as a living room area) on the functional area map to perform a cleaning task. The cleaning instruction can also include setting a working mode corresponding to the selected area in the selected area, such as a silent mode, a powerful mode, a repeated cleaning mode, etc. The specific implementation of the partitioning process can refer to the following embodiments.

[0097] Figure 2 A flowchart of a map partitioning method provided by an embodiment of the present disclosure is shown in FIG. 4. The execution subject of the method of the present embodiment can be an autonomous mobile device or a management system of the autonomous mobile device. As shown in FIG. 4, the method of the present embodiment can include: Figure 2

[0098] ​S201, obtaining a map of the target area; the map comprises a reachable boundary of the autonomous mobile device in the target area.

[0099] The target area is at least part of the area where the autonomous mobile device works, which can be the entire indoor space of a user, part of the entire indoor space, or even part of a closed space or a closed space.

[0100] In some embodiments, the map is a raster map (also known as a dot matrix map or a bitmap), and the autonomous mobile device creates the map of the target area while moving in the target area. To distinguish between unexplored positions, obstacle positions, and reachable positions / passing positions in the map, for example, during the creation of the map, the state values of all positions in the map are initially set to an initial value (usually consistent with the state value of the unexplored area) at the beginning, and the autonomous mobile device moves in the target area. When it reaches a position, it updates the state value of the position or can update the state values of the coordinates on the trajectory according to the trajectory it passes through in a period of time. For example, the state value of an unexplored position is initially set to 75 (such as a pixel value or a gray value; in this embodiment, the initial value of all positions at the beginning is 75); the state values of the coordinates corresponding to the reachable positions and the passing positions (which can be collectively referred to as passable positions) are set to 0 (such as a pixel value or a gray value), and the state values of the coordinates corresponding to the positions that are blocked by obstacles and cannot be reached (referred to as unreachable positions) are set or updated to 100 (such as a pixel value or a gray value). In this way, in the map of the target area, the intersection between the passable region with a state value of 0 and the unreachable region with a state value of 100 is the reachable boundary of the autonomous mobile device in the target area. Usually, for the sake of convenience, the position of the passable region with a state value of 0 adjacent to the position of the unreachable region with a state value of 100 can be defined as the reachable boundary, or the position of the unreachable region with a state value of 100 adjacent to the position of the passable region with a state value of 0 can be defined as the reachable boundary. Therefore, the points in the map of the target area are divided into two categories: the passable positions of the autonomous mobile device (referred to as internal points) and the positions that the autonomous mobile device cannot reach (such as obstacle positions and obstacle regions within a closed curve formed by sequentially connecting adjacent obstacle positions, which are all positions that the autonomous mobile device cannot reach, and the points within them can be referred to as unreachable points). It should be noted that the map can also be a vector map, and those skilled in the art should know the existing technologies that can implement the above scheme.

[0101] When the execution subject of the present embodiment is an autonomous mobile device, the map of the target area constructed by the device can be obtained, or the stored constructed map of the target area can be obtained from the autonomous mobile device itself or an external storage device (such as a server, a distributed storage system, a computer, a mobile phone, or another autonomous mobile device).

[0102] When the execution subject of the embodiment is a management system (such as a server) of the autonomous mobile device, a map of the target area stored by the system can be acquired, or a constructed map of the target area uploaded by the autonomous mobile device can be received.

[0103] In some other embodiments, the map of the target area can also be a map edited by a user. For example, a user can acquire a historical map on the autonomous mobile device in a mobile phone or a computer through a network or a wired manner, edit the historical map by adding, modifying, deleting, and the like, and save the edited historical map to a device used by the user, a server, or a storage device, or transmit the edited historical map to the autonomous mobile device.

[0104] S202, for a plurality of internal points in the reachable boundary, a representative distance value of the internal point is determined according to the distance between the internal point and each reachable boundary.

[0105] The plurality of internal points can select internal points of a part of the passable area in the map. Specifically, one or more internal points can be selected in each row and / or each column at every one or more internal points.

[0106] For example, the plurality of internal points can also select all internal points on the map of the target area. Specifically, each internal point can be traversed, such as starting from a certain internal point adjacent to a certain reachable boundary, selecting points in the row or column, and then gradually increasing or decreasing the row number or column number until all internal points in the passable area are traversed; in this process, the distance between each selected internal point and each reachable boundary is calculated.

[0107] The distance between the internal point and each reachable boundary refers to the distance between the internal point and the reachable boundary in the map of the target area; it can be the distance between the internal point and each point on the reachable boundary (which is necessarily multiple distances); it can also be the distance between the internal point and the point on the reachable boundary that meets certain preset conditions, such as a perpendicular line from the internal point to the reachable boundary intersecting at an intersection point, the length of the perpendicular segment between the internal point and the intersection point being the distance between the internal point and the reachable boundary. For the convenience of description, the distance between the internal point and a certain reachable boundary obtained by this method is referred to as the perpendicular distance, and the perpendicular distance is a single distance, i.e., the distance between one internal point and one reachable boundary can be one, i.e., the perpendicular distance, but usually the internal point has more than one reachable boundary, as shown in FIG. 1, each internal point A, B, C, and D has four reachable boundaries, so the distance between an internal point and a reachable boundary usually has multiple distances, and in the above example of the perpendicular distance, the number of perpendicular distances is the same as the number of reachable boundaries. Figure 3

[0108] ​The representative distance value for each interior point is determined based on its distances to all reachable boundaries. For example, the shortest distance between each interior point and all reachable boundaries can be used as the representative distance value; for instance... Figure 3 As shown, according to the above embodiment, the perpendicular distances between point C and each reachable boundary 1 to 4 are c1, c2, c3, and c4, respectively. Figure 3 It is evident that c1 is the shortest, and therefore the length of c1 is the representative distance value of the interior point C. Clearly, for an interior point, its representative distance value can only be one value.

