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

Abstract

The present application provides a method and device for dynamic object recognition, a storage medium, and an electronic device. The method comprises: determining a group of target grid cells from a target grid map based on the passable parameters of each grid cell in the target grid map, wherein the passable parameters of each target grid cell in the group of target grid cells meet preset parameter conditions; performing height detection on each target grid cell based on the unit position of the group of target grid cells in the target grid map to obtain a detected height value for each target grid cell; and performing dynamic object recognition on each target grid cell based on the detected height value of each target grid cell and the preset height value of each target grid cell. The above technical solution solves the problem of low dynamic object recognition efficiency in the dynamic object recognition method in the related art due to the need to traverse and check all map grid cells in the map.

Description

Technical field

[0001] The present application relates to the field of robotics, and more specifically, to a method and device for identifying dynamic objects, a storage medium, and an electronic device. [Background Technology]

[0002] Currently, achieving high-precision and reliable positioning in the fields of autonomous mobile robot navigation and autonomous driving often relies on high-precision maps. When constructing maps, dynamic objects, such as pedestrians and vehicles, are inevitably present in the environment. The introduction of dynamic objects can contaminate the map and adversely impact higher-level applications such as map-based positioning and path planning.

[0003] In related technologies, to create a clean map free of dynamic objects, map construction typically involves traversing all grid cells in the map, identifying and removing dynamic obstacles by determining whether each cell contains them. However, this method of map construction is very time-consuming and inefficient because it requires traversing all grid cells.

[0004] It can be seen that the dynamic object recognition method in the related art has the problem of low efficiency of dynamic object recognition due to the need to traverse and check all map grid cells in the map. [Summary of the invention]

[0005] The purpose of this application is to provide a dynamic object recognition method and device, a storage medium and an electronic device, so as to at least solve the problem of low efficiency of dynamic object recognition in the related art due to the need to traverse and check all map grid cells in the map.

[0006] The purpose of this application is achieved through the following technical solutions:

[0007] According to one aspect of an embodiment of the present application, a dynamic object recognition method is provided, comprising: determining a group of target grid cells from a target grid map based on a passable parameter of each grid cell in the target grid map, wherein the passable parameter of each target grid cell in the group of target grid cells satisfies a preset parameter condition; performing height detection on each target grid cell based on a cell position of the group of target grid cells in the target grid map to obtain a detected height value of each target grid cell; and performing dynamic object recognition on each target grid cell based on the detected height value of each target grid cell and a preset height value of each target grid cell.

[0008] In an exemplary embodiment, before determining a group of target grid cells from the target grid map based on the passable parameters of each grid cell in the target grid map, the method further includes: determining the passable parameters corresponding to the preset height value of each grid cell based on a mapping relationship between the height value of the grid cell and the passable value of the grid cell, wherein the passable parameters of each grid cell are the passable values ​​of each grid cell.

[0009] In an exemplary embodiment, the passable parameter of each grid cell is a passable value of each grid cell; and determining a group of target grid cells from the target grid map based on the passable parameter of each grid cell in the target grid map includes: determining each grid cell in the target grid map whose passable value is less than or equal to a target passable threshold as a target grid cell, thereby obtaining the group of target grid cells.

[0010] In an exemplary embodiment, determining a group of target grid cells from the target grid map based on the passable parameters of each grid cell in the target grid map includes: obtaining grid cells in the target grid map that are within the field of view of a perception sensor to obtain a group of candidate grid cells; and determining the group of target grid cells from the group of candidate grid cells based on the passable parameters of each candidate grid cell in the group of candidate grid cells.

[0011] In an exemplary embodiment, performing height detection on each target grid cell according to the cell position of the group of target grid cells in the target grid map to obtain the detection height value of each target grid cell includes: looping through the following steps until the detection height value of each target grid cell is determined; determining a reference grid cell from the grid cells within the field of view of the perception sensor, wherein a current light path formed by the reference grid cell and the sensing light emission position on the perception sensor passes through a target grid cell in the group of target grid cells for which a corresponding detection height value has not been determined; determining target grid cells in all grid cells passed by the vertical projection of the current light path on the target grid map for which a corresponding detection height value has not been determined, to obtain a group of grid cells to be detected; and determining the vertical distance between the current light path and each grid cell to be detected in the group of grid cells to be detected as the detection height value of each grid cell to be detected.

