Map construction method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a map construction method and device, electronic equipment and a storage medium, and the method comprises the steps: determining an alternative region according to the position of a non-transparent obstacle included in a three-dimensional map; acquiring environment information corresponding to the alternative area; in response to the environmental information conforming to at least one of the environmental characteristics, determining the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to each environmental transparent obstacle corresponding to the conforming environmental characteristics; and marking an environment transparent obstacle on the three-dimensional map to obtain the three-dimensional map comprising the transparent obstacle. Since the environment information around the transparent obstacle has certain environment characteristics, whether the environment information accords with the environment characteristics can be determined, and the environment transparent obstacle is determined from the alternative transparent obstacles. Therefore, the transparent obstacle can be accurately identified, a map including accurate transparent obstacle information is constructed, and subsequent normal operation of the robot is ensured.

Description

Technical Field

[0001] The present invention relates to the field of map exploration technology, and in particular to a map construction method, device, electronic equipment and storage medium. Background Art

[0002] When a robot is working in a new scene, it usually needs to record a 3D map of the scene, or build a 3D map by the robot itself. In this way, the robot can know the location of obstacles and the size of the scene, which is convenient for path planning when performing subsequent tasks. The traditional 3D map construction method can be achieved through binocular imaging or structured light, that is, collecting 2D images of the scene, determining the 3D information of the scene based on the 2D images, and then building a 3D map.

[0003] However, for transparent obstacles, such as floor-to-ceiling glass walls and floor-to-ceiling doors, since the surface of transparent obstacles hardly reflects light, it is difficult to effectively identify transparent obstacles using the above-mentioned traditional three-dimensional map construction method, resulting in a lack of transparent obstacle data in the three-dimensional map, or incorrect data recording of transparent obstacles, which will directly affect the mapping effect and thus affect the subsequent operations of the robot. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a map construction method, device, electronic device and storage medium to accurately identify transparent obstacles and construct a map including accurate transparent obstacle information, thereby ensuring the subsequent normal operation of the robot. The specific technical solution is as follows:

[0005] In a first aspect, an embodiment of the present invention provides a map construction method, the method comprising:

[0006] Determine a candidate area according to the position of the non-transparent obstacle included in the three-dimensional map, wherein the candidate area includes the area where the candidate transparent obstacle is located;

[0007] Obtaining environmental information corresponding to the candidate area;

[0008] In response to the environmental information being consistent with at least one of the environmental characteristics corresponding to the transparent obstacle, for each environmental transparent obstacle corresponding to the environmental characteristic, determining a spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle, wherein the spatial attribute includes at least size and position;

[0009] According to the spatial attributes of the environmental transparent obstacles, the environmental transparent obstacles are marked on the three-dimensional map to obtain a three-dimensional map including the transparent obstacles.

[0010] Optionally, the environmental information at least includes an environmental image collected within a preset range of the candidate area, and the environmental image includes a visible light image and a depth image;

[0011] After the step of obtaining the environment information corresponding to the candidate area, the method further includes:

[0012] Extracting image features of the associated object from the visible light image and the depth image, and determining the spatial attributes of the associated object based on the image features, wherein the function of the associated object is compatible with the function of the transparent obstacle and at least a portion of the associated object is made of a non-transparent material;

[0013] Based on the spatial relationship between the spatial attributes of the associated object and the candidate area, determining whether the connection relationship between the associated object and the candidate transparent obstacle conforms to the connection rule corresponding to the transparent obstacle and the associated object;

[0014] In response to the connection relationship between the associated object and the candidate transparent obstacle conforming to the connection rule corresponding to the transparent obstacle and the associated object, it is determined that the environmental information of the candidate transparent obstacle conforms to the environmental characteristics corresponding to the transparent obstacle.

[0015] Optionally, the step of determining whether the connection relationship between the associated object and the candidate transparent obstacle conforms to the connection rule corresponding to the transparent obstacle and the associated object based on the spatial relationship between the spatial attribute of the associated object and the candidate area includes:

[0016] Determining the category of the associated object according to the image features of the associated object extracted from the visible light image, and determining the connection rule corresponding to the category of the associated object based on a preset correspondence between the category and the connection rule;

[0017] Based on the spatial relationship between the spatial attributes of the associated object and the candidate area, it is determined whether the connection relationship between the associated object and the candidate transparent obstacle conforms to the determined connection rule.

[0018] Optionally, the step of determining the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle includes:

[0019] Determining the category of the associated object according to the image features of the associated object extracted from the visible light image, and determining the method for determining the spatial attribute corresponding to the category of the associated object based on a correspondence between a preset category and a method for determining the spatial attribute;

[0020] The spatial attribute of the environmental transparent obstacle is determined according to the environmental information corresponding to the environmental transparent obstacle and the method for determining the spatial attribute.

[0021] Optionally, the step of determining the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle and the method for determining the spatial attribute includes at least one of the following:

[0022] In response to the category of the associated object being a door or window frame, the size and position of the door or window frame and the area enclosed by it are respectively determined as the size and position of the environmental transparent obstacle;

[0023] In response to the category of the associated object not being a door or window frame, the edges of the non-transparent obstacles around the environmental transparent obstacle are determined based on the connection pattern between the environmental transparent obstacle and the associated object; the size and position of the area formed by the edges of the non-transparent obstacles are respectively determined as the size and position of the environmental transparent obstacle.

[0024] Optionally, the step of obtaining the environment information corresponding to the candidate area includes:

[0025] Taking a preset position in the candidate area as a center point, determining a region to be collected according to preset parameters, wherein the region to be collected includes the candidate area;

[0026] Obtain a visible light image and a depth image of the area to be collected.

[0027] Optionally, the environmental information includes at least historical trajectories of personnel collected within a preset range of the candidate area;

[0028] After the step of obtaining the environment information corresponding to the candidate area, the method further includes:

[0029] Determine whether there is a detour trajectory in the historical trajectory of the person; in response to the presence of a detour trajectory in the historical trajectory of the person, determine that the environmental information meets the environmental characteristics corresponding to the transparent obstacle, wherein the detour trajectory is a trajectory that moves from one side of the candidate area to the opposite side in a non-linear manner; or,

[0030] Determine whether there is a walking trajectory in the historical trajectory of the personnel; in response to the existence of a walking trajectory in the historical trajectory of the personnel, obtain the personnel action of the walking personnel corresponding to the walking trajectory; in response to the personnel action including a door opening action or a door closing action, determine that the environmental information meets the environmental characteristics corresponding to the transparent obstacle, wherein the walking trajectory is a trajectory moving in a straight line from one side of the alternative area to the opposite side.

[0031] Optionally, the step of determining the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle includes:

[0032] Determine the edge of the non-transparent obstacle around the transparent obstacle in the environment;

[0033] The size and position of the area formed by the edges of the non-transparent obstacles are respectively determined as the size and position of the transparent obstacles in the environment.

[0034] Optionally, the three-dimensional map is a map corresponding to a current scene in which the robot is located, and the robot includes a ranging sensor;

[0035] Before the step of marking the environmental transparent obstacles on the three-dimensional map according to the spatial attributes of the environmental transparent obstacles to obtain a three-dimensional map including the transparent obstacles, the method further includes:

[0036] Acquire a first distance between the robot and the transparent obstacle in the environment by using the distance measuring sensor;

[0037] Obtaining a second distance between the robot and the environmental transparent obstacle according to the position of the robot and the position included in the spatial attribute of the environmental transparent obstacle;

[0038] In response to the difference between the first distance and the second distance being not greater than a preset threshold, the step of marking the environmental transparent obstacle in the three-dimensional map according to the spatial attributes of the environmental transparent obstacle to obtain a three-dimensional map including the transparent obstacle is performed.

[0039] Optionally, the step of determining the candidate area according to the positions of the non-transparent obstacles included in the three-dimensional map includes at least one of the following:

[0040] Controlling the robot to move in the current scene and collect an environment image, and constructing a three-dimensional map of the current scene based on the position of the robot when collecting the environment image and the scene information included in the environment image; and determining, according to the positions of the non-transparent obstacles included in the three-dimensional map, an area in the three-dimensional map where at least part of the non-transparent obstacles do not exist as a candidate area;

[0041] Acquire a pre-built three-dimensional map of the current scene; and determine, according to the positions of the non-transparent obstacles included in the three-dimensional map, areas in the three-dimensional map where at least part of the non-transparent obstacles do not exist as candidate areas;

[0042] During the movement of the robot in the current scene, an environmental image is collected and the position of obstacles is detected by a ranging sensor; a three-dimensional map of the current scene is constructed based on the position of the robot when the environmental image is collected and the scene information included in the environmental image; the position of each obstacle is determined based on the distance of the obstacle detected by the ranging sensor; and an alternative area is determined based on the determined position of the obstacle and the position of the non-transparent obstacle included in the three-dimensional map.

[0043] In a second aspect, an embodiment of the present invention provides a map construction device, the device comprising:

[0044] An area determination module, used to determine a candidate area according to the position of the non-transparent obstacles included in the three-dimensional map, wherein the candidate area includes an area where the candidate transparent obstacles are located;

[0045] An information acquisition module, used to acquire environmental information corresponding to the candidate area;

[0046] an attribute determination module, configured to determine, in response to the environmental information being in accordance with at least one of the environmental characteristics corresponding to the transparent obstacle, for each environmental transparent obstacle corresponding to the environmental characteristic, a spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle, wherein the spatial attribute includes at least size and position;

[0047] The map construction module is used to mark the environmental transparent obstacles in the three-dimensional map according to the spatial attributes of the environmental transparent obstacles, so as to obtain a three-dimensional map including the transparent obstacles.