[0109] Within a closed planar space, points generally exhibit the following characteristics: the vertical distances between the center point and all reachable boundaries are relatively even, with small maxima and large minima; conversely, the closer a point is to a reachable boundary from the center, the greater the difference in distance between that point and the boundary, with increasingly larger maxima and smaller minima. Figure 3 For example, Figure 3 The black borders in the diagram represent reachable boundary lines, designated as reachable boundary line 1, 2, 3, and 4. Point A is an interior point near the center of the passable area. The four interior points B, C, and D progressively approach reachable boundary line 1. The perpendicular distances from point A to the four reachable boundaries are a1, a2, a3, and a4, respectively; from point B to the four reachable boundaries are b1, b2, b3, and b4, respectively; from point C to the four reachable boundaries are c1, c2, c3, and c4, respectively; and from point D to the four reachable boundaries are d1, d2, d3, and d4, respectively. As can be seen from the diagram, a1 (=a3) > b1 > c1 > d1, and a4 (=a2) < b4 < c4 < d4. That is, from point A to point D, the minimum value of the perpendicular distance from the interior point to the four reachable boundary lines decreases, while the maximum value increases.

[0110] Therefore, it can be determined that the distance between internal points within multiple reachable boundaries can, to some extent, characterize the distribution information of the planar space.

[0111] Correspondingly, the maximum distance between an interior point and each reachable boundary can be used as the representative distance value of that interior point. Therefore, within the map space of the target area, the representative distance value of interior points closer to the center is the smallest, while the representative distance value of interior points further outwards and closer to a certain reachable boundary is the largest.

[0112] Alternatively, the minimum distance between an interior point and each reachable boundary can be used as the representative distance value for that interior point. Therefore, within the target area map space, the representative distance value of an interior point closer to the center is the largest, and the representative distance value of an interior point decreases as it moves outwards closer to a certain reachable boundary.

[0113] S203, determining the main structure of each room candidate in the map according to the representative distance values of the plurality of internal points within the reachable boundary in the map.

[0114] In the present disclosure, the "room candidate in the map" refers to a sub-region temporarily divided from the map during the map processing process, which is not overlapped with each other according to the positions of the internal points and the representative distance values thereof. It should be noted that the present disclosure only uses "room" as a general term for each sub-region obtained during the map processing process, and does not limit the scheme of the present disclosure to only be applied to the interior of a home. The sub-regions do not necessarily correspond to the actual functional areas in the room one by one.

[0115] Since the positions of the internal points and the representative distance values thereof reflect the relative positional relationship between the internal points and the reachable boundaries, a plurality of representative internal points can be selected according to the representative distance values of the internal points to form the main structure of the room candidate in the map. The main structure of the room candidate in the map can roughly reflect the general shape and position of each sub-region in the map on a smaller scale.

[0116] S204, expanding the main structure of each room candidate in the map to obtain each room in the map.

[0117] The room in the map refers to the region in the map after the main structure of the room candidate in the map is expanded outward and cannot be expanded any more.

[0118] Specifically, the main structure of each room candidate in the map is expanded, and the condition that the main structure cannot be expanded any more is that the boundary of the expanded main structure of the room candidate in the map (also referred to as the expanded main structure) will be in contact with the expanded main structure of the other room candidate in the map, forming a main structure boundary, or the boundary of the expanded main structure will be in contact with the reachable boundary of the map, obtaining a main structure boundary. The main structure boundary between the rooms in the map refers to the connection between two rooms in the map that are connected to each other. In the actual target area, it may correspond to a "door" between two actual rooms, but it may also be a relatively narrow part between two relatively wide parts of the same room, rather than an actual door. The main structure boundary obtained by the expanded main structure of the room candidate in the map meeting the reachable boundary of the map refers to the limit of each room in the map that is not in contact with other rooms in the map. The corresponding area of the actual target area may be a "wall" or a wall-adjacent obstacle (such as a wall-adjacent floor cabinet, sofa, TV cabinet, refrigerator, air conditioner, bed, etc.). Since the position of the wall in the actual target area and the wall-adjacent furniture is an obstacle position, it will block the expansion of the main structure of the room candidate in the map, and the unreachable position where the expansion of the main structure is blocked constitutes the main structure boundary.

[0119] The map partitioning method provided in the embodiment comprises: obtaining a map of a target area; the map comprising reachable boundaries of an autonomous mobile device in the target area; determining a representative distance value of an internal point in the map according to a distance between the internal point and each reachable boundary, for a plurality of internal points in the map within the reachable boundaries; determining a main structure of each candidate room in the map according to the representative distance values of the plurality of internal points in the map within the reachable boundaries; and expanding the main structure of each candidate room in the map to obtain each room in the map. According to the rule that the distance from the boundary to the center is increasingly greater, the distance value of each point is determined, and according to the characteristics of the distribution of the distance values of the points, the main structure of each candidate room is determined, and the boundary of each room is further determined through the expansion of the main structure of each candidate room, thereby completing the automatic partitioning of the map. The method does not rely on actual walls, doors, and the like in the space as partitioning markers, and can be applied to partitioning in more scenarios, and has higher partitioning accuracy.