[0012] In an exemplary embodiment, the dynamic object recognition is performed on each target grid cell based on the detection height value of each target grid cell and the preset height value of each target grid cell, including: performing the following steps on each target grid cell respectively, wherein, when performing the following steps, each target grid cell is the current grid cell: when the detection height value of the current grid cell is less than the preset height value of the current grid cell, determining that a dynamic object is recognized from the current grid cell; when the detection height value of the current grid cell is greater than or equal to the preset height value of the current grid cell, determining that no dynamic object is recognized from the current grid cell.

[0013] In an exemplary embodiment, after performing dynamic object recognition on each target grid cell based on the detected height value of each target grid cell and the preset height value of each target grid cell, the method further includes: resetting the preset height value of the target grid cell in the group of target grid cells in which no dynamic object is recognized to a reference height value, to obtain an updated preset height value of each grid cell; determining a passable parameter of each grid cell based on the updated preset height value of each grid cell; determining a moving path of a mobile device based on the passable parameter of each grid cell, to obtain a target moving path; and controlling the mobile device to move within a target area corresponding to the target grid map according to the target moving path.

[0014] According to another aspect of an embodiment of the present application, a dynamic object recognition device is further provided, including: a first determination unit, configured to determine a group of target grid cells from a target grid map based on the passable parameters of each grid cell in the target grid map, wherein the passable parameters of each target grid cell in the group of target grid cells meet preset parameter conditions; a detection unit, configured to perform height detection on each target grid cell according to the unit position of the group of target grid cells in the target grid map to obtain a detected height value of each target grid cell; and an identification unit, configured to perform dynamic object recognition on each target grid cell based on the detected height value of each target grid cell and a preset height value of each target grid cell.

[0015] In an exemplary embodiment, the device further includes: a second determination unit for determining, before determining a group of target grid cells from the target grid map based on the passable parameters of each grid cell in the target grid map, the passable parameters corresponding to the preset height value of each grid cell according to a mapping relationship between the height value of the grid cell and the passable value of the grid cell, wherein the passable parameters of each grid cell are the passable values ​​of each grid cell.

[0016] In an exemplary embodiment, the passable parameter of each grid cell is the passable value of each grid cell; the first determination unit includes: a first determination module, configured to determine each grid cell in the target grid map whose passable value is less than or equal to the target passable threshold as a target grid cell, thereby obtaining the group of target grid cells.

[0017] In an exemplary embodiment, the first determination unit includes: an acquisition module for acquiring grid cells within the field of view of the perception sensor in the target grid map to obtain a group of candidate grid cells; and a second determination module for determining the group of target grid cells from the group of candidate grid cells based on the passable parameters of each candidate grid cell in the group of candidate grid cells.

[0018] In an exemplary embodiment, the second determination module includes: an execution submodule, which is used to loop through the following steps until the detection height value of each target grid cell is determined; determining a reference grid cell from the grid cells within the field of view of the perception sensor, wherein the current light path formed by the reference grid cell and the sensing light emission position on the perception sensor passes through the target grid cell for which the corresponding detection height value has not been determined in the group of target grid cells; determining the target grid cell for which the corresponding detection height value has not been determined among all the grid cells passed by the vertical projection of the current light path on the target grid map, to obtain a group of grid cells to be detected; and determining the vertical distance between the current light path and each grid cell to be detected in the group of grid cells to be detected as the detection height value of each grid cell to be detected.

[0019] In an exemplary embodiment, the first identification unit includes: an execution module for performing the following steps on each target grid cell, respectively, wherein, when performing the following steps, each target grid cell is a current grid cell: when the detection height value of the current grid cell is less than the preset height value of the current grid cell, determining that a dynamic object is identified from the current grid cell; when the detection height value of the current grid cell is greater than or equal to the preset height value of the current grid cell, determining that no dynamic object is identified from the current grid cell.

