Dynamic obstacle recognition method and device, storage medium and electronic device

Abstract

The application provides a dynamic obstacle identification method and device, a storage medium and an electronic device. The method comprises: mapping points with a height above the ground and less than or equal to a first height threshold to a graph according to each frame of point cloud of multiple frames of point cloud collected in a collection area, to obtain multiple local graphs; determining obstacles contained in each local graph by performing a clustering operation on points in each local graph; and sequentially performing obstacle identification on the obstacles contained in each local graph to obtain dynamic obstacles in the collection area. The application solves the problem of poor timeliness of dynamic obstacle identification caused by a large amount of data to be processed in the related art.

Description

[0001] This application is a divisional application of application number 202210550728.6, filed on May 20, 2022, entitled "Method and apparatus for identifying dynamic obstacles, storage medium and electronic device". [Technical Field] This application relates to the field of smart homes, and more specifically, to a method and apparatus for identifying dynamic obstacles, a storage medium, and an electronic device. [Background Technology] During the operation of the cleaning equipment, dynamic obstacle detection can be performed using sensors installed on the equipment to identify obstacles that are moving or moving away from the equipment. After identifying a dynamic obstacle, an obstacle avoidance strategy can be implemented to avoid collisions that could cause equipment damage or other abnormal situations.

[0002] Currently, dynamic obstacle recognition is performed by analyzing multiple frames of captured images. The dynamic obstacle is identified based on its positional changes across these frames. However, this method requires processing a large amount of data and cannot identify dynamic obstacles in a timely manner, leading to instances where cleaning equipment still collides with them.

[0003] Therefore, it is evident that the dynamic obstacle recognition methods in related technologies suffer from poor timeliness due to the large amount of data that needs to be processed. [Summary of the Invention] The purpose of this application is to provide a method, apparatus, storage medium, and electronic device for identifying dynamic obstacles, so as to at least solve the problem of poor timeliness of dynamic obstacle identification caused by the large amount of data to be processed in the related art.

[0004] The purpose of this application is to achieve the following technical solution: According to one aspect of the embodiments of this application, a method for identifying dynamic obstacles is provided, comprising: continuously acquiring point cloud data of the acquisition area of ​​a target sensor on a cleaning device to obtain multiple frames of point cloud; mapping points with a height above the ground and less than or equal to a first height threshold onto an image based on each frame of the multiple point cloud to obtain multiple local images; determining the obstacles contained in each local image by performing a clustering operation on the points in each local image; and sequentially identifying the obstacles contained in each local image to obtain the dynamic obstacles within the acquisition area.

[0005] In an exemplary embodiment, the step of continuously acquiring point cloud data of the acquisition area of ​​the target sensor through the target sensor on the cleaning device to obtain multiple frames of point cloud data includes: continuously acquiring point cloud data of the acquisition area of ​​the area array ToF sensor through the area array ToF sensor to obtain the multiple frames of point cloud data.

[0006] In an exemplary embodiment, the step of mapping points with heights above the ground and less than or equal to a first height threshold onto a single image based on each frame of the multi-frame point cloud to obtain multiple local images includes: performing the following steps on each frame of the multi-frame point cloud to obtain the multiple local images, wherein, when performing the following steps, each frame of point cloud is the current frame point cloud: performing ground detection on the current frame point cloud to obtain ground point clouds in the current frame point cloud; mapping points in the current frame point cloud whose heights are above the ground point clouds in the current frame point cloud and less than or equal to the first height threshold onto a single image to obtain a local image corresponding to the current frame point cloud.

[0007] In an exemplary embodiment, the step of performing ground detection on the current frame point cloud to obtain the ground point cloud in the current frame point cloud includes: mapping points in the current frame point cloud whose height is less than or equal to a second height threshold to an image to obtain a reference image corresponding to the current frame point cloud; calculating the height difference between any two adjacent points in the reference image; determining two adjacent points whose height difference is less than or equal to the height difference threshold as a set of candidate ground points corresponding to the current frame point cloud; performing a clustering operation on the set of candidate ground points to obtain a set of multiple candidate ground point clouds; and determining the ground point cloud in the current frame point cloud from the set of multiple candidate ground point clouds based on the point cloud parameters of each of the multiple candidate ground point clouds.

[0008] In an exemplary embodiment, determining the obstacles contained in each local map by performing a clustering operation on the points in each of the plurality of local maps includes: performing the following steps on each of the plurality of local maps to obtain the obstacles contained in each local map, wherein, when performing the following steps, each local map is a current local map: performing a clustering operation on the points in the current local map to obtain a plurality of candidate obstacles, wherein each of the plurality of candidate obstacles corresponds to a cluster obtained by clustering; selecting obstacles with a size greater than or equal to a target size threshold from the plurality of candidate obstacles to obtain the obstacles contained in the current local map.

[0009] In an exemplary embodiment, the step of sequentially identifying obstacles contained in each local image to obtain dynamic obstacles within the acquisition area includes: obtaining the position of any obstacle within the acquisition area in the plurality of local images by performing obstacle matching on the obstacles contained in each local image; determining the dynamic identification result of any obstacle based on the position of any obstacle in the plurality of local images, wherein the dynamic identification result is used to indicate whether the target obstacle is a dynamic obstacle.

[0010] In an exemplary embodiment, determining the target recognition result of any obstacle based on its position within the plurality of local images includes: determining the device position of the cleaning device based on its movement parameters when the cleaning device is in a moving state; converting the position of any obstacle within the plurality of local images into a position in a world coordinate system based on the device position of the cleaning device, thereby obtaining a set of position sequences of the obstacle; and determining the dynamic recognition result of the obstacle based on the distance between two adjacent positions in the set of position sequences.