[0120] In some embodiments, obtaining the map of the target area can further comprise: obtaining an original map of the target area; and pre-processing the original map to obtain the map of the target area.

[0121] The map constructed by the autonomous mobile device can not conform to the characteristics of the internal points and the boundary points, and thus, after obtaining the original map of the target area, the original map can be pre-processed to obtain the map of the target area.

[0122] The pre-processing manner can comprise binarization of an image, and the original map is converted into a binary image. In the binary image, the positions of obstacles such as walls that the autonomous mobile device cannot reach are processed as black (for example, pixel value 0), and the passable positions of the autonomous mobile device are processed as white (for example, pixel value 255). Specifically, a threshold T can be set, and the data of the image is divided into two parts by T: a pixel group greater than T and a pixel group less than T, thereby generating a corresponding binary image.

[0123] The original map can also be processed by a corrosion operation, a dilation operation, an opening operation, a closing operation, morphological processing, or the like to delete small outlines / obstacles or noise in the original map.

[0124] The principle of erosion operation is to use the convolution template B to perform convolution calculation on the image A to obtain the minimum value of the pixel points in the B covering area, and replace the pixel value of the reference point with the minimum value. A is referred to as the object to be processed, and B is referred to as the structure element. The convolution template B can be of any shape and size, and has a separately defined reference point, an anchor point, which is usually a square or a disc with a reference point, which can be referred to as a template or a mask. The processing process is to move the center point of the structure element B to a to obtain Ba, and if Ba is contained in A, the point a is recorded. The set of all a points satisfying the above condition is the result of the erosion of A by B. The set of a is within the range of A and smaller than A. Visually, the eroded image seems to have the outermost layer stripped off. For noise and small contours, the stripped-off part is eliminated. Based on the characteristics of the erosion operation, it can generally be used to eliminate noise and small contours in the image. The boundary in the map after the erosion operation is more clean, and the structure of the wall is more clear.

[0125] Dilation can be regarded as the dual operation of erosion, and its definition is: move the structure element B to a to obtain Ba, and if Ba hits the object to be processed X, record the point a. The set of all a points satisfying the above condition is referred to as the result of the dilation of X by B. The result of the dilation of X by B contains X, and visually, it is as if X has been dilated by one circle.

[0126] The opening operation is equivalent to first performing the erosion operation and then performing the dilation operation in principle. Based on the characteristics of the opening operation, it can generally be used to eliminate noise, such as isolated small points, burrs, etc., or separate two small objects connected by a thin line, while the total position and shape remain unchanged. Thus, the object contour is smoothed, the narrow discontinuity is disconnected, the small protrusions are eliminated, the small objects are eliminated, the objects are separated at the thin part, and the large object boundary is smoothed. The opening operation is equivalent to a filter based on geometric operation, and different sizes of the structure element will result in different filtering effects. Different selection of the structure element can result in different segmentation effects, i.e., different features in the image are extracted.

[0127] The closing operation is equivalent to first performing the dilation operation and then performing the erosion operation in principle. Based on the characteristics of the closing operation, it can generally be used to eliminate small holes in the foreground object, fill small cracks or small black points on the object, or close two small objects connected by a thin line, while the total position and shape remain unchanged. Thus, the narrow discontinuity is eliminated, the small holes are eliminated, the adjacent objects are connected, and the boundary is smoothed. The closing operation is equivalent to filtering by filling the concave corners of the image, and different sizes of the structure element will result in different filtering effects.

[0128] By preprocessing the original map, a target region map with clearer, more uniform boundaries and less noise can be obtained, which is more conducive to improving the accuracy of subsequent processing.

[0129] Reference Figure 4 The specific method for determining the main structure of each room candidate in the map according to the representative distance values of the plurality of internal points in the map can comprise: taking the shortest distance between the internal points and each reachable boundary as the representative distance value; repeatedly performing the following steps until there is no seed in the updated map: determining the internal point with the largest representative distance value in the map; determining a first threshold value according to the largest representative distance value in the map; the first threshold value is smaller than the largest representative distance value; determining at least one seed region in which the internal points with the representative distance values greater than the first threshold value and smaller than the largest representative distance value in the map are located; for each seed region, if the internal point with the largest representative distance value is located in the seed region, taking the seed region as a seed of a room candidate in the map; inflating the seed of the room candidate in the map to obtain the main structure of the room candidate in the map; and deleting the main structure of the room candidate in the map from the map to obtain the updated map.

[0130] According to the above, when the shortest distance between the internal points and each reachable boundary is taken as the representative distance value, if the internal point with the largest representative distance value appears in a certain region, the region is a room candidate in the map. Thus, in this embodiment, the main structure of the room candidate in the map is determined by searching for the internal point with the largest representative distance value.