[0020] In an exemplary embodiment, the device further includes: a reset unit for resetting the preset height value of a target grid cell in the group of target grid cells in which no dynamic object is identified to a reference height value after dynamic object identification is performed on each target grid cell based on the detected height value of each target grid cell and the preset height value of each target grid cell, thereby obtaining an updated preset height value of each grid cell; a third determination unit for determining a passable parameter of each grid cell based on the updated preset height value of each grid cell; a fourth determination unit for determining a moving path of a mobile device based on the passable parameter of each grid cell, thereby obtaining a target moving path; and a control unit for controlling the mobile device to move within a target area corresponding to the target grid map according to the target moving path.

[0021] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored. The computer program is configured to execute the above-mentioned dynamic object recognition method when running.

[0022] According to another aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the dynamic object recognition method through the computer program.

[0023] In an embodiment of the present application, a method of selecting some grid cells based on the passable parameters of the grid cells for dynamic object recognition is adopted. A group of target grid cells is determined from the target grid map based on the passable parameters of each grid cell in the target grid map, wherein the passable parameters of each target grid cell in the group of target grid cells meet preset parameter conditions; a height detection is performed on each target grid cell based on the unit position of the group of target grid cells in the target grid map to obtain a detected height value of each target grid cell; and dynamic object recognition is performed on each target grid cell based on the detected height value of each target grid cell and the preset height value of each target grid cell. Since dynamic object recognition only needs to be performed on the grid cells in the grid map whose passable parameters meet the preset parameter conditions, the purpose of reducing the grid cells that need to be traversed can be achieved, thereby achieving the technical effect of improving the recognition efficiency of dynamic objects, thereby solving the problem of low dynamic object recognition efficiency caused by the need to traverse and check all map grid cells in the map in the dynamic object recognition method in the related art.

Brief Description of the Drawings

[0024] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0025] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0026] Figure 1 is a schematic diagram of a hardware environment of an optional dynamic object recognition method according to an embodiment of the present application;

[0027] Figure 2 is a flow chart of an optional dynamic object recognition method according to an embodiment of the present application;

[0028] Figure 3 is a schematic diagram of an optional method for determining a reference grid cell according to an embodiment of the present application;

[0029] Figure 4 is a schematic diagram of another optional method for determining a reference grid cell according to an embodiment of the present application;

[0030] Figure 5 is a schematic diagram of an optional method of performing dynamic object recognition on a grid cell according to an embodiment of the present application;

[0031] Figure 6 is a schematic diagram of another optional dynamic object recognition method according to an embodiment of the present application;

[0032] Figure 7 is a flowchart of another optional dynamic object recognition method according to an embodiment of the present application;

[0033] Figure 8 is a structural block diagram of an optional dynamic object recognition device according to an embodiment of the present application;

[0034] Figure 9 This is a structural block diagram of an optional electronic device according to an embodiment of the present application. [Specific implementation method]

[0035] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0036] It should be noted that the terms "first", "second", etc. in the description and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0037] According to one aspect of the embodiment of the present application, a method for identifying a dynamic object is provided. Optionally, in this embodiment, the method for identifying a dynamic object can be applied to Figure 1 In the hardware environment shown in FIG. 1 , a mobile device 102 and a server 104 are configured. Figure 1 As shown, the server 104 is connected to the mobile device 102 via a network, and can be used to provide services (such as game services, application services, etc.) for the mobile device or the client installed on the mobile device. A database can be set up on the server or independently of the server to provide data storage services for the server 104.

[0038] The mobile device 102 may include one or more devices. Different mobile devices may communicate with each other through a server or directly without going through a server. Optionally, the mobile device 102 may include at least one of the following: a mobile robot, an autonomous vehicle, and the mobile robot may be a legged robot, including but not limited to at least one of the following: a delivery robot, a cleaning robot, and the delivery robot may include an object delivery robot or a food delivery robot, and the cleaning robot may include a sweeping robot (abbreviated as a sweeper), a floor scrubber (abbreviated as a scrubber), etc.