[0011] According to another aspect of the embodiments of this application, a dynamic obstacle identification device is also provided, comprising: a collection unit, configured to continuously collect point clouds from the collection area of ​​the target sensor on a cleaning device to obtain multiple frames of point clouds; a mapping unit, configured to map points above the ground and less than or equal to a first height threshold to an image based on each frame of the multiple frame point clouds to obtain multiple local images; a clustering unit, configured to determine the obstacles contained in each local image by performing a clustering operation on the points in each local image of the multiple local images; and an identification unit, configured to sequentially identify the obstacles contained in each local image to obtain the dynamic obstacles within the collection area.

[0012] In an exemplary embodiment, the acquisition unit includes: an acquisition module, configured to continuously acquire point clouds from the acquisition area of ​​the area array ToF sensor using an area array ToF sensor, thereby obtaining the multi-frame point cloud.

[0013] In an exemplary embodiment, the mapping unit includes: a first execution module, configured to perform the following steps on each frame of the multi-frame point cloud to obtain the multiple local maps, wherein, when performing the following steps, each frame of point cloud is a current frame point cloud: performing ground detection on the current frame point cloud to obtain ground point clouds in the current frame point cloud; mapping points in the current frame point cloud whose height is above the ground point clouds in the current frame point cloud and less than or equal to the first height threshold to a map to obtain a local map corresponding to the current frame point cloud.

[0014] In an exemplary embodiment, the first execution module includes: a mapping submodule, configured to map points in the current frame point cloud whose height is less than or equal to a second height threshold to an image to obtain a reference image corresponding to the current frame point cloud; a calculation submodule, configured to calculate the height difference between any two adjacent points in the reference image; a first determination submodule, configured to determine two adjacent points whose height difference is less than or equal to the height difference threshold as candidate ground points corresponding to the current frame point cloud; a clustering submodule, configured to perform a clustering operation on the candidate ground points to obtain multiple candidate ground point clouds; and a second determination submodule, configured to determine the ground point cloud in the current frame point cloud from the multiple candidate ground point clouds based on the point cloud parameters of the multiple candidate ground point clouds.

[0015] In an exemplary embodiment, the clustering unit includes: a second execution module, configured to perform the following steps on each of the plurality of local graphs to obtain obstacles contained in each local graph, wherein, when performing the following steps, each local graph is a current local graph: performing a clustering operation on the points within the current local graph to obtain a plurality of candidate obstacles, wherein each of the plurality of candidate obstacles corresponds to a cluster obtained by clustering; selecting obstacles with a size greater than or equal to a target size threshold from the plurality of candidate obstacles to obtain the obstacles contained in the current local graph.

[0016] In an exemplary embodiment, the identification unit includes: a matching module, configured to obtain the position of any obstacle in the acquisition area within the plurality of local images by performing obstacle matching on the obstacles contained in each local image; and a determination module, configured to determine the dynamic identification result of any obstacle based on the position of any obstacle within the plurality of local images, wherein the dynamic identification result is used to indicate whether the target obstacle is a dynamic obstacle.

[0017] In an exemplary embodiment, the determining module includes: a third determining submodule, configured to determine the device position of the cleaning device based on the movement parameters of the cleaning device when the cleaning device is in a moving state; a conversion submodule, configured to convert the position of any obstacle in the multiple local images into a position in the world coordinate system based on the device position of the cleaning device, thereby obtaining a set of position sequences of the any obstacle; and a fourth determining submodule, configured to determine the dynamic recognition result of the any obstacle based on the distance between two adjacent positions in the set of position sequences.

[0018] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer-readable storage medium, and the computer program is configured to execute the above-described dynamic obstacle recognition method at runtime.

[0019] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the dynamic obstacle recognition method through the computer program.

[0020] In this embodiment, dynamic obstacle identification is performed using local point cloud data. The target sensor on the cleaning device continuously collects point cloud data from the target sensor's acquisition area, obtaining multiple frames of point cloud. Based on each frame of the point cloud, points above the ground and less than or equal to a first height threshold are mapped to a single image, resulting in multiple local images. Clustering is performed on the points within each local image to determine the obstacles contained in each local image. Obstacle identification is then performed on the obstacles contained in each local image to obtain the dynamic obstacles within the acquisition area. Because points above the ground and not exceeding the height threshold in each frame of the point cloud are mapped to separate local images, the number of points in each local image is less than in the original point cloud. This reduces the amount of data required for dynamic obstacle identification, improving the timeliness of dynamic obstacle identification. This solves the problem of poor timeliness in dynamic obstacle identification methods in related technologies due to the large amount of data required for processing. [Attached Image Description] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the hardware environment of an optional dynamic obstacle recognition method according to an embodiment of this application; Figure 2 This is a flowchart illustrating an optional dynamic obstacle recognition method according to an embodiment of this application; Figure 3 This is a flowchart of another optional dynamic obstacle recognition method according to an embodiment of this application; Figure 4 This is a structural block diagram of an optional dynamic obstacle recognition device according to an embodiment of this application; Figure 5 This is a structural block diagram of an optional electronic device according to an embodiment of this application.

Detailed Implementation Methods

[0023] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0024] According to one aspect of the embodiments of this application, a method for identifying dynamic obstacles is provided. Optionally, in this embodiment, the above-described method for identifying dynamic obstacles can be applied to, for example... Figure 1 The hardware environment shown consists of cleaning equipment 102, base station 104, and cloud platform 106. For example... Figure 1 As shown, the cleaning device 102 can be connected to the base station 104 and / or the cloud platform 106 via a network to enable interaction between the cleaning device 102 and the base station 104 and / or the cloud platform 106.

[0025] The aforementioned networks may include, but are not limited to, at least one of the following: wired network, wireless network. The aforementioned wired network may include, but is not limited to, at least one of the following: wide area network (WAN), metropolitan area network (MAN), local area network (LAN). The aforementioned wireless network may include, but is not limited to, at least one of the following: Wi-Fi (Wireless Fidelity), Bluetooth, infrared. The network used by the cleaning device 102 to communicate with the base station 104 and / or the cloud platform 106 may be the same as or different from the network used by the base station 104 to communicate with the cloud platform 106. The cleaning device 102 may include, but is not limited to: sweeping machine, floor scrubber, etc.