[0048] In a third aspect, an embodiment of the present invention provides an electronic device, including a processor and a memory, wherein the processor communicates with the memory;

[0049] Memory, used to store computer programs;

[0050] The processor is used to implement any method step described in the first aspect when executing the program stored in the memory.

[0051] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps described in any one of the first aspects are implemented.

[0052] Beneficial effects of the embodiments of the present invention:

[0053] In the solution provided by the embodiment of the present invention, since the three-dimensional map includes the position of non-transparent obstacles, and the transparent obstacles in the scene do not overlap with the non-transparent obstacles, the alternative area where the alternative transparent obstacles may exist can be determined based on the position of the non-transparent obstacles included in the three-dimensional map. Since the environmental information around the transparent obstacle has certain environmental characteristics, it is possible to determine whether the environmental information corresponding to the alternative area meets the environmental characteristics, so as to accurately identify the environmental transparent obstacles among the alternative transparent obstacles. Furthermore, the size and position of the environmental transparent obstacle are determined, and the environmental transparent obstacle information is marked in the three-dimensional map. In this way, accurate identification of transparent obstacles can be achieved, and a map including accurate transparent obstacle information can be constructed, thereby ensuring the subsequent normal operation of the robot.

[0054] Of course, it is not necessary to achieve all of the advantages described above at the same time to implement any product or method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0056] Figure 1 A flowchart of a map construction method provided by an embodiment of the present invention;

[0057] Figure 2 Based on Figure 1 A flow chart of determining whether environmental information meets environmental characteristics in the illustrated embodiment;

[0058] Figure 3 for Figure 2 A specific flow chart of step S202 in the illustrated embodiment;

[0059] Figure 4 for Figure 1 A specific flow chart of step S103 in the illustrated embodiment;

[0060] Figure 5 for Figure 1 A specific flow chart of step S102 in the illustrated embodiment;

[0061] Figure 6 Based on Figure 1 A schematic diagram of the personnel history trajectory of the illustrated embodiment;

[0062] Figure 7 for Figure 1 Another specific flow chart of step S103 in the illustrated embodiment;

[0063] Figure 8 Based on Figure 1 A flow chart of detecting transparent obstacles in the environment according to the embodiment shown;

[0064] Fig. 9 Based on Figure 1 A schematic diagram of detecting transparent obstacles in the environment according to the embodiment shown;

[0065] Fig.10 A schematic diagram of the structure of a map construction device provided by an embodiment of the present invention;

[0066] Fig.11 The present invention is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0067] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field based on the present invention belong to the scope of protection of the present invention.

[0068] In order to accurately identify transparent obstacles and construct a map including accurate transparent obstacle information, thereby ensuring the subsequent normal operation of the robot, an embodiment of the present invention provides a map construction method, device, electronic device, computer-readable storage medium, and computer program product. First, a map construction method provided by an embodiment of the present invention is introduced.

[0069] A map construction method provided in an embodiment of the present invention can be applied to a controller, processor, etc. carried by the robot itself, and can also be applied to a control device, processing device, management device, etc. that is communicatively connected to the robot. For the sake of clarity, they will be referred to as electronic devices in this article.

[0070] like Figure 1 As shown, it is a flowchart of a map construction method, including S101 to S105.

[0071] S101, determining a candidate area according to the positions of non-transparent obstacles included in the three-dimensional map.

[0072] The candidate area includes the area where the candidate transparent obstacle is located.

[0073] S102, obtaining the environment information corresponding to the candidate area, and determining whether the environment information meets the environment characteristics corresponding to the transparent obstacle. In response to the environment information meeting at least one of the environment characteristics corresponding to the transparent obstacle, executing steps S103 and S104; in response to the environment information not meeting any of the environment characteristics corresponding to the transparent obstacle, executing step S105.

[0074] S103, for each environmental transparent obstacle corresponding to the environmental characteristics, determine the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle.

[0075] The spatial attributes include at least size and position.

[0076] S104: Mark the environmental transparent obstacles on the three-dimensional map according to the spatial attributes of the environmental transparent obstacles to obtain a three-dimensional map including the transparent obstacles.

[0077] S105 , for each candidate transparent obstacle whose environmental information does not conform to any environmental characteristic, determine that the candidate transparent obstacle is not an environmental transparent obstacle.

[0078] It can be seen that in the solution provided by the embodiment of the present invention, since the three-dimensional map includes the position of non-transparent obstacles, and the transparent obstacles in the scene do not overlap with the non-transparent obstacles, the alternative area where the alternative transparent obstacles may exist can be determined based on the position of the non-transparent obstacles included in the three-dimensional map. Since the environmental information around the transparent obstacle has certain environmental characteristics, it is possible to determine whether the environmental information corresponding to the alternative area meets the environmental characteristics, so as to accurately identify the environmental transparent obstacles among the alternative transparent obstacles. Furthermore, the size and position of the environmental transparent obstacle are determined, and the environmental transparent obstacle information is marked in the three-dimensional map. In this way, accurate identification of transparent obstacles can be achieved, and a map including accurate transparent obstacle information can be constructed, thereby ensuring the subsequent normal operation of the robot.

[0079] Since the surface of non-transparent obstacles reflects light, the two-dimensional image captured for the current scene includes non-transparent obstacles. Accordingly, the three-dimensional map constructed using the traditional three-dimensional map construction method includes the location of the non-transparent obstacles. Since the non-transparent obstacles and transparent obstacles in the scene do not overlap, the transparent obstacles can only be located outside the location of the non-transparent obstacles. The electronic device can perform the above step S101, that is, based on the location of the non-transparent obstacles included in the three-dimensional map, preliminarily determine the candidate area, and the candidate area can include the area where the candidate transparent obstacles are located.

[0080] Among them, the above-mentioned three-dimensional map is a map constructed by a traditional three-dimensional map construction method, for example, by binocular imaging or structured light. In one embodiment, the above-mentioned three-dimensional map can be a map corresponding to the current scene where the robot is located, for example, the current scene can be a home, a factory, a pharmacy, a supermarket, etc. In different scenes, the types of transparent obstacles may also be different. For example, in a supermarket scene, the transparent obstacle can be a glass door at the entrance and exit; in a home scene, the transparent obstacle can be a floor-to-ceiling glass window, a bathroom door, and so on.

[0081] After determining the possible candidate transparent obstacles, it is necessary to determine the actual existing environmental transparent obstacles from the candidate transparent obstacles. Since the environmental information around the transparent obstacles usually has certain environmental characteristics, the electronic device can perform the above step S102, that is, obtain the environmental information corresponding to the candidate area, and determine whether the environmental information meets the environmental characteristics corresponding to the transparent obstacles, so as to screen the candidate transparent obstacles.

[0082] The environment information corresponding to the candidate area may at least include the environment information around the candidate area. Specifically, the environment information corresponding to the candidate area may be the environment information around the candidate area, or may be richer and more extensive information including the environment information around the candidate area, which is not specifically limited here.

[0083] In response to the environmental information meeting at least one of the environmental characteristics corresponding to the transparent obstacle, the candidate transparent obstacle corresponding to the environmental characteristic is determined to be an environmental transparent obstacle, and then the above steps S103 and S104 are executed. In response to the environmental information not meeting any of the environmental characteristics corresponding to the transparent obstacle, the candidate transparent obstacle whose environmental information does not meet any of the environmental characteristics is determined not to be an environmental transparent obstacle, and the above step S105 is executed.

[0084] The environmental characteristics corresponding to the above-mentioned transparent obstacles include various situations. For example, assuming that the environmental information corresponding to the candidate area is the size and position of the door handle, the environmental characteristics may be that the door handle is in the air in the candidate area. In response to the door handle meeting the environmental characteristics, it indicates that there is a floor-to-ceiling glass door in the candidate area. For another example, in response to the environmental information corresponding to the candidate area being the historical trajectory of a person, the environmental characteristics may be that there is a detour trajectory traveling in the opposite direction in the historical trajectory of the person. In response to the historical trajectory of the person meeting the environmental characteristics, it indicates that there is a glass wall in the candidate area.

[0085] In the case where the environmental information meets the environmental characteristics corresponding to the transparent obstacle, it is indicated that the candidate transparent obstacle corresponding to the environmental characteristics is an environmental transparent obstacle actually existing in the current scene. In order to accurately mark the information of the environmental transparent obstacle in the three-dimensional map, the spatial attributes of the environmental transparent obstacle can be determined. Among them, the spatial attributes may include at least size and position, and the position may be expressed as the coordinates of the corner points of the environmental transparent obstacle. The spatial attributes may also include surface shape, degree of inclination, etc., which are not specifically limited here.