[0039] The aforementioned network may include, but is not limited to, at least one of the following: a wired network and a wireless network. The aforementioned wired network may include, but is not limited to, at least one of the following: a wide area network, a metropolitan area network, and a local area network. The aforementioned wireless network may include, but is not limited to, at least one of the following: Wi-Fi (Wireless Fidelity) and Bluetooth.

[0040] The dynamic object recognition method of the embodiment of the present application can be executed by the server 104, or by the mobile device 102, or by both the server 104 and the mobile device 102. The dynamic object recognition method of the embodiment of the present application can also be executed by the client installed on the mobile device 102.

[0041] Taking the mobile device 102 as an example, Figure 2 FIG. 1 is a flow chart of an optional dynamic object recognition method according to an embodiment of the present application, such as Figure 2 As shown, the process of the method may include the following steps:

[0042] Step S202 : determining a group of target grid cells from the target grid map according to the passable parameters of each grid cell in the target grid map, wherein the passable parameters of each target grid cell in the group of target grid cells meet preset parameter conditions.

[0043] The dynamic object recognition method in this embodiment can be applied to scenarios where dynamic objects in a target area are identified when controlling a mobile device to perform autonomous navigation or unmanned driving in the target area. The identified dynamic objects can be applied to scenarios where an area map of the target area is constructed. The target area can be an outdoor area, such as a highway, a main road, etc., or an indoor area, such as an office, a factory workshop, etc.

[0044] In order to reduce the pollution of the map caused by the introduction of dynamic objects and the adverse effects on upper-layer applications such as map-based positioning and path planning, and thereby improve the accuracy of mobile devices during autonomous navigation or unmanned driving, when constructing a passable map, the mobile device can identify and remove dynamic objects (dynamic obstacles, such as pedestrians, vehicles, etc.) in the passable map. In related technologies, it is usually adopted to traverse and check all grid cells in the map, and identify and remove dynamic objects by judging whether each grid cell contains dynamic objects. However, the method of traversing and checking all grid cells in the map is very time-consuming and labor-intensive, and the efficiency of map construction is low.

[0045] In this embodiment, grid cells that require dynamic object recognition can be selected from the constructed traversable map based on the traversable parameters of the grid cells, and then dynamic object recognition can be performed on the selected grid cells, so that dynamic objects can be automatically and efficiently removed, thereby solving at least part of the above-mentioned problems.

[0046] The mobile device can obtain a constructed regional map of the target area. The constructed map type is a raster map, that is, a target raster map. The target raster map supports multi-layer maps and can include at least the following types of map layers: a passable map layer. The passable map layer contains the passability parameters of each raster map. For example, the passable map layer can save the passability score of the terrain, that is, the passability value. The valid range of the passability value can be [0,1], where 0 represents completely impassable and 1 represents completely passable.

[0047] There are one or more ways to obtain the target grid map. This can include obtaining a regional map of the target area from a database to obtain the target grid map. Alternatively, the target grid map can be obtained by invoking a regional map provided by a third-party module to obtain various types of required map layers, and then fusing the obtained map layers to obtain the target grid map. Alternatively, the target grid map can be obtained by other means, which are not limited in this embodiment. With respect to the traversable map layer or traversable map, the traversable map can be a pre-built offline traversable map or a traversable map built online in real time.

[0048] After acquiring the target grid map, the mobile device may acquire a set of target grid cells in order to clear the information generated by dynamic objects in the target grid map. The target grid cells are grid cells whose passable parameters meet preset parameter conditions. The preset parameter conditions may be that the parameter value of the passable parameter is within a set parameter value range, or other preset parameter conditions. The preset parameter conditions are not limited in this embodiment.

[0049] Step S204 : performing height detection on each target grid cell according to the cell positions of a group of target grid cells in the target grid map to obtain a detected height value of each target grid cell.

[0050] The mobile device can detect the height of each grid cell in the set of target grid cells based on the cell position of the set of target grid cells in the target grid cells to obtain a detected height value of each target grid cell. Here, the detected height value of each target grid cell can be the maximum height within the grid cell at the time of detection, or the average height, or a height determined by other means, which is not limited in this embodiment.