[0026] The dynamic obstacle identification method of this application embodiment can be executed by the cleaning device 102, the base station 104, or the cloud platform 106 individually, or by at least two of the cleaning device 102, the base station 104, and the cloud platform 106 together. The dynamic obstacle identification method of this application embodiment can also be executed by a client installed on the cleaning device 102 or the base station 104.

[0027] Taking the dynamic obstacle recognition method in this embodiment as an example, which is performed by the cleaning device 102, Figure 2This is a flowchart illustrating an optional dynamic obstacle recognition method according to an embodiment of this application, as shown below. Figure 2 As shown, the process of this method may include the following steps: In step S202, point cloud data is continuously collected from the target sensor's acquisition area using the target sensor on the cleaning equipment to obtain multiple frames of point cloud data.

[0028] The dynamic obstacle recognition method in this embodiment can be applied to scenarios where cleaning equipment identifies dynamic obstacles. Cleaning equipment may include robotic vacuum cleaners, robotic floor scrubbers, or other devices with area cleaning functions. Dynamic obstacle recognition can be performed while the cleaning equipment is stationary or in motion; for example, it can be performed during the cleaning process, while traveling from a base station to the cleaning area, or while returning to the base station. Dynamic obstacles can be obstacles with a certain speed of movement, such as obstacles moving away from or near the cleaning equipment.

[0029] In this embodiment, a target sensor installed on the cleaning equipment can be used for dynamic obstacle recognition. The target sensor can be a sensor used for point cloud data acquisition, such as a dot matrix or linear array laser sensor. The target sensor can be located at the front end, top, left side, right side, or other front-end areas of the cleaning equipment that can collect data in the direction of movement of the cleaning equipment. This embodiment does not limit the type or location of the target sensor.

[0030] Optionally, the target sensor can be a rotatable sensor. For example, the bottom of the target sensor can be connected to a rotatable base, and the acquisition direction of the target sensor can be controlled by rotating the rotatable base. Before continuously acquiring point cloud data of the acquisition area of ​​the target sensor on the cleaning equipment, the rotation angle of the rotatable base can be determined according to the moving direction of the cleaning equipment. After rotation, the target sensor can acquire data of the front area of ​​the cleaning equipment in the moving direction; the rotatable base is rotated according to the determined rotation angle.

[0031] In this embodiment, the cleaning device can continuously collect point cloud data from the target sensor's acquisition area using its onboard target sensor, obtaining multiple frames of point cloud data. The acquisition interval for each frame of point cloud data can be a preset interval, such as acquiring data once per second, acquiring data multiple times per second, etc. When acquiring each frame of point cloud data, the target sensor can send a detection signal to its acquisition area, and the current frame of point cloud data is generated based on the received reflected signal.

[0032] Step S204: Based on each frame of the multi-frame point cloud, map the points whose height is above the ground and less than or equal to the first height threshold to a single image to obtain multiple local images.

[0033] In this embodiment, for each frame of point cloud, the cleaning device can first identify the ground and the height of each point in each frame of point cloud; then, points whose height is above the ground and less than or equal to a first height threshold are mapped to an image, such as a blank image, to obtain a local image. Here, a local image refers to an image containing local point cloud data in a frame of point cloud. By performing the above mapping, the amount of point cloud data that needs to be processed in each frame of point cloud can be reduced, thereby improving the efficiency of obstacle recognition.

[0034] Optionally, the first height threshold can be a preset height threshold, which is at least greater than the height of the cleaning equipment. Considering that the cleaning equipment itself has a certain height, obstacles at higher positions, whether dynamic or static, do not affect the operation of the cleaning equipment; therefore, there is no need to process this part of the data. By filtering out points with heights above the first height threshold, the efficiency of dynamic obstacle recognition can be improved without affecting the operation of the equipment.

[0035] Step S206: By performing a clustering operation on the points in each of the multiple local maps, the obstacles contained in each local map are determined.

[0036] For each local map, the cleaning device can perform a clustering operation on the points within each local map to determine the obstacles contained in each local map. The clustering algorithm used to perform the clustering operation can be a clustering algorithm that specifies the number of clusters (e.g., K-means clustering algorithm) or a clustering algorithm that does not specify the number of clusters (e.g., hierarchical clustering algorithm), and this embodiment does not limit this.

[0037] By performing clustering operations on the points within each local image, multiple clusters corresponding to each local image can be obtained. Each cluster can correspond to one obstacle, multiple obstacles (for example, obstacles overlap at the angle of the cleaning equipment), or it may not be an obstacle. Combining other information of the cluster, such as length, width, and height, obstacle identification can be further performed.

[0038] The number of obstacles in each local view is not fixed, and adjacent local views can contain the same or different numbers of obstacles. When the number of obstacles is the same, the obstacles themselves can be the same or different.

[0039] Step S208: Obstacle identification is performed on the obstacles contained in each local image in sequence to obtain the dynamic obstacles in the acquisition area.

[0040] After identifying the obstacles in each local image, the cleaning equipment matches these obstacles to determine their positions in different local images and at different times. Based on the changes in the position of the same obstacle at different times, it can be determined whether the obstacle is a dynamic obstacle. By identifying whether each obstacle appearing in multiple local images is a dynamic obstacle, dynamic obstacles within the data collection area can be identified.

[0041] After identifying dynamic obstacles within the data collection area, the movement parameters of the dynamic obstacles can be determined. Based on the movement parameters of the cleaning equipment and the dynamic obstacles, it can be determined whether a collision will occur between the cleaning equipment and the dynamic obstacles. If so, an obstacle avoidance strategy can be implemented in advance to reduce the risk of a collision between the cleaning equipment and the dynamic obstacles. The aforementioned movement parameters may include, but are not limited to, at least one of the following: movement speed and movement direction.