[0086] Since transparent obstacles hardly reflect light, it is impossible to directly determine their spatial properties based on the image features of the transparent obstacles. Since the environmental information is collected around the environmental transparent obstacle, the environmental information can reflect the spatial features of the environmental transparent obstacle, and then the electronic device can perform the above step S103, that is, determine the spatial properties of the environmental transparent obstacle based on the environmental information corresponding to the environmental transparent obstacle, and thus infer the spatial properties of the environmental transparent obstacle based on the environmental information around the environmental transparent obstacle.

[0087] For example, assuming that the environmental information around the environmental transparent obstacle A includes the size and position of the door frame, the size and position of the door frame and the area surrounded by it can be determined as the size and position of the environmental transparent obstacle. For another example, assuming that the environmental information around the environmental transparent obstacle B includes the information of the non-transparent obstacle, the edge of the non-transparent obstacle around the environmental transparent obstacle can be determined, and then the size and position of the area composed of the above edges can be determined as the size and position of the environmental transparent obstacle.

[0088] After determining the spatial attributes of the environmental transparent obstacles, the electronic device can execute the above step S104, that is, marking the environmental transparent obstacles in the three-dimensional map according to the spatial attributes of the environmental transparent obstacles, and obtaining a three-dimensional map including the transparent obstacles, thereby correcting the three-dimensional map that lacks transparent obstacles or has incorrect transparent obstacle markings.

[0089] For example, assuming that candidate transparent obstacles 1-5 and corresponding candidate areas 1-5 are determined from the three-dimensional map, the environmental information around each candidate area, i.e., environmental information 1-5, can be obtained respectively, and it can be determined whether environmental information 1-5 conforms to the environmental characteristics corresponding to the transparent obstacle. In response to environmental information 1-3 conforming to the environmental characteristics corresponding to the transparent obstacle, it can be determined that candidate transparent obstacles 1-3 are environmental transparent obstacles, and according to environmental information 1-3, the spatial attributes 1-3 of candidate transparent obstacles 1-3 are determined respectively, and then according to spatial attributes 1-3, candidate transparent obstacles 1-3 are marked in the three-dimensional map respectively. In response to environmental information 4 and environmental information 5 not conforming to the environmental characteristics corresponding to the transparent obstacle, it can be determined that candidate transparent obstacles 4 and 5 are not environmental transparent obstacles.

[0090] The above identification result of whether the candidate transparent obstacle is an environmental transparent obstacle may be wrong. For example, the candidate transparent obstacle itself is a transparent obstacle in the physical space, but it is identified as not being an environmental transparent obstacle. For another example, the candidate transparent obstacle itself is not a transparent obstacle in the physical space, but it is identified as an environmental transparent obstacle. Since transparent obstacles can be detected by ranging sensors, when the robot includes ranging sensors, the identification result of environmental transparent obstacles can be reviewed through the information detected by the ranging sensors. Specifically:

[0091] For an alternative transparent obstacle determined to be an environmental transparent obstacle, a first distance between the alternative transparent obstacle and the robot can be determined through information detected by the ranging sensor, and then a second distance between the alternative transparent obstacle and the robot can be determined based on the position of the robot and the position of the alternative transparent obstacle.

[0092] If the difference between the first distance and the second distance is not greater than the preset threshold, it means that the candidate transparent obstacle actually exists in the physical space, and it can be determined that the candidate transparent obstacle is indeed an environmental transparent obstacle. If the difference between the first distance and the second distance is greater than the preset threshold, it means that the candidate transparent obstacle does not exist in the physical space, and the candidate transparent obstacle is mistakenly identified as an environmental transparent obstacle, and the mark corresponding to the candidate transparent obstacle can be removed from the three-dimensional map.

[0093] For alternative transparent obstacles that have not been determined as environmental transparent obstacles, the third distance between the robot and the alternative area where the alternative transparent obstacle is located can be determined through information detected by the ranging sensor, and then the fourth distance between the alternative area and the robot can be determined based on the position of the robot and the position of the alternative area. The fourth distance is the minimum value of the distance between each boundary point of the alternative area and the robot.

[0094] When the difference between the third distance and the fourth distance is not greater than the preset threshold, it means that the candidate transparent obstacle actually exists in the physical space, and the candidate transparent obstacle can be determined to be an environmental transparent obstacle, and then the above steps S103 and S104 can be performed for the candidate transparent obstacle, so as to mark the candidate transparent obstacle in the three-dimensional map. When the difference between the third distance and the fourth distance is greater than the preset threshold, it means that the candidate transparent obstacle does not really exist in the physical space, and the candidate transparent obstacle is really not an environmental transparent obstacle.

[0095] Continuing with the previous example, the electronic device can review whether candidate transparent obstacles 1 to 3 are mistakenly identified as environmental transparent obstacles, and whether candidate transparent obstacles 4 and 5 are mistakenly identified as not being environmental transparent obstacles. Assuming that after review, it is determined that candidate transparent obstacle 1 is not actually an environmental transparent obstacle, and candidate transparent obstacle 5 is actually an environmental transparent obstacle, then the mark corresponding to candidate transparent obstacle 1 can be removed from the three-dimensional map, and the mark corresponding to candidate transparent obstacle 5 can be added.

[0096] It can be seen that in this solution, since the three-dimensional map includes the position of non-transparent obstacles, and the transparent obstacles in the scene do not overlap with the non-transparent obstacles, the candidate area where the candidate transparent obstacles may exist can be determined according to the position of the non-transparent obstacles included in the three-dimensional map. Since the environmental information around the transparent obstacle has certain environmental characteristics, it can be determined whether the environmental information corresponding to the candidate area meets the environmental characteristics, so as to accurately identify the environmental transparent obstacles among the candidate transparent obstacles. Then, the size and position of the environmental transparent obstacle are determined, and the environmental transparent obstacle information is marked in the three-dimensional map. Through the above method, it is possible to avoid identifying only the border of the transparent obstacle as an obstacle, thereby avoiding the robot from colliding with the actual transparent obstacle when leaning forward, leaning back or extending its arm, and accurately identifying the transparent obstacle, forming a three-dimensional map with a higher degree of restoration of the current scene, thereby ensuring that the robot can normally complete the tasks of tallying, sorting, picking up on demand, indoor cleaning, and clothing sorting in the future.

[0097] As an implementation of an embodiment of the present invention, the above-mentioned environmental information may at least include an environmental image collected within a preset range of the candidate area, and the above-mentioned environmental image may include a visible light image and a depth image. Figure 2 As shown, after the step of obtaining the environmental information corresponding to the candidate area, the method may further include S201 to S204:

[0098] S201, extracting image features of associated objects from the visible light image and the depth image, and determining spatial attributes of the associated objects based on the image features.

[0099] There are usually associated items around transparent obstacles, and the associated items are compatible with the functions of the transparent obstacles. In the case where the transparent obstacle is a floor-to-ceiling glass door, the associated items may be items that support the floor-to-ceiling glass door, such as door frames, door thresholds, etc. The associated items may also be items that serve as decorations and identification, such as film patterns, hanging decorations, floor mats, etc. The associated items may also be items used to open the door, such as door handles, smart door locks, etc. The associated items may also be items used to achieve opening and closing operations, such as hardware accessories for door hinges. Assuming that the floor-to-ceiling glass door is an induction door, the associated item may be the induction device above the door. Assuming that it is a floor-to-ceiling glass door that requires facial recognition to start, the associated item may be a facial image acquisition device around the floor-to-ceiling glass door.

[0100] Furthermore, at least part of the associated items may be made of non-transparent materials. In one embodiment, the associated items may be all made of non-transparent materials. For example, the associated items may be metal door frames, wooden door sills, hardware accessories for door hinges, etc. In another embodiment, the associated items may be partially made of non-transparent materials. For example, the associated items may be a protective film attached to tempered glass, most of the protective film is transparent, and a small part of the area with a colored pattern is non-transparent, and the colored pattern usually serves as a decoration or identification.

[0101] When the transparent obstacle is a floor-to-ceiling window, the associated items may be curtains, window frames, platforms, etc. When the transparent obstacle is a floor-to-ceiling glass fish tank, the associated items may be the fish tank's filter, lighting equipment, cover, etc. When the transparent obstacle is a floor-to-ceiling glass cabinet, the associated items may be exhibits, cabinet frames, locks, etc.

[0102] Since the associated items are usually connected to or close to the transparent obstacles, the associated items and the transparent obstacles satisfy a certain connection rule. By determining whether the connection relationship between the associated items and the alternative transparent obstacles conforms to the above connection rule, it can be determined whether the environmental information around the alternative transparent obstacles conforms to the environmental characteristics corresponding to the transparent obstacles.

[0103] In order to determine the connection relationship between the associated item and the candidate transparent obstacle, the electronic device can control the image acquisition device of the robot to acquire an environmental image within a preset range of the candidate area, and the environmental image may include a visible light image and a depth image. Since the associated item is made of a non-transparent material, the associated item can be acquired by the image acquisition device, and the environmental image acquired by the image acquisition device includes image information of the associated item.

[0104] Furthermore, the electronic device can extract image features of the associated items from the visible light image and the depth image, and determine the spatial attributes of the associated items based on the image features. Specifically, the size of the associated items can be determined by the image features of the associated items extracted from the depth image, and the position of the associated items can be determined by combining the image features of the associated items extracted from the visible light image and the depth image.