[0051] Each time the height of a target grid cell is detected, the number of target grid cells detected at a time may be one or more. The detected height value of each target grid cell may be detected by a perception sensor on a mobile device. The perception sensor may be any sensor capable of measuring distance, including but not limited to at least one of the following: a lidar, a depth camera, a monocular camera, a binocular stereo camera, etc., or other sensors for detecting grid cell height, which are not limited in this embodiment.

[0052] Step S206 : performing dynamic object recognition on each target grid cell according to the detected height value of each target grid cell and the preset height value of each target grid cell.

[0053] In addition to the traversable map layer, the target grid map may also include an elevation map layer. The elevation map layer may store the height value data of the environment to form a 2.5D map (pseudo-three-dimensional map). The above two layers of grid map interfaces are universal, and the elevation map and traversable map may be provided by any third-party module. In this embodiment, there is no limitation on the generation method of the traversable map layer and the elevation map layer.

[0054] In this embodiment, after obtaining the detected height value of each grid cell, the mobile device can perform dynamic object recognition for each target grid cell based on the preset height value of each target detection cell. The preset height value here can be the height value of the grid cell in the elevation map layer. The mobile device can determine the grid cell containing the dynamic object based on the detected height value of each grid cell and the preset height value of each target detection cell.

[0055] The detected height value of the target grid cell can be the actual height value of the target grid cell, or it can be a range of height values ​​of the target grid cell, for example, greater than a certain height value, less than a certain height value, etc. Correspondingly, when dynamic object recognition is performed on the target grid cell, whether the target grid cell contains a dynamic object can be determined based on the relationship between the detected height value of the target grid cell and the preset height value of the target grid cell.

[0056] Through the above steps S202 to S206, a group of target grid cells are determined from the target grid map based on the passable parameters of each grid cell in the target grid map, wherein the passable parameters of each target grid cell in the group of target grid cells meet preset parameter conditions; a height detection is performed on each target grid cell based on the unit position of the group of target grid cells in the target grid map to obtain a detected height value of each target grid cell; and dynamic object recognition is performed on each target grid cell based on the detected height value of each target grid cell and the preset height value of each target grid cell. This solves the problem of low dynamic object recognition efficiency caused by the need to traverse and check all map grid cells in the map in dynamic object recognition methods in related arts, thereby improving the efficiency of dynamic object recognition.

[0057] In an exemplary embodiment, before determining a group of target grid cells from the target grid map based on the traversable parameters of each grid cell in the target grid map, the method further includes:

[0058] S11, determining a passable parameter corresponding to a preset height value of each grid cell according to a mapping relationship between the height value of the grid cell and the passable value of the grid cell, wherein the passable parameter of each grid cell is the passable value of each grid cell.

[0059] In this embodiment, to use the passability parameters of the grid cells in the target grid map, the passability parameters of each grid cell may be determined first. Alternatively, the mobile device may determine the passability parameters of the grid cell based on the height value of the grid cell. Here, the passability parameters of the grid cell may be the passability values ​​of the grid cell, which may be used to represent the probability that the grid cell is passable.

[0060] For each grid cell, based on the mapping relationship between the height value of the grid cell and the passability value of the grid cell, the preset height value of each grid cell is mapped to the corresponding passability value, thereby obtaining the passability value of each grid cell, that is, the passability parameter of each grid cell. Correspondingly, the preset parameter condition can be that the passability value is within the passability value range, for example, the passability value is not 1, the passability value is less than a set parameter threshold (for example, 0.3), or other conditions related to the passability value, which are not limited in this embodiment.

[0061] Through this embodiment, based on the mapping relationship between the height value of the grid cell and the passable value of the grid cell, the height value of the grid cell is mapped to the corresponding passable value, which can improve the convenience of determining the passable map layer.

[0062] In an exemplary embodiment, the passability parameter of each grid cell is a passability value for each grid cell. The meaning of the passability value is similar to that in the aforementioned embodiment and is not further described here. Correspondingly, based on the passability parameter of each grid cell in the target grid map, a group of target grid cells is determined from the target grid map, including:

[0063] S21 , determining each grid cell in the target grid map whose passability value is less than or equal to the target passability threshold as a target grid cell, thereby obtaining a group of target grid cells.