[0042] The dynamic obstacle recognition method provided in this embodiment can improve the efficiency of dynamic obstacle recognition, improve the timeliness of obstacle avoidance during obstacle avoidance, reduce the risk of collision between cleaning equipment and dynamic obstacles, and improve the safety of cleaning equipment operation.

[0043] Through steps S202 to S208, point cloud data is continuously collected from the target sensor's acquisition area using the target sensor on the cleaning equipment, resulting in multiple frames of point cloud data. Based on each frame of the multiple point cloud data, points with a height above the ground and less than or equal to a first height threshold are mapped onto a single image to obtain multiple local images. By performing a clustering operation on the points within each local image, obstacles contained in each local image are determined. Obstacle identification is then performed on the obstacles contained in each local image sequentially to obtain the dynamic obstacles within the acquisition area. This solves the problem of poor timeliness in dynamic obstacle identification due to the large amount of data required for processing in related technologies, thus improving the timeliness of dynamic obstacle identification.

[0044] In one exemplary embodiment, point cloud data is continuously acquired from the acquisition area of ​​the target sensor using a target sensor on a cleaning device to obtain multiple frames of point cloud data, including: S11: The point cloud is continuously acquired by the area array ToF sensor to obtain multiple frames of point cloud.

[0045] Using dot matrix or linear array laser sensors to collect point cloud data results in insufficient accuracy in identifying dynamic obstacles due to sensor and calibration errors. Furthermore, the slow scanning speed of these sensors leads to poor timeliness in dynamic obstacle identification.

[0046] In this embodiment, the target sensor is an area array ToF (Time of Flight) sensor. The ToF sensor continuously sends light pulses to the target object (the object within the ToF sensor's acquisition area) and receives the light signals returned from the target object. The distance between the ToF sensor and the target object is obtained by determining the flight (round trip) time of the transmitted and received light pulses.

[0047] During point cloud acquisition, the cleaning equipment can continuously acquire point cloud data from the acquisition area of ​​the array ToF sensor using a Time-of-Flight (ToF) sensor, obtaining multiple frames of point cloud. During each point cloud acquisition, the ToF sensor emits a probe signal into its acquisition area and receives the reflected signal. Based on the time difference between transmitting the probe signal and receiving the reflected signal, the distance between the ToF sensor and the target corresponding to the reflected signal is determined, and a point cloud is generated based on this determined distance. Since the time it takes for the reflected signal from different objects to reach the ToF sensor is different, the point cloud matching the acquisition area can be determined based on the reception time of each reflected signal.

[0048] In this embodiment, point cloud acquisition using an area array ToF sensor can improve the speed and accuracy of point cloud acquisition and enhance the timeliness of dynamic obstacle recognition.

[0049] In an exemplary embodiment, based on each frame of the multi-frame point cloud, points with a height above the ground and less than or equal to a first height threshold are mapped to a single image to obtain multiple local images, including: S21, perform the following steps on each frame of the multi-frame point cloud to obtain multiple local images, wherein each frame of point cloud is the current frame point cloud when performing the following steps: Perform ground detection on the current frame point cloud to obtain the ground point cloud in the current frame point cloud; Map the points in the current frame point cloud whose height is above the ground point cloud in the current frame point cloud and less than or equal to the first height threshold to a map to obtain a local map corresponding to the current frame point cloud.

[0050] In this embodiment, to obtain a local image corresponding to each frame of point cloud, a mapping operation can be performed on each frame of point cloud to obtain a local image corresponding to each frame of point cloud. For example, for each frame of point cloud, it can be used as the current frame of point cloud to perform the following mapping operation to obtain a local image corresponding to the current frame of point cloud: First, perform ground detection on the current frame point cloud to obtain the ground point cloud in the current frame point cloud. Ground detection can be performed using the ground detection methods provided in related technologies, which will not be elaborated here. Then, for each point in the current frame point cloud, the height of the ground point at the same location is determined. For points whose height is not lower than the height of the corresponding ground point and is less than or equal to the first height threshold, they can be mapped onto a map.

[0051] After all points have been processed, a local map corresponding to the current frame point cloud can be obtained. Optionally, to improve the rationality of dynamic obstacle recognition, points in the current frame point cloud whose height is not lower than the height of the corresponding ground point and whose height difference with the corresponding ground point is not greater than a first height threshold can be mapped to a map, thereby obtaining a local map corresponding to the current frame point cloud.

[0052] In this embodiment, by performing ground recognition and determining the points mapped to the local map based on the height relationship between each point in the point cloud and the corresponding ground point, the rationality of the local map determination can be improved.

[0053] In an exemplary embodiment, ground detection is performed on the current frame point cloud to obtain the ground point cloud in the current frame point cloud, including: S31, map the points in the current frame point cloud whose height is less than or equal to the second height threshold to a map to obtain a reference map corresponding to the current frame point cloud; S32, calculate the height difference between any two adjacent points in the reference diagram; S33, determine two adjacent points whose height difference is less than or equal to the height difference threshold as candidate ground points corresponding to the current frame point cloud; S34, perform clustering operation on candidate ground points to obtain multiple candidate ground point clouds; S35, based on the point cloud parameters of multiple candidate ground point clouds, determine the ground point cloud in the current frame point cloud from the multiple candidate ground point clouds.

[0054] When conducting ground inspection, if the ground is assumed to be a plane with a height of 0, due to sensor errors, calibration errors, etc., the ground may not actually be a plane with a height of 0. This can easily lead to the ground being identified as an obstacle, resulting in incomplete cleaning.

[0055] In this embodiment, a certain height redundancy space can be set for the ground, that is, points with a height not higher than the second height threshold are all processed as candidate ground points. For the current frame point cloud, the points in the current frame point cloud with a height not higher than the second height threshold can be mapped to an image to obtain a reference image corresponding to the current frame point cloud. Here, the reference image is an image used for ground detection (or can be considered as a local image), and the second height threshold can be less than the first height threshold.