[0105] S202, based on the spatial relationship between the spatial attributes of the associated items and the candidate regions, determine whether the connection relationship between the associated items and the candidate transparent obstacles conforms to the connection rule corresponding to transparent obstacles and associated items. In response to the connection relationship between the associated items and the candidate transparent obstacles conforming to the connection rule corresponding to transparent obstacles and associated items, step S203 is executed; in response to the connection relationship between the associated items and the candidate transparent obstacles not conforming to the connection rule corresponding to transparent obstacles and associated items, step S204 is executed.

[0106] Since the spatial attributes of the associated items can reflect the size and position of the associated items, and the spatial position of the candidate area can reflect the size and position of the candidate transparent obstacle, the spatial relationship between the spatial attributes of the associated items and the candidate area can reflect the connection relationship between the associated items and the candidate transparent obstacle. Furthermore, the electronic device can determine whether the connection relationship between the associated items and the candidate transparent obstacle conforms to the connection rule corresponding to the transparent obstacle and the associated items based on the spatial relationship between the spatial attributes of the associated items and the candidate area.

[0107] For example, if the associated item is a door handle, then the size and position of the door handle and the spatial relationship between the candidate areas can be used to determine whether the door handle is located in the middle of the candidate transparent obstacle in the vertical direction. If the associated item is a threshold, then the size and position of the threshold and the spatial relationship between the candidate areas can be used to determine whether the threshold is located at the bottom of the candidate transparent obstacle. If the associated item is a hanging decoration, then the size and position of the hanging decoration and the spatial relationship between the candidate areas can be used to determine whether the hanging decoration is located on the outer surface of the candidate transparent obstacle.

[0108] S203: Determine whether the environmental information of the candidate transparent obstacle meets the environmental characteristics corresponding to the transparent obstacle.

[0109] In response to the connection relationship between the associated object and the candidate transparent obstacle conforming to the connection rule corresponding to the transparent obstacle and the associated object, the electronic device may determine that the environmental information of the candidate transparent obstacle conforms to the environmental characteristics corresponding to the transparent obstacle.

[0110] For example, assuming that the associated item is a hanging decoration, through the size and position of the hanging decoration and the spatial relationship between the alternative areas, it is determined that the hanging decoration is located on the outer surface of the alternative area. Then it can be determined that the environmental information of the alternative transparent obstacle meets the environmental characteristics corresponding to the transparent obstacle, and there may be a transparent door in the alternative area.

[0111] S204: Determine that the environmental information of the candidate transparent obstacle does not conform to the environmental characteristics corresponding to the transparent obstacle.

[0112] In response to the connection relationship between the associated object and the candidate transparent obstacle not conforming to the connection rule corresponding to the transparent obstacle and the associated object, the electronic device may determine that the environmental information of the candidate transparent obstacle does not conform to the environmental characteristics corresponding to the transparent obstacle.

[0113] For example, assuming that the associated object is a hanging decoration, through the size and position of the hanging decoration and the spatial relationship between the alternative areas, it is determined that the hanging decoration is located inside the alternative area. Then it can be determined that the environmental information of the alternative transparent obstacle does not conform to the environmental characteristics corresponding to the transparent obstacle, and the alternative area may be a hollow doorway.

[0114] It can be seen that in this embodiment, the electronic device can determine the spatial attributes of the associated object through the environmental image, and then determine whether the environmental information of the candidate transparent obstacle meets the environmental characteristics corresponding to the transparent obstacle based on the connection relationship between the spatial attributes of the associated object and the candidate area. Since the associated object usually exists around the transparent obstacle and satisfies a certain connection rule with the transparent obstacle, the associated object can be targetedly identified through the above method, thereby accurately determining whether the environmental information of the candidate transparent obstacle meets the environmental characteristics corresponding to the transparent obstacle, and then accurately determining whether the candidate transparent obstacle is an environmental transparent obstacle.

[0115] As an implementation method of the embodiment of the present invention, Figure 3 As shown, the step of determining whether the connection relationship between the associated object and the candidate transparent obstacle conforms to the connection rule corresponding to the transparent obstacle and the associated object based on the spatial relationship between the spatial attribute of the associated object and the candidate area may include S301 and S302:

[0116] S301, determining the category of the associated items according to the image features of the associated items extracted from the visible light image, and determining the connection rule corresponding to the category of the associated items based on the correspondence between the preset category and the connection rule.

[0117] Since related items of different categories may be located at different positions of the transparent obstacle, the corresponding relationship between the category and the connection rule can be preset. For example, the corresponding relationship can be shown in the following Table 1:

[0118] Table 1

[0119] Category of related items Connection rules Thresholds, floor mats, platforms Located at the bottom of the optional transparent obstacle Lighting equipment, covers Located on top of the optional transparent barrier Sensing device, face image acquisition device, curtain Outside the optional transparent barrier Displays, filters Located inside the optional transparent barrier Hardware accessories Located on the side edge of the alternative transparent obstacle Door handles, smart door locks, locks Located in the middle of the alternative transparent obstacle in the vertical direction Film patterns, hanging decorations Located on the outer surface of the alternative transparent obstacle Door frames, window frames, cabinet frames Surround and fit the optional transparent barrier

[0120] Since the connection rules corresponding to associated items of different categories may be different, when it is necessary to determine whether the connection relationship between the associated items and the alternative transparent obstacles conforms to the connection rules, the electronic device can first determine the category of the associated items based on the image features of the associated items extracted from the visible light image, and then determine the connection rule corresponding to the category of the associated items based on the above correspondence.

[0121] S302: Based on the spatial properties of the associated items and the spatial relationship between the candidate areas, determine whether the connection relationship between the associated items and the candidate transparent obstacles conforms to the determined connection rule.

[0122] Since the determined connection rule corresponds to the associated object, the electronic device can determine whether the connection relationship between the associated object and the candidate transparent obstacle conforms to the determined connection rule based on the spatial relationship between the spatial attributes of the associated object and the candidate area.

[0123] For example, assuming that the category of the associated item is door frame, then it can be determined that the connection rule corresponding to the door frame is surrounding and fitting the alternative transparent obstacle, and then based on the spatial relationship between the spatial attributes of the door frame and the alternative area, it can be determined whether the door frame surrounds and fits the alternative transparent obstacle.

[0124] It can be seen that in this embodiment, corresponding connection rules are set in advance for the categories of various related items. Then, when related items of different categories are identified, the connection rules suitable for the categories of the related items can be used to accurately determine whether the connection relationship between the related items and the alternative transparent obstacles conforms to the corresponding connection rules.

[0125] As an implementation method of the embodiment of the present invention, Figure 4 As shown, the step of determining the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle may include S401 and S402:

[0126] S401, determining the category of the associated item based on the image features of the associated item extracted from the visible light image, and determining the method for determining the spatial attribute corresponding to the category of the associated item based on the correspondence between the preset category and the method for determining the spatial attribute.

[0127] Since the door frame surrounds and fits the environmental transparent obstacle, when the associated object is the door frame, the spatial properties of the environmental transparent obstacle can be directly determined based on the spatial properties of the door frame. However, other types of associated objects are not in a surrounding relationship with the environmental transparent obstacle, so the spatial properties of the environmental transparent obstacle cannot be directly determined based on the spatial properties of the associated objects. Since the environmental transparent obstacle is usually in contact with the surrounding non-transparent obstacles and is in the same plane, the spatial properties of the environmental transparent obstacle can be determined through the non-transparent obstacles.

[0128] It can be seen that the determination methods of spatial attributes corresponding to different categories of associated items are different, so the correspondence between the categories and the determination methods of spatial attributes can be preset. When it is necessary to determine the spatial attributes of environmental transparent obstacles, the electronic device can first determine the category of the associated items based on the image features of the associated items extracted from the visible light image, and then determine the determination method of the spatial attributes corresponding to the category of the associated items based on the above correspondence.

[0129] S402: Determine the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle and the method for determining the spatial attribute.

[0130] Since the method for determining the spatial attribute is corresponding to the associated object, the electronic device can determine the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle and the method for determining the spatial attribute.

[0131] For example, assuming that the associated object is a door frame, the electronic device can determine the size and position of the door frame and the area it surrounds as the size and position of the environmental transparent obstacle. For another example, assuming that the associated object is a door handle, the electronic device can determine the size and position of the area formed by the edges of the non-transparent obstacles around the environmental transparent obstacle as the size and position of the environmental transparent obstacle.

[0132] It can be seen that in this embodiment, the corresponding relationship between the category of the associated object and the determination method of the spatial attribute is preset. When the category of the associated object is different, the determination method suitable for the category can be used to determine the spatial attribute of the environmental transparent obstacle. In this way, the spatial attribute of the environmental transparent obstacle can be determined quickly and accurately.

[0133] As an implementation manner of an embodiment of the present invention, the step of determining the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle and the method for determining the spatial attribute may include at least one of the following:

[0134] In response to the category of the associated object being a door or window frame, the size and position of the door or window frame and the area enclosed by it are respectively determined as the size and position of the environmental transparent obstacle.

[0135] In response to the category of the associated object not being a door or window frame, the edges of the non-transparent obstacles around the environmental transparent obstacle are determined based on the connection pattern between the environmental transparent obstacle and the associated object; the size and position of the area formed by the edges of the non-transparent obstacles are respectively determined as the size and position of the environmental transparent obstacle.