[0131] Specifically, the position of the internal point with the largest representative distance value in the map can be determined first, and the position is the center position of one of the room candidates. Then, a first threshold value is determined according to the largest representative distance value, and the region in which the points with the representative distance values between the largest representative distance value and the first threshold value in the map are located is the central region of the candidate room and is taken as a seed region. There can be multiple seed regions in the map, but only the seed region in which the point corresponding to the largest representative distance value is located is the seed of the candidate room. The main structure of the room candidate in the map can be obtained by inflating the seed. After the main structure of the first room candidate in the map is saved, it can be deleted from the map, and an updated map can be obtained. If there is still a seed in the updated map, the above operation can be repeated based on the updated map to determine the internal point with the largest representative distance value in the remaining part of the map, and then the main structure of a second room candidate, the main structure of a third room candidate, and so on until there is no seed in the updated map, for example, the size of the largest representative distance value determined in the Nth time does not meet the preset condition of the main structure, or the main structure of the room candidate determined in the Nth time does not meet the preset condition of the main structure, and it can be determined that there is no seed in the updated map.

[0132] For example, the first threshold value can be set by including a number of (e.g., 6) interior points in the seed region whose representative distance values are greater than the first threshold value, so that these interior points in the seed region can form a general outline of a seed region.

[0133] Another way to determine the seed of a candidate room in a map is to determine the position of an interior point with the maximum representative distance value in the map, and then determine the first threshold value according to the maximum representative distance value. The first threshold value is less than the maximum representative distance value. Then, taking the position of the interior point with the maximum representative distance value as a reference, points are traversed one by one outward, and the traversal stops when the representative distance value of any point and other points around it are all less than or equal to the first threshold value. The region of the traversed points includes the interior point with the maximum representative distance value, and thus the seed region is the seed of the candidate room in the map.

[0134] In this embodiment, the first threshold value can be n times the maximum representative distance value, and n is an empirical value, which is usually a positive number less than 1. For example, if the maximum representative distance value is 1000 and n is 0.8, then the first threshold value is 800. In some embodiments, the value of n is also limited by the minimum representative distance value in the map, that is, the value of n cannot make the first threshold value less than the minimum representative distance value in the map.

[0135] In addition, a region that can be referred to as a single room can have geometric parameters such as length and area that need to meet certain conditions. For example, the area is not less than 4 square meters. Correspondingly, conditions such as the maximum representative distance value being not less than 2 meters or the main structure area of the candidate room being not less than 3 square meters can be required. Correspondingly, this condition can be used as Figure 4 The specific judgment condition of "whether there is a seed in the updated map" in the illustrated embodiment.

[0136] In some embodiments, corresponding to the above-mentioned embodiments, the main structure of each candidate room in the map is determined according to the representative distance values of the plurality of interior points in the reachable boundary of the map, including: taking the longest distance between the interior point and each reachable boundary as the representative distance value; repeatedly performing the following steps until there is no seed in the updated map: determining the interior point with the minimum representative distance value in the map; determining the second threshold value according to the minimum representative distance value in the map, the second threshold value being greater than the minimum representative distance value; determining at least one seed region in which the interior points with representative distance values less than the second threshold value and greater than the minimum representative distance value are located; for each seed region, if the interior point with the minimum representative distance value is located in the seed region, the seed region is taken as the seed of the candidate room in the map; the main structure of the candidate room in the map is dilated to obtain the main structure of the candidate room in the map; after the main structure of the candidate room in the map is saved, the main structure of the candidate room in the map is deleted from the map to obtain the updated map.

[0137] According to the above, when the shortest distance between the interior point and each reachable boundary is taken as the representative distance value, the area in which the interior point with the minimum representative distance value appears is the candidate room in the graph. Thus, in this embodiment, the main structure of the candidate room in the graph is determined by searching for the interior point with the minimum representative distance value.

[0138] Specifically, the position of the interior point with the minimum representative distance value in the graph can be determined first, and this position is the position of the center point of one of the candidate rooms in the graph. Then, a second threshold value is determined according to the minimum representative distance value. The area in which the points with representative distance values between the minimum representative distance value and the second threshold value in the graph is the central area of the candidate room in the graph, and is taken as the seed area. There can be multiple seed areas in the graph, but only the seed area in which the point with the minimum representative distance value is located is the area in the candidate room in the graph, i.e., the seed of the candidate room in the graph. The seed is dilated to obtain the main structure of the candidate room in the graph. After the main structure of the first room is saved, it is exemplary to be deleted from the graph to obtain an updated graph. If there is still a seed in the updated graph, the above operation can be repeated based on the updated graph to continue to determine the interior point with the minimum representative distance value in the remaining part of the graph, and further determine the main structure of the second candidate room, the main structure of the third candidate room, and so on, until there is no longer a seed in the updated graph, for example, the size of the minimum representative distance value determined in the Nth time does not meet the preset condition of the main structure, or the main structure of the candidate room determined in the Nth time does not meet the preset condition of the main structure.

[0139] Exemplarily, the setting condition of the second threshold value can be that the seed area includes a plurality of (such as 6) interior points with representative distance values less than the second threshold value, so that these interior points in the seed area can constitute the general outline of the seed area.

[0140] Another way to determine the seed of the candidate room is to determine the position of the interior point with the minimum representative distance value in the graph, and then determine the second threshold value according to the minimum representative distance value. Then, the position of the interior point with the minimum representative distance value is taken as the reference, and the points are traversed one by one outward, and the traversal is stopped when the representative distance value of any point and other points around it is greater than or equal to the second threshold value. The area in which the traversed points are located includes the interior point with the minimum representative distance value, and thus the seed area is the seed of the candidate room.