[0056] Considering the relatively small undulations of the ground, points with small height differences from their neighbors can be identified as candidate ground points. For a reference map corresponding to the current frame's point cloud, the height difference between any two adjacent points in the reference map can be calculated. Two adjacent points in the reference map whose height difference is less than or equal to a height difference threshold are identified as candidate ground points, resulting in multiple candidate ground points corresponding to the current frame's point cloud.

[0057] For multiple candidate ground points, clustering can be performed to obtain multiple candidate ground point clouds. Each candidate ground point cloud can contain a subset of candidate ground points from the multiple candidate ground points. The ground, compared to other obstacles, has unique point clouds; for example, its density, length, width, and area are larger than those of other obstacles. The ground in the current frame's point cloud can be determined from the multiple candidate ground point clouds based on the point cloud parameters of each candidate point cloud. For example, the point cloud parameters of each candidate point cloud can be statistically analyzed, and these parameters can include, but are not limited to, at least one of the following: density, length, width, and area. Point clouds whose point cloud parameters meet a set threshold are identified as ground.

[0058] In this embodiment, candidate ground points are determined by judging the height difference between points with heights below a height threshold and their adjacent points, and the ground is determined by clustering the candidate ground points, which can improve the accuracy of ground identification.

[0059] In one exemplary embodiment, obstacles contained in each local map are determined by performing a clustering operation on points within each of multiple local maps, including: S41, Perform the following steps for each of the multiple local maps to obtain the obstacles contained in each local map, wherein each local map is the current local map when performing the following steps: Perform a clustering operation on the points within the current local graph to obtain multiple candidate obstacles, where each candidate obstacle corresponds to a cluster obtained from the clustering. From multiple candidate obstacles, obstacles with a size greater than or equal to the target size threshold are selected to obtain the obstacles contained in the current local graph.

[0060] In this embodiment, to determine the obstacles contained in each local graph, a clustering operation can be performed on each local graph separately to identify the obstacles contained in each local graph. For example, for each local graph, the following clustering operation can be performed as the current local graph to determine the obstacles contained in the current local graph: First, perform a clustering operation on the points within the current local graph to obtain multiple clusters. Each cluster is then identified as a candidate obstacle, thus obtaining multiple candidate obstacles. Then, determine the obstacle dimensions for each candidate obstacle, for example, by calculating at least one of the length, width, and height for each candidate obstacle; Finally, candidate obstacles whose size is greater than or equal to the target size threshold are identified as obstacles included in the current local graph. That is, each candidate obstacle whose size is greater than or equal to the target size threshold is identified as an obstacle included in the current local graph, thereby obtaining all obstacles included in the current local graph.

[0061] Optionally, the obstacle size may include at least one of the length, width, and height of the candidate obstacle. When determining whether the obstacle size is greater than or equal to the target size threshold, the relationship between each parameter (i.e., length, width, and height) and each parameter threshold (the target size threshold includes each parameter threshold) can be determined separately. Only candidate obstacles whose parameters are all greater than or equal to their corresponding parameter thresholds are determined as actual obstacles. Optionally, the relationship between the product of each parameter and a set size threshold (i.e., the target size threshold) can also be determined. Candidate obstacles whose product of each parameter is greater than or equal to the set size threshold are determined as actual obstacles.

[0062] In this embodiment, by selecting candidate obstacles whose size is greater than or equal to a set size threshold and identifying them as actual obstacles, the accuracy of obstacle identification can be improved.

[0063] In an exemplary embodiment, obstacles are sequentially identified in each local image to obtain dynamic obstacles within the acquisition area, including: S51, by performing obstacle matching on the obstacles contained in each local image, the position of any obstacle in the acquisition area within multiple local images is obtained; S52, determine the dynamic recognition result of any obstacle based on its position in multiple local images.

[0064] In this embodiment, when performing obstacle recognition, obstacle matching can be performed on the obstacles contained in each local image to determine the local images containing the same obstacle in multiple local images, as well as the position of the same obstacle in its local image. Then, based on its position in its corresponding local image, it can be determined whether the same obstacle is a dynamic obstacle. If a dynamic obstacle exists, the dynamic motion state of the dynamic obstacle can be determined, such as the direction of movement and the speed of movement.

[0065] For any obstacle in a local map, such as the current obstacle, the cleaning device can determine the position of the current obstacle across multiple local maps by performing obstacle matching on the obstacles contained in each local map. For local maps that do not contain the current obstacle, the position of the current obstacle in these local maps is empty; for local maps that do contain the current obstacle, the position of the current obstacle in these local maps is the position of the current obstacle in the sensor coordinate system.

[0066] Based on the current obstacle's position within multiple local images, the cleaning device can determine the dynamic recognition result of the current obstacle. This dynamic recognition result can be used to indicate whether the current obstacle is a dynamic obstacle. If the position change of the current obstacle within the multiple local images is small (e.g., the position change range is less than a preset range threshold; considering detection accuracy, a small position change can be considered as no position change), the current obstacle can be considered a static obstacle. If the position change of the current obstacle is large (e.g., the position change range is greater than or equal to the preset range threshold), the current obstacle can be considered a dynamic obstacle. By performing the above dynamic obstacle recognition process on any obstacle as the current obstacle, all dynamic obstacles in the multi-frame point cloud can be identified.

[0067] This embodiment improves the accuracy of obstacle recognition by dynamically identifying the same obstacle based on its position in multiple local images.

[0068] In one exemplary embodiment, determining the target recognition result of any obstacle based on its position within multiple local images includes: S61, when the cleaning equipment is in a moving state, determine the equipment position of the cleaning equipment according to the moving parameters of the cleaning equipment; S62, based on the location of the cleaning equipment, convert the position of any obstacle in multiple local images into its position in the world coordinate system, and obtain a set of position sequences for any obstacle; S63, determine the dynamic recognition result of any obstacle based on the distance between two adjacent positions in a set of position sequences.