[0136] Door and window frames are door frames or window frames. When the category of the associated items is door and window frames, since the door and window frames are in close contact with the outer contour of the environmental transparent obstacles, and the door and window frames are components of floor-to-ceiling glass doors or floor-to-ceiling glass windows, when the category of the associated items is door and window frames, the electronic device can determine the size and position of the door and window frames and the area enclosed by them as the size and position of the environmental transparent obstacles, respectively.

[0137] When the category of the associated object is not a door or window frame, the outer contour of the associated object and the environmental transparent obstacle are not fitted, while the non-transparent obstacle around the environmental transparent obstacle is usually fitted with the environmental transparent obstacle, and the connection pattern between the environmental transparent obstacle and the associated object can reflect the spatial properties of the environmental transparent obstacle. Therefore, the electronic device can determine the edge of the non-transparent obstacle around the environmental transparent obstacle based on the connection pattern between the environmental transparent obstacle and the associated object, and determine the size and position of the area formed by the edge of the non-transparent obstacle as the size and position of the environmental transparent obstacle, respectively.

[0138] For example, assuming that the associated object is a threshold, the electronic device can determine the wall edge around the environmental transparent obstacle based on the connection relationship between the threshold and the environmental transparent obstacle, and then determine the size and position of the area formed by the wall edge as the size and position of the environmental transparent obstacle.

[0139] It can be seen that in this embodiment, for the two cases where the category of the associated items is a door and window frame and the category of the associated items is not a door and window frame, different methods can be used to determine the spatial attributes of the environmental transparent obstacle. In the case where the category of the associated items is a door and window frame, the spatial attributes of the environmental transparent obstacle can be determined directly based on the spatial attributes of the associated items. In the case where the category of the associated items is not a door and window frame, the spatial attributes of the environmental transparent obstacle can be indirectly determined based on the edges of the non-transparent obstacles around the environmental transparent obstacle. In this way, when the associated items are of different categories, the spatial attributes of the environmental transparent obstacle can be quickly and accurately determined using corresponding methods.

[0140] As an implementation method of the embodiment of the present invention, Figure 5 As shown, the above step of obtaining the environment information corresponding to the candidate area may include S501 and S502:

[0141] S501, taking a preset position in the candidate area as a center point, and determining an area to be collected according to preset parameters.

[0142] After determining the candidate area of ​​the candidate transparent obstacle, in order to further determine whether the candidate transparent obstacle is an environmental transparent obstacle, the electronic device can use a preset position in the candidate area as the center point and determine the area to be collected according to preset parameters. The preset position can be the center point of the candidate area or a non-center point of the candidate area.

[0143] In order to make the environmental information of the area to be collected completely include the environmental information of the candidate area, the area to be collected may include the candidate area, that is, the area to be collected covers the candidate area.

[0144] The above preset parameters can be set according to actual needs. In one embodiment, the preset parameter can be a circular radius. In this case, the preset position can be used as the center of the circle and the preset parameter can be used as the radius to define a circular area to be collected. In another embodiment, the preset parameter can be used as the side length of a rectangle. In this case, the preset position can be used as the center of the rectangle and the preset parameter can be used as the side length to define a square area to be collected.

[0145] Of course, it is also possible to use the preset position as the center point and expand outward according to the preset parameters to delineate other regular or irregular shaped areas to be collected, which is all reasonable.

[0146] S502: Acquire a visible light image and a depth image of the area to be collected.

[0147] After determining the area to be collected, the electronic device can obtain the visible light image and the depth image of the area to be collected, wherein the visible light image and the depth image can be collected by the image collection device of the robot.

[0148] It can be seen that in this embodiment, the electronic device can demarcate the area to be collected that covers the candidate area according to the pre-set parameters, and then obtain the environmental image of the area to be collected. In this way, the area to be collected can be determined efficiently, and the environmental image can contain complete environmental information of the candidate area, which helps to improve the accuracy of determining whether the candidate transparent obstacle is an environmental transparent obstacle.

[0149] In scenes such as factories or workshops, the transparent obstacle may be a whole glass wall used for regional isolation. Since the whole glass wall may not have any identifiable associated items, it is impossible to determine whether the candidate transparent obstacle is an environmental transparent obstacle through the environmental image. However, in scenes with a whole glass wall, the movement trajectory of the personnel will be greatly affected by the whole glass wall. Therefore, the historical trajectory of the personnel can be used to determine whether the environmental information of the candidate area meets the environmental characteristics corresponding to the transparent obstacle, thereby determining whether the candidate transparent obstacle is an environmental transparent obstacle.

[0150] As an implementation of an embodiment of the present invention, the above-mentioned environmental information may at least include historical trajectories of personnel collected within a preset range of the candidate area. In this case, after the above-mentioned step of obtaining the environmental information corresponding to the candidate area, the above-mentioned method may further include:

[0151] Determine whether there is a detour trajectory in the historical trajectory of the personnel; in response to the existence of a detour trajectory in the historical trajectory of the personnel, determine that the environmental information meets the environmental characteristics corresponding to the transparent obstacle; or, determine whether there is a passing trajectory in the historical trajectory of the personnel; in response to the existence of a passing trajectory in the historical trajectory of the personnel, obtain the personnel action of the passing personnel corresponding to the passing trajectory; in response to the personnel action including a door opening action or a door closing action, determine that the environmental information meets the environmental characteristics corresponding to the transparent obstacle.

[0152] In a first implementation, the electronic device may determine whether there is a detour track in the historical trajectory of the person, wherein the detour track may be a track that moves from one side of the candidate area to the opposite side in a non-linear manner, and is usually a track that moves in the opposite direction. For example, the detour track may be in an arc shape, a square shape, or other irregular shapes. A schematic diagram of the historical trajectory of the person may be as follows: Figure 6 As shown, the first position 601 and the second position 602 are located on both sides of the alternative area 603, and the first trajectory 604 from the first position 601 to the second position 602 is a detour trajectory. The first half of the first trajectory 604 is a trajectory moving to the right, and the second half is a trajectory moving to the left, that is, a reverse trajectory occurs.

[0153] In response to the presence of a detour trajectory in the personnel's historical trajectory, it means that the personnel's movements in the alternative area are restricted and they cannot pass through the alternative area, but can only go around from one side of the alternative area to the other side. Therefore, it can be determined that the environmental information meets the environmental characteristics corresponding to the transparent obstacle.

[0154] In one embodiment, the degree of conformity of the environmental information to the environmental characteristics is positively correlated with the number of detour tracks. The number of detour tracks may be one or more, and the greater the number of detour tracks, the higher the degree of conformity of the environmental information to the environmental characteristics. When the degree of conformity is greater than a preset degree of conformity threshold, it can be determined that the environmental information conforms to the environmental characteristics corresponding to the transparent obstacle.

[0155] In response to the fact that there is no detour trajectory in the historical trajectory of the person, it means that the person can normally travel through the alternative area and will not be restricted by the alternative area. Therefore, it can be determined that the environmental information does not meet the environmental characteristics corresponding to the transparent obstacle.

[0156] In a second embodiment, the electronic device can determine whether there is a travel trajectory in the historical trajectory of the person. The travel trajectory is a trajectory that moves in a straight line from one side of the candidate area to the opposite side. It should be understood that since the movement trajectory of a person is usually not straight, the above-mentioned travel trajectory refers to a trajectory that is straight or close to a straight line, not an absolute straight line. Figure 6 As shown, the second trajectory 605 from the first position 601 to the second position 602 is a travel trajectory.

[0157] In response to the presence of a walking track in the historical trajectory of the person, it indicates that there is a hollow door opening or a floor-to-ceiling glass door at the location of the walking track. Since the person will not open or close the door when passing through a hollow door opening, but may open or close the door when passing through a floor-to-ceiling glass door, in order to further determine whether the location of the walking track is a hollow door opening or a floor-to-ceiling glass door, the electronic device can obtain the person's action of the walking person corresponding to the walking track. In response to the person's action including the door opening action or the door closing action, the electronic device can determine that the environmental information meets the environmental characteristics corresponding to the transparent obstacle.

[0158] In response to the absence of a travel trajectory in the historical trajectory of the person, or the person's actions do not include door opening or door closing, indicating that the candidate transparent obstacle is not an environmental transparent obstacle, the electronic device can determine that the environmental information does not meet the environmental characteristics corresponding to the transparent obstacle.

[0159] It can be seen that in this embodiment, the electronic device can make a forward inference by determining whether there is a detour track in the historical trajectory of the person, and can also make a reverse inference by determining whether there is a through track in the historical trajectory of the person. In this way, it can be determined whether the candidate transparent obstacle is an environmental transparent obstacle based on the historical trajectory of the person, and the accuracy can be improved.

[0160] As an implementation method of the embodiment of the present invention, Figure 7 As shown, the step of determining the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle may include S701 and S702:

[0161] S701: Determine the edge of the non-transparent obstacles around the environmental transparent obstacle.

[0162] Since there are usually non-transparent obstacles around the environmental transparent obstacle, and the environmental transparent obstacle needs to be connected to the non-transparent obstacle to be fixed, in order to determine the size and position of the environmental transparent obstacle, the electronic device can determine the edges of the non-transparent obstacles around the environmental transparent obstacle.