[0141] The second threshold value can be n times of the minimum representative distance value, where n is an empirical value, and is usually a positive number greater than 1, for example, when the minimum representative distance value is 100 and n is 2, the second threshold value is 200. Meanwhile, in this case, the value of n is also limited by the maximum representative distance value in the map, that is, the value of n cannot make the second threshold value greater than the maximum representative distance value in the map.

[0142] In some embodiments, the specific manner of expanding the main structure of the candidate room in each graph to obtain the room in each graph can include: simultaneously expanding the main structure of the candidate room in each graph at the same preset rate to become an expanded main structure; if the expanded main structure of the candidate room in the graph meets the expanded main structure of the candidate room in another graph to generate a main structure intersection, or if the expanded main structure of the candidate room in the graph meets the accessible boundary of the map to generate a main structure boundary, then stopping the expansion to obtain the room in each graph.

[0143] The main structure of the candidate room in each graph simultaneously expands at the same preset rate to become an expanded main structure, which means that the main structure of the candidate room in each graph simultaneously expands outward in each direction while maintaining the same expansion rate. The main structure of the candidate room in each graph during expansion is referred to as an expanded main structure. The condition for stopping the expansion is that a certain expanded main structure meets another expanded main structure to generate a main structure intersection, or the expanded main structure meets the accessible boundary in the map to generate a main structure boundary.

[0144] Specifically, based on the initially obtained map of the target area, the main structure of the candidate room in each graph that has been determined and saved can be filled into the corresponding position in the map. When filling, for example, the main structure of each candidate room in the graph is filled with a unique identifier that is different from the main structure of the candidate room in other graphs. For example, it is filled with different pixel values or different characters (different letters or different numbers).

[0145] Then, for the main structure of each candidate room in each graph, expansion is performed outward based on each point in the main structure. For each point, it is determined whether the neighboring points around the point have the unique identifier of the main structure of the candidate room in the graph, or have the unique identifier of the main structure of the candidate room in another graph, or are points on the reachable boundary, or are interior points. During the expansion, if it is determined that there are interior points among the neighboring points around the point, the interior points are added with the unique identifier of the main structure of the candidate room in the graph, so as to become points of the main structure of the candidate room in the graph. In this way, the outward expansion of the main structure of the candidate room in the graph is achieved. During the expansion, if it is determined that there are no interior points among the neighboring points around the edge of the expanded main structure of the candidate room in the graph, the expansion is stopped. There are usually two cases to stop the expansion, the first case is that the points outside the expanded main structure of the candidate room in the graph have the unique identifier of the candidate room in another graph, at this time the corresponding case is that the expanded main structure of the candidate room in the graph meets the expanded main structure of the candidate room in another graph, and a main structure boundary is generated; the second case is that the points outside the expanded main structure of the candidate room in the graph are unreachable points, that is, the main structure of the candidate room in the graph meets the reachable boundary of the map to form a main structure boundary.

[0146] During the expansion, the expansion rate can be controlled by the range of the point modified each time. Referring to Figure 5 , Figure 5 The point marked 1 in FIG. 1 is a point of the main structure of a candidate room a in a graph. Figure 5 (a) in FIG. 1 is a schematic diagram of the initial state of the main structure of a, Figure 5 (b) in FIG. 1 is a schematic diagram of the expanded main structure of a after expansion once at a first rate, the first rate being outward expansion by one unit (for example, the unit can be one pixel point, and the first rate can be 1 pixel / time or 1 pixel / second). Figure 5 (c) in FIG. 1 is a schematic diagram of the expanded main structure of a after expansion once at a second rate, the second rate being outward expansion by two units (for example, the unit can be two pixel points, and the second rate can be 2 pixels / time or 2 pixels / second).

[0147] After the expansion is completed, the rooms in the graphs are obtained. For the main structure boundary between the rooms in the graphs or the main structure boundary between the rooms in the graphs and the reachable boundary of the target map, further correction can be performed.

[0148] In other embodiments, after the main structure of each candidate room in each graph is expanded to obtain the rooms in the graphs, the rooms in the graphs can be fused to obtain the rooms in the fused graphs, so that the final map is more consistent with the functional area division in the actual target area.

[0149] In the process of expanding the candidate room in the map to the room in the map, there can be some rooms that are incorrectly divided. The characteristics of the candidate room can be analyzed, and the rooms that can be incorrectly divided can be merged to correct the division result. The main structure intersection between the rooms in the map, as described above, refers to the connection between two connected rooms in the map in the actual target area, which can correspond to a "door" between two actual rooms, but can also be a narrow part between two wide parts of the same room, rather than an actual door. The purpose of merging is to merge the two rooms in the map on both sides of the main structure intersection that does not correspond to the "door" in the actual functional area into one room in the map, so that the divided map is more consistent and accurate with the division of the functional area of the actual target area. The main structure boundary obtained by the expansion of the main structure of the candidate room in the map and the reachable boundary of the map, as described above, refers to the limit of each room in the map that does not contact other rooms in the map. Its corresponding area in the actual target area can be a "wall" or a wall-mounted obstacle (such as a wall-mounted floor cabinet, sofa, TV cabinet, refrigerator, air conditioner, bed, etc. furniture); Since the position of the wall and the wall-mounted furniture in the actual target area is the position of the obstacle, it will block the expansion of the main structure of the candidate room in the map, and the unreachable position where the main structure is blocked constitutes the main structure boundary. In the case where the wall or wall-mounted furniture divides two actual rooms into two functional areas, the position corresponding to the obstacle in the map is not a line, but a two-dimensional shape, so the obstacle must form a main structure boundary with the rooms on both sides; Therefore, the two areas where the main structure boundaries of the obstacle on both sides meet cannot be merged. The following describes the merging of the rooms in the map based on the definition and description of the main structure intersection and the main structure boundary.