[0069] The position of any obstacle within multiple local images represents its position in the sensor coordinate system. The actual position of any obstacle depends not only on its position within the local images but also on the position of the cleaning equipment. When the cleaning equipment is in motion, the dynamic recognition result of any obstacle can be determined based on the equipment's position and the obstacle's position within the multiple local images: Based on the equipment's position, the obstacle's position within the local images is converted to its position in the world coordinate system, resulting in a set of position sequences for each obstacle; the dynamic recognition result is then determined based on the distance between two adjacent positions within this sequence.

[0070] The location of the cleaning equipment can be its position in a world coordinate system, which can be a three-dimensional coordinate system with the equipment's initial position as its origin. When converting the position of any obstacle across multiple local images to its position in the world coordinate system based on the equipment's location, a set of reference positions for any obstacle in the equipment coordinate system can be determined using the obstacle's position in the local images and the coordinate transformation relationship between the sensor coordinate system and the cleaning equipment's coordinate system. Based on the cleaning equipment's location and the coordinate transformation relationship between the equipment coordinate system and the world coordinate system, this set of reference positions is then converted into a set of positions in the world coordinate system.

[0071] Based on the distance between two adjacent positions in a set of positions, the positional change of any obstacle can be determined. Similar to the previous embodiment, the magnitude of the positional change of any obstacle can be used to determine whether any obstacle is a dynamic obstacle, thereby obtaining the dynamic recognition result of any obstacle.

[0072] In this embodiment, obstacle identification is performed based on the location of the cleaning equipment and the location of the obstacle within multiple local images, which can improve the accuracy of dynamic obstacle identification.

[0073] The method for identifying dynamic obstacles in this application embodiment will be explained below with reference to optional examples. In this optional example, the cleaning device is a robot vacuum cleaner (i.e., a robotic vacuum cleaner), and the target sensor is an area array ToF sensor.

[0074] This optional example provides an obstacle recognition method based on an area array ToF sensor, which can identify obstacles based on multiple frames of ToF data, thereby identifying objects that are moving or moving away from the robot vacuum.

[0075] Combination Figure 3 As shown, the obstacle recognition method in this optional example may include the following steps: Step S302: Obstacle recognition is performed on each frame of point cloud in the consecutive frames of point cloud to obtain multi-frame results, and the multi-frame results are saved.

[0076] The single-frame ToF data is processed as follows: a local map is created, and ToF points that are above the ground but below a certain threshold above the fuselage are mapped onto it; the points in the local map are clustered; the length, width, and height of each clustered point cloud are calculated, and if the length, width, and height of a point cloud are greater than a certain threshold, the point cloud is saved.

[0077] Step S304: Compare the saved multi-frame results. If movement or disappearance occurs between frames, it is determined to be a dynamic obstacle.

[0078] This optional example demonstrates how dynamic obstacle recognition based on an area array ToF sensor can improve both the accuracy and efficiency of dynamic obstacle recognition.

[0079] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0080] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM (Read-Only Memory) / RAM (Random Access Memory), magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0081] According to another aspect of the embodiments of this application, a dynamic obstacle identification device for implementing the above-described dynamic obstacle identification method is also provided. Figure 4 This is a structural block diagram of an optional dynamic obstacle recognition device according to an embodiment of this application, such as... Figure 4 As shown, the device may include: The acquisition unit 402 is used to continuously acquire point cloud data of the acquisition area of ​​the target sensor through the target sensor on the cleaning equipment, and obtain multiple frames of point cloud data. The mapping unit 404 is connected to the acquisition unit 402 and is used to map points that are above the ground and less than or equal to a first height threshold onto a map based on each frame of the multi-frame point cloud, so as to obtain multiple local maps. Clustering unit 406, connected to mapping unit 404, is used to determine the obstacles contained in each local map by performing clustering operations on the points in each local map among multiple local maps. The identification unit 408, connected to the clustering unit 406, is used to sequentially identify obstacles contained in each local image to obtain dynamic obstacles within the acquisition area.

[0082] It should be noted that the acquisition unit 402 in this embodiment can be used to perform the above step S202, the mapping unit 404 in this embodiment can be used to perform the above step S204, the clustering unit 406 in this embodiment can be used to perform the above step S206, and the identification unit 408 in this embodiment can be used to perform the above step S208.

[0083] Through the above modules, point cloud data is continuously collected from the target sensor's acquisition area using the target sensor on the cleaning equipment, resulting in multiple frames of point cloud data. Based on each frame of the point cloud data, points above the ground and less than or equal to a first height threshold are mapped onto a single image to obtain multiple local images. By performing clustering operations on the points within each local image, obstacles contained in each local image are determined. Obstacle identification is then performed on the obstacles contained in each local image to obtain the dynamic obstacles within the acquisition area. This solves the problem of poor timeliness in dynamic obstacle identification due to the large amount of data required in related technologies, thus improving the timeliness of dynamic obstacle identification.

[0084] In one exemplary embodiment, the acquisition unit includes: The acquisition module is used to continuously acquire point clouds from the acquisition area of ​​the array ToF sensor using an array ToF sensor, and obtain multiple frames of point clouds.

[0085] Optional examples of this implementation scheme can be found in the examples shown in the above-described equipment operation control method, which will not be repeated here.

[0086] In one exemplary embodiment, the mapping unit includes: The first execution module is used to perform the following steps on each frame of the multi-frame point cloud to obtain multiple local images, wherein each frame of point cloud is the current frame point cloud when performing the following steps: Perform ground detection on the current frame point cloud to obtain the ground point cloud in the current frame point cloud; Map the points in the current frame point cloud whose height is above the ground point cloud in the current frame point cloud and less than or equal to the first height threshold to a map to obtain a local map corresponding to the current frame point cloud.

[0087] Optional examples of this implementation scheme can be found in the examples shown in the above-described equipment operation control method, which will not be repeated here.