[0163] For example, assuming that the environmental transparent obstacle is a whole glass wall, the surrounding non-transparent obstacles may be the wall, and the edge of the non-transparent obstacle may be the boundary between the whole glass wall and the wall. For another example, assuming that the environmental transparent obstacle is a glass door, the surrounding non-transparent obstacles may be door frames, and the edge of the non-transparent obstacle may be the outer contour of the door frame.

[0164] S702: Determine the size and position of the area formed by the edge of the non-transparent obstacle as the size and position of the environmental transparent obstacle.

[0165] Since the edges of non-transparent obstacles and environmental transparent obstacles are usually close to each other, after determining the edges of non-transparent obstacles around the environmental transparent obstacle, the electronic device can determine the size and position of the area formed by the edges of the non-transparent obstacles as the size and position of the environmental transparent obstacle.

[0166] It can be seen that in this embodiment, the electronic device can determine the size and position of the environmental transparent obstacle through the edge of the non-transparent obstacle. Since the edges of the non-transparent obstacle and the environmental transparent obstacle are usually in close contact, the size and position of the environmental transparent obstacle can be quickly and accurately determined through the above method, so that the information of the environmental transparent obstacle subsequently marked in the three-dimensional map is accurate.

[0167] As an implementation of an embodiment of the present invention, the three-dimensional map may be a map corresponding to the current scene where the robot is located, and the robot may include a distance measuring sensor. Figure 8 As shown, before the step of marking the environmental transparent obstacles on the three-dimensional map according to the spatial attributes of the environmental transparent obstacles to obtain the three-dimensional map including the transparent obstacles, the method may further include S801 to S804:

[0168] S801, obtaining a first distance between the robot and the transparent obstacle in the environment through the distance measuring sensor.

[0169] Since the distance measuring sensor can detect transparent obstacles, in order to verify whether there are actually transparent obstacles in the environment, the distance measuring sensor can be deployed on the robot. In this way, the electronic device can obtain the first distance between the robot and the transparent obstacle in the environment through the distance measuring sensor.

[0170] The above-mentioned distance measuring sensors can be deployed in one or more of the following parts: head, chest, back, side, chassis. If multiple distance measuring sensors are deployed on the robot, then when measuring the distance, it is necessary to control each distance measuring sensor to send signals in sequence at a certain time interval to avoid the distance measuring sensor receiving the signals sent and returned by other distance measuring sensors in advance after sending the signal, thereby misjudging the distance.

[0171] Among them, the above-mentioned ranging sensor can be a sensor that realizes the ranging function by sending and receiving signals, and can specifically be an ultrasonic transceiver, an infrared ranging sensor, a laser ranging sensor, a radar ranging sensor, etc., which is not specifically limited here.

[0172] For example, a schematic diagram of a robot detecting transparent obstacles in the environment through a range sensor can be as follows: Fig. 9 As shown, a distance measuring sensor 902 is installed on the chassis of the robot 901 , and the distance measuring sensor 902 can be used to determine the first distance between the robot 901 and the transparent obstacle 903 in the environment.

[0173] S802, according to the position of the robot and the position included in the spatial attribute of the environmental transparent obstacle, obtain a second distance between the robot and the environmental transparent obstacle. In response to the difference between the first distance and the second distance being not greater than a preset threshold, execute step S803; in response to the difference between the first distance and the second distance being greater than the preset threshold, execute step S804.

[0174] The electronic device can obtain the second distance between the robot and the environmental transparent obstacle according to the position of the robot and the position included in the spatial attributes of the environmental transparent obstacle. Since the first distance is the distance between the environmental transparent obstacle actually existing in the current scene and the robot, and the second distance is the distance between the environmental transparent obstacle and the robot obtained by inference, it can be determined whether the difference between the first distance and the second distance is not greater than a preset threshold, thereby determining whether the environmental transparent obstacle obtained by inference actually exists in the current scene.

[0175] The preset threshold value may be set according to actual needs of the user, for example, 3 cm, 5 cm, 6 cm, etc., which are all reasonable.

[0176] S803, executing the step of marking the environmental transparent obstacles in the three-dimensional map according to the spatial attributes of the environmental transparent obstacles to obtain a three-dimensional map including the transparent obstacles.

[0177] In response to the difference between the first distance and the second distance being not greater than the preset threshold, it indicates that the position of the inferred environmental transparent obstacle is relatively close to the position of the actual environmental transparent obstacle, and it can be determined that the inferred environmental transparent obstacle actually exists. Then, the electronic device can perform the above step S104, that is, mark the environmental transparent obstacle in the three-dimensional map.

[0178] S804: Determine whether the environmental transparent obstacle actually does not exist.

[0179] In response to the difference between the first distance and the second distance being greater than the preset threshold, it indicates that the position of the inferred environmental transparent obstacle is far away from the position of the actually existing environmental transparent obstacle, and it can be determined that the inferred environmental transparent obstacle does not actually exist. The environmental transparent obstacle may actually be a door sill, a threshold, a hollow simple door opening with a edging, etc., which is mistakenly identified as a transparent obstacle, and the environmental transparent obstacle does not need to be marked in the three-dimensional map.

[0180] It can be seen that in this embodiment, the electronic device can determine the first distance corresponding to the actual environmental transparent obstacle in the current scene through the robot's distance measurement sensor, and determine the second distance corresponding to the environmental transparent obstacle obtained by inference through the spatial attribute of the environmental transparent obstacle. Furthermore, based on the difference between the first distance and the second distance, it is verified whether the environmental transparent obstacle obtained by inference actually exists, which can avoid misidentifying the decorative frame as a transparent obstacle.

[0181] As an implementation manner of an embodiment of the present invention, the step of determining the candidate area according to the position of the non-transparent obstacle included in the three-dimensional map may include at least one of the following:

[0182] The robot is controlled to move in the current scene and collect an environmental image, and a three-dimensional map of the current scene is constructed based on the position of the robot when collecting the environmental image and the scene information included in the environmental image; and according to the position of the non-transparent obstacles included in the three-dimensional map, an area in the three-dimensional map where at least part of the non-transparent obstacles do not exist is determined as a candidate area.

[0183] A pre-constructed three-dimensional map of the current scene is obtained; and according to the positions of the non-transparent obstacles included in the three-dimensional map, an area in the three-dimensional map where at least part of the non-transparent obstacles do not exist is determined as a candidate area.

[0184] During the movement of the robot in the current scene, an environmental image is collected and the position of obstacles is detected by a ranging sensor; a three-dimensional map of the current scene is constructed based on the position of the robot when the environmental image is collected and the scene information included in the environmental image; the position of each obstacle is determined based on the distance of the obstacle detected by the ranging sensor; and an alternative area is determined based on the determined position of the obstacle and the position of the non-transparent obstacle included in the three-dimensional map.

[0185] In the first embodiment, the electronic device can control the robot to move in the current scene and collect environmental images through the robot's image acquisition device. According to the position of the robot when collecting the environmental image and the scene information included in the environmental image, a three-dimensional map corresponding to the current scene is constructed, that is, the robot is controlled to construct a three-dimensional map corresponding to the current scene through a traditional three-dimensional space reconstruction method.

[0186] Since transparent obstacles and non-transparent obstacles in the scene do not overlap, it can be determined that there are no transparent obstacles in the area where non-transparent obstacles exist. Accordingly, it can be determined that transparent obstacles may be located in the area where non-transparent obstacles do not exist.

[0187] On the one hand, in the area where there are no non-transparent obstacles, only some areas may have transparent obstacles, rather than all areas. On the other hand, in the process of building a three-dimensional map, there may be certain restrictions on the areas where transparent obstacles exist. For example, assuming that it is predetermined that transparent obstacles are locked only in a preset area, the overlapping part between the preset area and the area where there are no non-transparent obstacles can be determined as an alternative area.

[0188] Therefore, according to the positions of the non-transparent obstacles included in the three-dimensional map, at least a part of the area in the three-dimensional map where the non-transparent obstacles do not exist can be determined as the candidate area. That is, in the area in the three-dimensional map where the non-transparent obstacles do not exist, only a part of the area can be determined as the candidate area, or the whole area can be determined as the candidate area.

[0189] In the second embodiment, the electronic device can obtain a pre-constructed three-dimensional map of the current scene, and then determine the area in the three-dimensional map where at least part of the non-transparent obstacles do not exist as a candidate area based on the positions of the non-transparent obstacles included in the three-dimensional map, that is, the information of the transparent obstacles is supplemented on the basis of the original three-dimensional map.

[0190] In a third embodiment, when the robot moves in the current scene, the electronic device can control the robot's image acquisition device to acquire environmental images and detect the position of obstacles through the ranging sensor. The electronic device can construct a three-dimensional map of the current scene based on the robot's position when acquiring the environmental image and the scene information included in the environmental image, that is, control the robot to construct a three-dimensional map corresponding to the current scene through a traditional three-dimensional space reconstruction method.

[0191] Next, the electronic device can determine the position of each obstacle based on the distance of the obstacle detected by the ranging sensor. Since the distance detected by the ranging sensor includes the distances corresponding to all obstacles, that is, the distances corresponding to non-transparent obstacles and the distances corresponding to transparent obstacles, the position of each obstacle determined by the ranging sensor includes both the position of the non-transparent obstacle and the position of the transparent obstacle.