[0150] In one embodiment, with reference to Figure 6 The way of merging the rooms in the map to obtain the merged room in the map can include: obtaining the area of each room in the map and the main structure intersection between the rooms in the map; determining the adjacent relationship between the rooms in the map according to the main structure intersection between the rooms in the map, to determine adjacent rooms; comparing the area of the room in the map with the area threshold value, if the area of the room in the map is less than the area threshold value, the room in the map is determined as a target room; merging the target room and one of its adjacent rooms into a merged room in the map.

[0151] In the embodiment, whether a room in a map is a room in a map that is wrongly divided is determined by judging the area of the room in a map. If the area of a room in a map is too small, it is impossible to be a single room in reality, but a part of a room that is wrongly divided. In the embodiment, a room in a map whose area is obviously smaller than a reasonable threshold is called a target room. After the target room is determined, the room in a map that the target room can belong to can be determined based on the rooms in a map that are adjacent to the target room, and the two rooms in a map are fused into one room in a map. The room in a map obtained after the fusion is called a fused room in a map for distinction. Therefore, the final map after the fusion step is a map in which the rooms in a map that do not need to be fused and the fused rooms in a map constitute the divided map. Since the main structure intersection represents the connectivity between the two rooms in a map on both sides, that is, if the two rooms in a map are connected, there must be a main structure intersection between them, the rooms in a map that can be fused can be determined based on the target room and the rooms in a map that are adjacent to the target room and share the main structure intersection. Since the target room can have multiple rooms in a map as adjacent rooms, one of the adjacent rooms can be selected to be fused with the target room, or the adjacent room with the largest area can be selected to be fused with the target room.

[0152] For example, there is a main structure intersection L1 between a room in a map H1 and a room in a map H2, which indicates that the room in a map H1 and the room in a map H2 are connected to each other in the actual target area, and thus the room in a map H1 and the room in a map H2 are adjacent to each other. If the area of the room in a map H1 is smaller than the area threshold, the room in a map H1 is a target room, and the room in a map H1 and the room in a map H2 adjacent to the room in a map H1 are fused into one room, that is, a fused room in a map H3 is obtained.

[0153] In addition, Figure 6 The order in the above embodiment is only an example. The area of a room in a map can be compared with the area threshold to determine whether the room in a map is a target room, and after it is determined that the area of the room in a map is smaller than the area threshold, the room in a map is determined to be a target room. The adjacent relationship between the rooms in a map can be determined based on the main structure intersection between the rooms in a map to determine the adjacent rooms, and finally the target room and the adjacent rooms are fused.

[0154] After the partitioning method of each of the above embodiments is executed, the final map corresponding to the partitioning result can be displayed on the terminal device of the user for the user to view. Meanwhile, the user can also be provided with an adjustment operation of the partitioning result. If the user considers that the partitioning is wrong, the user can input an adjustment instruction on the interactive interface or directly adjust the final map on the UI interface to further adjust the size and position of the room. In addition, the user can input a label instruction through the interactive interface to add a label to a certain room to indicate the room attribute. When establishing a task for the autonomous mobile device, a targeted task can be created based on the room attribute. For example, "clean the living room". The autonomous mobile device can determine, through recognition of the instruction, that the task to be executed is "cleaning" and the range of executing the task is "living room".

[0155] In one specific embodiment, the target area is partitioned, and a map of the target area is as shown in Figure 7a . This map is a pre-processed image, and it can be seen that the boundaries in the image are relatively clear and almost no noise.

[0156] For the handled map, the minimum value of the distance between each point in the boundary and each boundary is calculated as the distance value of the point, and a corresponding distance map is generated, as shown in Figure 7b .

[0157] The position of the point with the maximum distance value in the map is determined; the area of the point with a distance value greater than 60% of the maximum distance value around the point is taken as the seed of the first room. The seed area of the first room is greater than 2 square meters, the length is greater than 1 meter, and the width is greater than 0.5 meter. The obtained seed of the first room is dilated to obtain the position and shape of the main structure of the first room, and then the part of the main structure of the first room in the handled map is removed to obtain a new handled map, as shown in Figure 7c . The corresponding new distance map is as shown in Figure 7d .

[0158] The above operations are repeatedly executed until there is no seed meeting the above geometric conditions in the handled map. The handled maps generated in the process are as shown in Figures 7e to 7l .

[0159] The main structures of all the rooms obtained are superimposed on the handled map to construct a seed map, and each room is marked with a different gray value, as shown in Figure 7m .

[0160] Based on the seed map, the main structure is expanded using the flood fill algorithm to obtain a label map of each room, as shown inFigure 7n As such, the partitioning of the target area is completed.

[0161] Figure 8 A structural schematic diagram of a map partitioning apparatus provided for an embodiment of the present disclosure is shown in FIG. 8. As shown in FIG. 8, the map partitioning apparatus 800 of the present embodiment can include an acquisition module 801, a representative distance value determination module 802, a main structure determination module 803, and a main structure expansion module 804. Figure 8

[0162] The acquisition module 801 is configured to acquire a map of a target area; the map includes reachable boundaries of a self-moving device in the target area.