[0088] In one exemplary embodiment, the first execution module includes: The mapping submodule is used to map points in the current frame point cloud whose height is less than or equal to the second height threshold to a map, so as to obtain a reference map corresponding to the current frame point cloud. The calculation submodule is used to calculate the height difference between any two adjacent points in the reference map. The first determination submodule is used to determine two adjacent points whose height difference is less than or equal to the height difference threshold as candidate ground points corresponding to the current frame point cloud. The clustering submodule is used to perform clustering operations on candidate ground points to obtain multiple candidate ground point clouds; The second determining submodule is used to determine the ground point cloud in the current frame point cloud from multiple candidate ground point clouds based on the point cloud parameters of multiple candidate ground point clouds.

[0089] Optional examples of this implementation scheme can be found in the examples shown in the above-described equipment operation control method, which will not be repeated here.

[0090] In one exemplary embodiment, the clustering unit includes: The second execution module is used to perform the following steps on each of the multiple local graphs to obtain the obstacles contained in each local graph, wherein each local graph is the current local graph when performing the following steps: Perform a clustering operation on the points within the current local graph to obtain multiple candidate obstacles, where each candidate obstacle corresponds to a cluster obtained from the clustering. From multiple candidate obstacles, obstacles with a size greater than or equal to the target size threshold are selected to obtain the obstacles contained in the current local graph.

[0091] Optional examples of this implementation scheme can be found in the examples shown in the above-described equipment operation control method, which will not be repeated here.

[0092] In one exemplary embodiment, the identification unit includes: The matching module is used to obtain the position of any obstacle in the acquisition area within multiple local images by performing obstacle matching on the obstacles contained in each local image; The determination module is used to determine the dynamic recognition result of any obstacle based on its position in multiple local images. The dynamic recognition result is used to indicate whether the target obstacle is a dynamic obstacle.

[0093] Optional examples of this implementation scheme can be found in the examples shown in the above-described equipment operation control method, which will not be repeated here.

[0094] In one exemplary embodiment, the determining module includes: The third determination submodule is used to determine the position of the cleaning equipment based on its movement parameters when the cleaning equipment is in motion. The conversion submodule is used to convert the position of any obstacle in multiple local images into its position in the world coordinate system based on the location of the cleaning equipment, thus obtaining a set of position sequences for any obstacle; The fourth determination submodule is used to determine the dynamic recognition result of any obstacle based on the distance between two adjacent positions in a set of position sequences.

[0095] Optional examples of this implementation scheme can be found in the examples shown in the above-described equipment operation control method, which will not be repeated here.

[0096] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should also be noted that the above modules, as part of a device, can operate in environments such as... Figure 1 The hardware environment shown can be implemented through software or hardware, and the hardware environment includes the network environment.

[0097] According to another aspect of the embodiments of this application, a storage medium is also provided. Optionally, in this embodiment, the storage medium can be used to execute program code for any of the dynamic obstacle recognition methods described in the embodiments of this application.

[0098] Optionally, in this embodiment, the storage medium may be located on at least one of the network devices in the network shown in the above embodiment.

[0099] Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps: S1, continuously collect point cloud data of the target sensor's acquisition area through the target sensor on the cleaning equipment to obtain multiple frames of point cloud data; S2, based on each frame of the multi-frame point cloud, map the points whose height is above the ground and less than or equal to the first height threshold to a map to obtain multiple local maps. S3, by performing a clustering operation on the points in each of the multiple local maps, determines the obstacles contained in each local map; S4. Obstacle identification is performed on the obstacles contained in each local image in sequence to obtain the dynamic obstacles in the acquisition area.

[0100] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and will not be repeated in this embodiment.

[0101] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, ROMs, RAMs, portable hard drives, magnetic disks, or optical disks.

[0102] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described dynamic obstacle recognition method is also provided. The electronic device may be a server, a terminal, or a combination thereof.

[0103] Figure 5 This is a structural block diagram of an optional electronic device according to an embodiment of this application, such as... Figure 5 As shown, the aforementioned electronic device includes a processor 502, a communication interface 504, a memory 506, and a communication bus 508. The processor 502, communication interface 504, and memory 506 communicate with each other via the communication bus 508. Memory 506 is used to store computer programs; When processor 502 executes a computer program stored in memory 506, it performs the following steps: S1, continuously collect point cloud data of the target sensor's acquisition area through the target sensor on the cleaning equipment to obtain multiple frames of point cloud data; S2, based on each frame of the multi-frame point cloud, map the points whose height is above the ground and less than or equal to the first height threshold to a map to obtain multiple local maps. S3, by performing a clustering operation on the points in each of the multiple local maps, determines the obstacles contained in each local map; S4. Obstacle identification is performed on the obstacles contained in each local image in sequence to obtain the dynamic obstacles in the acquisition area.

[0104] Optionally, in this embodiment, the communication bus can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, Figure 5 The symbol is represented by a single thick line, but this does not indicate that there is only one bus or one type of bus. The communication interface is used for communication between the aforementioned electronic device and other devices.

[0105] The aforementioned memory may include RAM, or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0106] As an example, the memory 506 described above may include, but is not limited to, the acquisition unit 402, mapping unit 404, clustering unit 406, and identification unit 408 from the control device of the aforementioned device. Furthermore, it may include, but is not limited to, other module units from the control device of the aforementioned device, which will not be elaborated upon in this example.

[0107] The processor mentioned above can be a general-purpose processor, including but not limited to: CPU (Central Processing Unit), NP (Network Processor), etc.; it can also be DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0108] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and will not be repeated here.

[0109] Those skilled in the art will understand that Figure 5The structure shown is for illustrative purposes only. The device that implements the above-described dynamic obstacle recognition method can be a terminal device, such as a smartphone (e.g., Android phone, iOS phone), tablet computer, PDA, mobile Internet Devices (MID), PAD, etc. Figure 5 This does not limit the structure of the aforementioned electronic device. For example, the electronic device may also include components that are more... Figure 5 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 5 The different configurations shown.