[0192] Since only the information of non-transparent obstacles can be collected through the traditional three-dimensional space reconstruction method, the three-dimensional map only includes the positions of non-transparent obstacles, and does not include the positions of transparent obstacles. The electronic device can determine the candidate area based on the determined positions of the obstacles and the positions of the non-transparent obstacles included in the three-dimensional map. Specifically, the positions of the non-transparent obstacles included in the three-dimensional map can be filtered out from the determined positions of the obstacles to obtain the positions of the transparent obstacles, and the candidate area can be determined based on the positions of the transparent obstacles.

[0193] It can be seen that in this embodiment, the electronic device can determine the candidate area in the three-dimensional map through the above three implementation methods. Since non-transparent obstacles and transparent obstacles do not overlap, the position of the non-transparent obstacle can be determined through the traditional three-dimensional space reconstruction method, and then the candidate area can be determined according to the position of the non-transparent obstacle. In this way, the candidate area can be determined quickly and accurately.

[0194] Corresponding to the above-mentioned map construction method, an embodiment of the present invention further provides a map construction device. The map construction device provided by the embodiment of the present invention is introduced below.

[0195] like Fig.10 FIG. 1 is a schematic diagram of a map construction device, wherein the device comprises:

[0196] The region determination module 1001 is used to determine a candidate region according to the positions of non-transparent obstacles included in the three-dimensional map.

[0197] The candidate area includes the area where the candidate transparent obstacle is located.

[0198] The information acquisition module 1002 is used to acquire the environment information corresponding to the candidate area.

[0199] The attribute determination module 1003 is used to determine the spatial attribute of each environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle in response to the environmental information corresponding to at least one of the environmental characteristics corresponding to the transparent obstacle.

[0200] The spatial attributes include at least size and position.

[0201] The map construction module 1004 is used to mark the environmental transparent obstacles in the three-dimensional map according to the spatial attributes of the environmental transparent obstacles, so as to obtain a three-dimensional map including the transparent obstacles.

[0202] It can be seen that in the solution provided by the embodiment of the present invention, since the three-dimensional map includes the position of non-transparent obstacles, and the transparent obstacles in the scene do not overlap with the non-transparent obstacles, the alternative area where the alternative transparent obstacles may exist can be determined based on the position of the non-transparent obstacles included in the three-dimensional map. Since the environmental information around the transparent obstacle has certain environmental characteristics, it is possible to determine whether the environmental information corresponding to the alternative area meets the environmental characteristics, so as to accurately identify the environmental transparent obstacles among the alternative transparent obstacles. Furthermore, the size and position of the environmental transparent obstacle are determined, and the environmental transparent obstacle information is marked in the three-dimensional map. In this way, accurate identification of transparent obstacles can be achieved, and a map including accurate transparent obstacle information can be constructed, thereby ensuring the subsequent normal operation of the robot.

[0203] As an implementation of an embodiment of the present invention, the above-mentioned environmental information may at least include an environmental image collected within a preset range of the candidate area, and the environmental image includes a visible light image and a depth image. In this case, the above-mentioned device may also include:

[0204] The associated item determination module is used to extract image features of the associated items from the visible light image and the depth image, and determine the spatial attributes of the associated items based on the image features.

[0205] The function of the associated object is compatible with the function of the transparent obstacle and at least a part of the associated object is made of non-transparent material.

[0206] The connection relationship determination module is used to determine whether the connection relationship between the associated object and the candidate transparent obstacle conforms to the connection rule corresponding to the transparent obstacle and the associated object based on the spatial attribute of the associated object and the spatial relationship between the candidate area.

[0207] The first determination module is used to determine that the environmental information of the candidate transparent obstacle meets the environmental characteristics corresponding to the transparent obstacle in response to the connection relationship between the associated object and the candidate transparent obstacle meeting the connection rule corresponding to the transparent obstacle and the associated object.

[0208] As an implementation manner of an embodiment of the present invention, the above-mentioned connection relationship determination module may include:

[0209] The category determination submodule is used to determine the category of the associated object based on the image features of the associated object extracted from the visible light image, and determine the connection rule corresponding to the category of the associated object based on the corresponding relationship between the preset category and the connection rule.

[0210] The connection relationship determination submodule is used to determine whether the connection relationship between the associated object and the candidate transparent obstacle conforms to the determined connection rule based on the spatial attribute of the associated object and the spatial relationship between the candidate area.

[0211] As an implementation of the embodiment of the present invention, the attribute determination module 1003 may include:

[0212] The method determination submodule is used to determine the category of the associated object based on the image features of the associated object extracted from the visible light image, and determine the method for determining the spatial attribute corresponding to the category of the associated object based on the correspondence between the preset category and the method for determining the spatial attribute.

[0213] The first attribute determination submodule is used to determine the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle and the method for determining the spatial attribute.

[0214] As an implementation manner of an embodiment of the present invention, the first attribute determination submodule may include:

[0215] The first attribute determination unit is used to determine the size and position of the door and window frame and the area surrounded by it as the size and position of the environmental transparent obstacle in response to the category of the associated object being the door and window frame.

[0216] The second attribute determination unit is used to determine the edge of the non-transparent obstacle around the environmental transparent obstacle in response to the category of the associated object not being a door or window frame, based on the connection rule between the environmental transparent obstacle and the associated object; and determine the size and position of the area composed of the edges of the non-transparent obstacle as the size and position of the environmental transparent obstacle, respectively.

[0217] As an implementation of an embodiment of the present invention, the information acquisition module 1002 may include:

[0218] The submodule for determining the area to be collected is used to determine the area to be collected according to preset parameters with a preset position in the candidate area as the center point.

[0219] Wherein, the area to be collected includes the candidate area.

[0220] The image acquisition submodule is used to acquire the visible light image and the depth image of the area to be collected.

[0221] As an implementation of an embodiment of the present invention, the above-mentioned environmental information may at least include historical trajectories of personnel collected within a preset range of the candidate area. In this case, the above-mentioned device may also include:

[0222] The first trajectory determination submodule is used to determine whether there is a detour trajectory in the historical trajectory of the person; in response to the presence of a detour trajectory in the historical trajectory of the person, determine that the environmental information meets the environmental characteristics corresponding to the transparent obstacle.

[0223] The detour trajectory is a trajectory that moves in a non-linear manner from one side of the candidate area to the opposite side.

[0224] The second trajectory determination submodule is used to determine whether there is a walking trajectory in the historical trajectory of the personnel; in response to the existence of a walking trajectory in the historical trajectory of the personnel, obtain the personnel action of the walking personnel corresponding to the walking trajectory; in response to the personnel action including a door opening action or a door closing action, determine that the environmental information meets the environmental characteristics corresponding to the transparent obstacle.

[0225] The travel trajectory is a trajectory that moves in a straight line from one side of the candidate area to the opposite side.

[0226] As an implementation of the embodiment of the present invention, the attribute determination module 1003 may include:

[0227] The edge determination submodule is used to determine the edges of the non-transparent obstacles around the transparent obstacles in the environment.

[0228] The second attribute determination submodule is used to determine the size and position of the area formed by the edge of the non-transparent obstacle as the size and position of the transparent obstacle in the environment.

[0229] As an implementation of an embodiment of the present invention, the three-dimensional map may be a map corresponding to the current scene where the robot is located, and the robot may include a distance measuring sensor. In this case, the device may also include:

[0230] The first distance determination module is used to obtain a first distance between the robot and the transparent obstacle in the environment through the distance measuring sensor.

[0231] The second distance determination module is used to obtain a second distance between the robot and the environmental transparent obstacle according to the position of the robot and the position included in the spatial attributes of the environmental transparent obstacle.

[0232] An execution module is used to execute the step of marking the environmental transparent obstacles in the three-dimensional map according to the spatial attributes of the environmental transparent obstacles in response to the difference between the first distance and the second distance being no greater than a preset threshold to obtain a three-dimensional map including the transparent obstacles.

[0233] As an implementation of an embodiment of the present invention, the region determination module 1001 may include:

[0234] The first candidate area determination submodule is used to control the robot to move in the current scene and collect environmental images, and to construct a three-dimensional map of the current scene based on the position of the robot when collecting the environmental images and the scene information included in the environmental images; and to determine, according to the positions of the non-transparent obstacles included in the three-dimensional map, an area in the three-dimensional map where at least part of the non-transparent obstacles do not exist as a candidate area.

[0235] The second candidate area determination submodule is used to obtain a pre-built three-dimensional map of the current scene; according to the positions of the non-transparent obstacles included in the three-dimensional map, determine at least part of the area in the three-dimensional map where no non-transparent obstacles exist as a candidate area.

[0236] The third candidate area determination submodule is used to collect environmental images and detect the positions of obstacles through ranging sensors during the movement of the robot in the current scene; construct a three-dimensional map of the current scene based on the position of the robot when collecting the environmental images and the scene information included in the environmental images; determine the position of each obstacle based on the distance of the obstacle detected by the ranging sensor; and determine the candidate area based on the determined positions of the obstacles and the positions of the non-transparent obstacles included in the three-dimensional map.

[0237] like Fig.11 FIG. 1 is a schematic diagram of a structure of an electronic device, comprising:

[0238] Memory 1101, used for storing computer programs;

[0239] The processor 1102 is used to implement the map construction method steps described in any of the above embodiments when executing the program stored in the memory 1101.