[0163] The representative distance value determination module 802 is configured to, for a plurality of internal points within the reachable boundaries in the map, determine a representative distance value of each internal point according to a distance between the internal point and each reachable boundary.

[0164] The main structure determination module 803 is configured to determine a main structure of each map candidate room according to the representative distance values of the plurality of internal points within the reachable boundaries in the map.

[0165] The main structure expansion module 804 is configured to expand the main structure of each map candidate room to obtain a map room.

[0166] Optionally, the main structure determination module 803 is specifically configured to:

[0167] determine the shortest distance between the internal point and each reachable boundary as the representative distance value.

[0168] repeat the following steps until there is no seed in the updated map:

[0169] determine an internal point with the largest representative distance value in the map;

[0170] determine a first threshold value according to the largest representative distance value in the map; the first threshold value is smaller than the largest representative distance value.

[0171] determine at least one seed region in which the internal points with the representative distance values greater than the first threshold value and smaller than the largest representative distance value in the map are located.

[0172] for each seed region, if the internal point with the largest representative distance value is located in the seed region, take the seed region as a seed of a map candidate room.

[0173] inflate the seed of the map candidate room to obtain the main structure of the map candidate room.

[0174] delete the main structure of the map candidate room from the map to obtain an updated map.

[0175] ​Optionally, the main structure determination module 803 is specifically configured to:

[0176] taking the longest distance among distances of the internal point to each reachable boundary as a representative distance value;

[0177] repeating the following steps until there is no seed in the updated map:

[0178] determining an internal point with the minimum representative distance value in the map;

[0179] determining a second threshold according to the minimum representative distance value in the map, the second threshold being greater than the minimum representative distance value;

[0180] determining at least one seed region in which an internal point with a representative distance value less than the second threshold and greater than the minimum representative distance value is located;

[0181] for each seed region, if the internal point with the minimum representative distance value is located in the seed region, taking the seed region as a seed of a candidate room in the map;

[0182] performing inflation on the seed of the candidate room in the map to obtain a main structure of the candidate room in the map;

[0183] deleting the main structure of the candidate room in the map from the map to obtain an updated map.

[0184] Optionally, the main structure expansion module 804 is specifically configured to:

[0185] each main structure of the candidate room in the map expands at the same preset rate to become an expanded main structure;

[0186] if the expanded main structure of the candidate room in the map meets the expanded main structure of another candidate room in the map to generate a main structure boundary, or if the expanded main structure of the candidate room in the map meets a reachable boundary of the map, the expansion is stopped to obtain each room in the map.

[0187] Optionally, the apparatus further includes a fusion module 805 configured to, after the main structure of each candidate room in the map is expanded to obtain each room in the map, fuse each room in the map to obtain a fused room in the map.

[0188] Optionally, the fusion module 805, when fusing each candidate room in the map, is specifically configured to:

[0189] obtain an area of each room in the map and a main structure boundary between the rooms in the map;

[0190] determine a neighboring relationship between the rooms in the map according to the main structure boundary between the rooms in the map to determine adjacent rooms;

[0191] The area of the room in the graph is compared with the size of the area threshold value, and if the area of the room in the graph is less than the area threshold value, the room in the graph is determined as the target room.

[0192] The target room and one adjacent room thereof are fused into a room in a fused graph.

[0193] Optionally, the obtaining module 801 is specifically configured to:

[0194] obtain an original map of the target area;

[0195] perform preprocessing on the original map to obtain the map of the target area.

[0196] The device of the embodiment can be used to execute the method of any of the above embodiments, and has similar implementation principles and technical effects, which will not be described here.

[0197] Figure 9 A structural schematic diagram of an autonomous mobile device according to an embodiment of the present disclosure is shown in FIG. 9. Figure 9 As shown in FIG. 9, the autonomous mobile device 900 according to the embodiment can include a memory 901 and a processor 902.

[0198] The memory 901 is configured to store program instructions.

[0199] The processor 902 is configured to invoke and execute the program instructions in the memory 901 to execute the method of any of the above embodiments, and has similar implementation principles and technical effects, which will not be described here.

[0200] The present disclosure further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the method according to any of the above embodiments is implemented.

[0201] The present disclosure further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the method according to any of the above embodiments is implemented.

[0202] In all the above embodiments, the map of the target area mentioned is explained and described as a dot matrix, but is not limited to a dot matrix, and can also be a vector graph. In fact, those skilled in the art can explicitly apply the technical solutions described in the present disclosure to dot matrix, vector graph, and any kind of existing technology known to those skilled in the art that can apply the above technical solutions.

[0203] Those skilled in the art can understand that all or part of the steps of the foregoing method embodiments can be completed by program instruction related hardware. The foregoing program can be stored in a computer readable storage medium. The program executes to perform the steps of the foregoing method embodiments; and the foregoing storage medium includes various storage media that can store program codes, such as ROM, RAM, magnetic disk, or optical disk.

[0204] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than limit them; although the present disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present disclosure.