[0110] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, ROM, RAM, disk or optical disk, etc.

[0111] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0112] If the integrated units in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in the aforementioned computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0113] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0114] In the several embodiments provided in this application, it should be understood that the disclosed client can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between units or modules, and may be electrical or other forms.

[0115] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the solution provided in this embodiment, depending on actual needs.

[0116] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0117] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for identifying dynamic obstacles, characterized in that, include: Based on each frame of point cloud collected in the collection area, points with a height above the ground and less than or equal to a first height threshold are mapped onto a single image to obtain multiple local images. By performing a clustering operation on the points within each of the multiple local maps, the obstacles contained in each local map are determined. Obstacle identification is performed sequentially on the obstacles contained in each local image to obtain the dynamic obstacles within the acquisition area. The step of determining the obstacles contained in each local map by performing a clustering operation on the points within each of the multiple local maps includes: For each of the plurality of local images, the following steps are performed to obtain the obstacles contained in each local image, wherein each local image is the current local image when performing the following steps: Clustering is performed on the points within the current local graph to obtain multiple candidate obstacles, wherein each of the multiple candidate obstacles corresponds to a cluster obtained by clustering. From the plurality of candidate obstacles, obstacles with a size greater than or equal to the target size threshold are selected to obtain the obstacles contained in the current local graph.

2. The method according to claim 1, characterized in that, The method for acquiring multi-frame point clouds includes: The target sensor on the cleaning equipment continuously collects point cloud data of the acquisition area of ​​the target sensor to obtain multiple frames of point cloud data. The target sensor includes a ToF area array sensor.

3. The method according to claim 1, characterized in that, The step of mapping points with a height above the ground and less than or equal to a first height threshold to a single image based on each frame of the multi-frame point cloud to obtain multiple local images includes: The following steps are performed on each of the multiple point cloud frames to obtain the multiple local images, wherein each point cloud frame is the current frame point cloud during the following steps: Perform ground detection on the current frame point cloud to obtain the ground point cloud in the current frame point cloud; In the current frame point cloud, points whose height is above the ground point cloud in the current frame point cloud and less than or equal to the first height threshold are mapped onto a map to obtain a local map corresponding to the current frame point cloud.

4. The method according to claim 3, characterized in that, The step of performing ground detection on the current frame point cloud to obtain the ground point cloud in the current frame point cloud includes: In the current frame point cloud, points whose height is less than or equal to the second height threshold are mapped onto a map to obtain a reference map corresponding to the current frame point cloud. Calculate the height difference between any two adjacent points in the reference diagram; Two adjacent points whose height difference is less than or equal to the height difference threshold are identified as candidate ground points corresponding to the current frame point cloud. Clustering is performed on the candidate ground points to obtain multiple candidate ground point clouds; Based on the point cloud parameters of the multiple candidate ground point clouds, the ground point cloud in the current frame point cloud is determined from the multiple candidate ground point clouds.

5. The method according to claim 1, characterized in that, The size parameters of the candidate obstacles include at least one of length, width, and height. The step of selecting obstacles with sizes greater than or equal to a target size threshold from the plurality of candidate obstacles to obtain the obstacles included in the current local image includes: Determine the relationship between each size parameter and its corresponding threshold value, and identify candidate obstacles whose size parameters are all greater than or equal to their respective threshold values ​​as the obstacles; or Determine the relationship between the product of each size parameter and the target size threshold, and identify candidate obstacles whose product of each size parameter is greater than or equal to the target size threshold as obstacles.

6. The method according to any one of claims 1 to 5, characterized in that, The step of sequentially identifying obstacles in each local image to obtain dynamic obstacles within the acquisition area includes: By performing obstacle matching on the obstacles contained in each of the local images, the position of any obstacle in the acquisition area within the multiple local images can be obtained; Based on the position of any obstacle within the multiple local images, determine the dynamic recognition result of any obstacle.

7. The method according to claim 6, characterized in that, The step of determining the target recognition result of any obstacle based on its position within the multiple local images includes: When the cleaning equipment is in a moving state, the equipment position of the cleaning equipment is determined according to the moving parameters of the cleaning equipment; Based on the location of the cleaning equipment, the position of any obstacle in the multiple local images is converted into the position in the world coordinate system, resulting in a set of position sequences of any obstacle. The dynamic recognition result of any obstacle is determined based on the distance between two adjacent positions in the set of position sequences.

8. A device for identifying dynamic obstacles, characterized in that, include: The mapping unit is used to map points that are above the ground and less than or equal to a first height threshold to a single image based on each frame of point cloud collected in the acquisition area, so as to obtain multiple local images. A clustering unit is used to determine the obstacles contained in each local map by performing a clustering operation on the points in each of the multiple local maps; The identification unit is used to sequentially identify obstacles contained in each local image to obtain the dynamic obstacles within the acquisition area. The step of determining the obstacles contained in each local map by performing a clustering operation on the points within each of the multiple local maps includes: For each of the plurality of local images, the following steps are performed to obtain the obstacles contained in each local image, wherein each local image is the current local image when performing the following steps: Clustering is performed on the points within the current local graph to obtain multiple candidate obstacles, wherein each of the multiple candidate obstacles corresponds to a cluster obtained by clustering. From the plurality of candidate obstacles, obstacles with a size greater than or equal to the target size threshold are selected to obtain the obstacles contained in the current local graph.

9. The apparatus according to claim 8, characterized in that, Also includes: The acquisition unit is used to continuously acquire point cloud data of the acquisition area of ​​the target sensor through the target sensor on the cleaning equipment to obtain multiple frames of point cloud data. The target sensor includes a ToF area array sensor.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program, when executed, performs the method of any one of claims 1 to 7.

11. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to execute the method of any one of claims 1 to 7 through the computer program.

Images