[0240] Furthermore, the electronic device may further include a communication bus and / or a communication interface, and the processor 1102, the communication interface, and the memory 1101 communicate with each other via the communication bus.

[0241] It can be seen that in the solution provided by the embodiment of the present invention, since the three-dimensional map includes the position of non-transparent obstacles, and the transparent obstacles in the scene do not overlap with the non-transparent obstacles, the alternative area where the alternative transparent obstacles may exist can be determined based on the position of the non-transparent obstacles included in the three-dimensional map. Since the environmental information around the transparent obstacle has certain environmental characteristics, it is possible to determine whether the environmental information corresponding to the alternative area meets the environmental characteristics, so as to accurately identify the environmental transparent obstacles among the alternative transparent obstacles. Furthermore, the size and position of the environmental transparent obstacle are determined, and the environmental transparent obstacle information is marked in the three-dimensional map. In this way, accurate identification of transparent obstacles can be achieved, and a map including accurate transparent obstacle information can be constructed, thereby ensuring the subsequent normal operation of the robot.

[0242] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0243] The communication interface is used for communication between the above electronic device and other devices.

[0244] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0245] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0246] In another embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the steps of the map construction method described in any of the above embodiments are implemented.

[0247] In another embodiment of the present invention, a computer program product including instructions is provided, which, when executed on a computer, enables the computer to execute the map construction method described in any one of the above embodiments.

[0248] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk Solid State Disk (SSD)), etc.

[0249] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0250] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, electronic device, computer-readable storage medium, and computer program product embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0251] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A map construction method, characterized in that: The method comprises: Determine a candidate area according to the position of the non-transparent obstacle included in the three-dimensional map, wherein the candidate area includes the area where the candidate transparent obstacle is located; Obtaining environmental information corresponding to the candidate area; In response to the environmental information being consistent with at least one of the environmental characteristics corresponding to the transparent obstacle, for each environmental transparent obstacle corresponding to the environmental characteristic, determining a spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle, wherein the spatial attribute includes at least size and position; According to the spatial attributes of the environmental transparent obstacles, the environmental transparent obstacles are marked on the three-dimensional map to obtain a three-dimensional map including the transparent obstacles.

2. The method according to claim 1, characterized in that The environmental information at least includes an environmental image collected within a preset range of the candidate area, and the environmental image includes a visible light image and a depth image; After the step of obtaining the environment information corresponding to the candidate area, the method further includes: Extracting image features of the associated item from the visible light image and the depth image, and determining the spatial attributes of the associated item based on the image features; determining whether the connection relationship between the associated item and the candidate transparent obstacle conforms to the connection rule corresponding to the transparent obstacle and the associated item based on the spatial relationship between the spatial attributes of the associated item and the candidate area; in response to the connection relationship between the associated item and the candidate transparent obstacle conforming to the connection rule corresponding to the transparent obstacle and the associated item, determining that the environmental information of the candidate transparent obstacle conforms to the environmental characteristics corresponding to the transparent obstacle, wherein the function of the associated item is compatible with the function of the transparent obstacle and at least a portion of the associated item is made of non-transparent material; and / or, The environmental information at least includes historical trajectories of personnel collected within a preset range of the candidate area; After the step of obtaining the environment information corresponding to the candidate area, the method further includes: Determine whether there is a detour trajectory in the personnel's historical trajectory; in response to the existence of a detour trajectory in the personnel's historical trajectory, determine that the environmental information meets the environmental characteristics corresponding to the transparent obstacle; or, determine whether there is a passing trajectory in the personnel's historical trajectory; in response to the existence of a passing trajectory in the personnel's historical trajectory, obtain the personnel action of the passing personnel corresponding to the passing trajectory; in response to the personnel action including a door opening action or a door closing action, determine that the environmental information meets the environmental characteristics corresponding to the transparent obstacle, wherein the detour trajectory is a trajectory moving from one side of the alternative area to the opposite side in a non-straight line, and the passing trajectory is a trajectory moving from one side of the alternative area to the opposite side in a straight line.

3. The method according to claim 2, characterized in that The step of determining whether the connection relationship between the associated object and the candidate transparent obstacle conforms to the connection rule corresponding to the transparent obstacle and the associated object based on the spatial relationship between the spatial attribute of the associated object and the candidate area comprises: Determine the category of the associated item based on the image features of the associated item extracted from the visible light image, and determine the connection rule corresponding to the category of the associated item based on the correspondence between the preset category and the connection rule; determine whether the connection relationship between the associated item and the candidate transparent obstacle conforms to the determined connection rule based on the spatial relationship between the spatial attribute of the associated item and the candidate area; and / or, The step of determining the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle comprises: According to the image features of the associated objects extracted from the visible light image, the category of the associated objects is determined, and based on the correspondence between the preset categories and the methods for determining the spatial attributes, the method for determining the spatial attributes corresponding to the category of the associated objects is determined; according to the environmental information corresponding to the environmental transparent obstacle and the method for determining the spatial attributes, the spatial attributes of the environmental transparent obstacle are determined.

4. The method according to claim 3, characterized in that The step of determining the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle and the method for determining the spatial attribute comprises at least one of the following: In response to the category of the associated object being a door or window frame, the size and position of the door or window frame and the area enclosed by it are respectively determined as the size and position of the environmental transparent obstacle; In response to the category of the associated object not being a door or window frame, the edges of the non-transparent obstacles around the environmental transparent obstacle are determined based on the connection pattern between the environmental transparent obstacle and the associated object; the size and position of the area formed by the edges of the non-transparent obstacles are respectively determined as the size and position of the environmental transparent obstacle.

5. The method according to claim 2, characterized in that: The step of obtaining the environment information corresponding to the candidate area includes: Taking a preset position in the candidate area as a center point, determining a region to be collected according to preset parameters, wherein the region to be collected includes the candidate area; Obtain a visible light image and a depth image of the area to be collected.

6. The method according to claim 2, characterized in that The step of determining the spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle comprises: Determine the edge of the non-transparent obstacle around the transparent obstacle in the environment; The size and position of the area formed by the edges of the non-transparent obstacles are respectively determined as the size and position of the transparent obstacles in the environment.

7. The method according to any one of claims 1 to 6, characterized in that: The three-dimensional map is a map corresponding to the current scene where the robot is located, and the robot includes a distance measuring sensor; Before the step of marking the environmental transparent obstacles on the three-dimensional map according to the spatial attributes of the environmental transparent obstacles to obtain a three-dimensional map including the transparent obstacles, the method further includes: By means of the distance measuring sensor, a first distance between the robot and the environmental transparent obstacle is obtained; according to the position of the robot and the position included in the spatial attribute of the environmental transparent obstacle, a second distance between the robot and the environmental transparent obstacle is obtained; in response to the difference between the first distance and the second distance being not greater than a preset threshold, the step of marking the environmental transparent obstacle in the three-dimensional map according to the spatial attribute of the environmental transparent obstacle to obtain a three-dimensional map including the transparent obstacle is performed; and / or, The step of determining the candidate area according to the positions of the non-transparent obstacles included in the three-dimensional map includes at least one of the following: Controlling the robot to move in the current scene and collect an environment image, and constructing a three-dimensional map of the current scene based on the position of the robot when collecting the environment image and the scene information included in the environment image; and determining, according to the positions of the non-transparent obstacles included in the three-dimensional map, an area in the three-dimensional map where at least part of the non-transparent obstacles do not exist as a candidate area; Acquire a pre-built three-dimensional map of the current scene; and determine, according to the positions of the non-transparent obstacles included in the three-dimensional map, areas in the three-dimensional map where at least part of the non-transparent obstacles do not exist as candidate areas; During the movement of the robot in the current scene, an environmental image is collected and the position of obstacles is detected by a ranging sensor; a three-dimensional map of the current scene is constructed based on the position of the robot when the environmental image is collected and the scene information included in the environmental image; the position of each obstacle is determined based on the distance of the obstacle detected by the ranging sensor; and an alternative area is determined based on the determined position of the obstacle and the position of the non-transparent obstacle included in the three-dimensional map.

8. A map construction device, characterized in that: The device comprises: An area determination module, used to determine a candidate area according to the position of the non-transparent obstacles included in the three-dimensional map, wherein the candidate area includes an area where the candidate transparent obstacles are located; An information acquisition module, used to acquire environmental information corresponding to the candidate area; an attribute determination module, configured to determine, in response to the environmental information being in accordance with at least one of the environmental characteristics corresponding to the transparent obstacle, for each environmental transparent obstacle corresponding to the environmental characteristic, a spatial attribute of the environmental transparent obstacle according to the environmental information corresponding to the environmental transparent obstacle, wherein the spatial attribute includes at least size and position; The map construction module is used to mark the environmental transparent obstacles in the three-dimensional map according to the spatial attributes of the environmental transparent obstacles, so as to obtain a three-dimensional map including the transparent obstacles.

9. An electronic device, characterized in that: comprising a processor and a memory, wherein the processor communicates with the memory; Memory, used to store computer programs; A processor, for implementing the method steps described in any one of claims 1 to 7 when executing a program stored in a memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps of any one of claims 1 to 7 are implemented.

Images