Claims

1. A method of map partitioning, characterized by, The method comprises: obtaining a map of a target area; the map comprises reachable boundaries of an autonomous mobile device in the target area; the reachable boundaries are intersection limits of unreachable areas and passable areas in the target area; for a plurality of internal points within the reachable boundaries in the map, determining a representative distance value of the internal points according to distances of the internal points to each reachable boundary; wherein the representative distance value of the internal point is the shortest distance or the longest distance among the distances of the internal point to each reachable boundary; determining a main structure of each map candidate room according to the representative distance values of the plurality of internal points within the reachable boundaries in the map; expanding the main structure of each map candidate room to obtain each map room; if the representative distance value of the internal point is the shortest distance among the distances of the internal point to each reachable boundary, the determining of the main structure of each map candidate room according to the representative distance values of the plurality of internal points within the reachable boundaries in the map comprises: repeating the following steps until there is no seed in the updated map: determining an internal point with the maximum representative distance value in the map; determining a first threshold value according to the maximum representative distance value in the map; the first threshold value is less than the maximum representative distance value; determining at least one seed area in which the internal point with the maximum representative distance value is located and in which the representative distance value is greater than the first threshold value and less than the maximum representative distance value; for each seed area, if the internal point with the maximum representative distance value is located in the seed area, taking the seed area as a seed of a map candidate room; inflating the seed of the map candidate room to obtain the main structure of the map candidate room; deleting the main structure of the map candidate room from the map to obtain an updated map.

2. The method of claim 1, wherein, if the representative distance value of the internal point is the longest distance among the distances of the internal point to each reachable boundary, the determining of the main structure of each map candidate room according to the representative distance values of the plurality of internal points within the reachable boundaries in the map comprises: repeating the following steps until there is no seed in the updated map: determining an internal point with the minimum representative distance value in the map; determining a second threshold value according to the minimum representative distance value in the map; the second threshold value is greater than the minimum representative distance value; determining at least one seed area in which the internal point with the minimum representative distance value is located and in which the representative distance value is less than the second threshold value and greater than the minimum representative distance value; for each seed area, if the internal point with the minimum representative distance value is located in the seed area, taking the seed area as a seed of a map candidate room; inflating the seed of the map candidate room to obtain the main structure of the map candidate room; deleting the main structure of the map candidate room from the map to obtain an updated map.

3. The method according to any of claims 1-2, characterized in that, the expanding of the main structure of each map candidate room to obtain each map room comprises: the main structure of each map candidate room is expanded at the same preset rate to become an expanded main structure; If the expansion main structure of the candidate room in the graph meets the expansion main structure of the candidate room in the other graph to generate a main structure intersection, or if the expansion main structure of the candidate room in the graph meets the accessible boundary of the map, the expansion is stopped to obtain the room in each graph.

4. The method according to any one of claims 1-2, characterized in that, After the main structure of the candidate room in each graph is expanded to obtain the room in each graph, the method further comprises: fusing the room in each graph to obtain a room in a fused graph.

5. The method of claim 4, wherein, The fusing the room in each graph to obtain a room in a fused graph comprises: obtaining the area of each room in the graph and the main structure intersection between the rooms in the graph; determining the adjacent relationship between the rooms in the graph according to the main structure intersection between the rooms in the graph, to determine adjacent rooms; comparing the area of the room in the graph with the area threshold value, and if the area of the room in the graph is less than the area threshold value, determining the room in the graph as a target room; fusing the target room with one adjacent room thereof to obtain a room in a fused graph.

6. The method according to any one of claims 1-2, characterized in that, The obtaining the map of the target region comprises: obtaining an original map of the target region; preprocessing the original map to obtain the map of the target region.

7. A map partitioning apparatus characterized by comprising: The method comprises: a obtaining module, configured to obtain the map of the target region; the map comprises the accessible boundary of the autonomous mobile device in the target region; the accessible boundary is the intersection limit between the unreachable region and the passable region in the target region; a representative distance value determining module, configured to, for a plurality of internal points in the accessible boundary in the map, determine the representative distance value of the internal point according to the distance between the internal point and each accessible boundary; wherein the representative distance value of the internal point is the shortest or longest distance between the internal point and each accessible boundary; a main structure determining module, configured to determine the main structure of each candidate room in the graph according to the representative distance value of the plurality of internal points in the accessible boundary in the map; a main structure expansion module, configured to expand the main structure of each candidate room in the graph to obtain a room in each graph; if the shortest distance between the internal point and each accessible boundary is taken as the representative distance value, the main structure determining module is specifically configured to: repeat the following steps until there is no seed in the updated map: determine the internal point with the maximum representative distance value in the map; determine a first threshold value according to the maximum representative distance value in the map; the first threshold value is less than the maximum representative distance value; determine at least one seed region in which the internal point with the representative distance value greater than the first threshold value and less than the maximum representative distance value is located; for each seed region, if the internal point with the maximum representative distance value is located in the seed region, take the seed region as a seed of the candidate room in the graph; dilate the seed of the candidate room in the graph to obtain the main structure of the candidate room in the graph; delete the main structure of the candidate room in the graph from the map to obtain an updated map.

8. An autonomous mobile device, comprising: The method comprises: a memory, configured to store program instructions; a processor, configured to invoke and execute the program instructions in the memory, and execute the method in any one of claims 1-6. The method comprises: a memory, configured to store program instructions; a processor, configured to invoke and execute the program instructions in the memory, and execute the method in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, and the computer program, when executed by a processor, implements the method in any one of claims 1-6.

10. A computer program product comprising a computer program, characterized in that, The computer program, when executed by a processor, implements the method in any one of claims 1-6.

Images