[0064] In this embodiment, in order to improve the efficiency of dynamic object recognition, the mobile device can first obtain the target pass threshold. The above-mentioned target pass threshold is used to filter out the grid cells that need to be traversed in the target grid map. It can be pre-configured, and can be a threshold set by the user according to needs, a default threshold, or other thresholds. This embodiment does not limit the setting method of the target pass threshold.

[0065] Based on the target passability threshold, the mobile device can determine from the target grid map each grid cell whose passability value is less than or equal to the target passability threshold, obtain the target grid cell, and then obtain a group of target grid cells. For example, if the target passability threshold is 0.3, the legged robot can determine the grid cells with passability values ​​less than or equal to 0.3 as a group of grid cells.

[0066] According to this embodiment, the accuracy and convenience of dynamic object recognition can be improved by screening grid cells by setting the pass threshold and the passable value of the grid cells.

[0067] In an exemplary embodiment, determining a set of target grid cells from the target grid map based on the traversable parameters of each grid cell in the target grid map includes:

[0068] S31, obtaining grid cells within the viewing range of the perception sensor in the target grid map to obtain a set of candidate grid cells;

[0069] S32 : determining a group of target grid cells from the group of candidate grid cells according to the accessible parameters of each candidate grid cell in the group of candidate grid cells.

[0070] In this embodiment, due to the limited viewing angle of the perception sensor on the mobile device, grid cells outside the viewing angle of the perception sensor require the mobile device to be moved in order to be perceived. Therefore, at the current location, only grid cells within the viewing angle of the perception sensor need to be processed. Accordingly, the mobile device can first obtain grid cells within the viewing angle of the perception sensor to obtain a set of candidate grid cells. The obtained set of candidate grid cells can be grid cells within the viewing angle of the perception sensor for which the grid height can be determined.

[0071] After obtaining a set of candidate grid cells, the mobile device can determine a set of target grid cells from the set of candidate grid cells based on the traversability parameters of each candidate grid cell in the set of candidate grid cells. The method of selecting target grid cells based on the traversability parameters is similar to the method of determining a set of target grid cells from the target grid map in the aforementioned embodiment, and will not be repeated here.

[0072] For example, all map grid cells within the effective viewing angle of the perception sensor and having a lower passability value (eg, 0.3, where the threshold value can be determined by the user based on an assessment of the robot's basic movement capabilities) may be recorded.

[0073] Through this embodiment, the grid cells to be identified are determined from the grid cells within the viewing angle of the sensor, which can improve the convenience and efficiency of dynamic object recognition.

[0074] In an exemplary embodiment, according to the unit positions of a group of target grid cells in the target grid map, performing height detection on each target grid cell to obtain the detected height value of each target grid cell includes:

[0075] S41, looping through the following steps until the detection height value of each target grid cell is determined;

[0076] Determining a reference grid cell from grid cells within a viewing angle of the perception sensor, wherein a current light path formed by the reference grid cell and a sensing light emission position on the perception sensor passes through a target grid cell for which no corresponding detection height value is determined in a group of target grid cells;

[0077] Determine the target grid cells whose corresponding detection height values ​​are not determined among all grid cells passed by the vertical projection of the current light path on the target grid map, and obtain a group of grid cells to be detected;

[0078] The vertical distance between the current light path and each grid cell to be detected in a group of grid cells to be detected is determined as a detection height value of each grid cell to be detected.

[0079] In this embodiment, in order to determine the detection height value of each target grid cell, the following operations may be performed on a group of grid cells to obtain the detection height value of each target detection cell:

[0080] Step 1: Determine a reference grid cell from the grid cells within the viewing angle of the perception sensor.

[0081] The mobile device can determine a reference grid cell from the grid cells within the field of view of the perception sensor. The reference grid cell and the perception light emission position of the perception sensor form a current light path. The current light path can be a light path formed by connecting the two points from the light emission point position of the perception sensor as the starting point and the reference grid cell as the end point. The current light path passes through at least one of the target grid cells in a group of target grid cells whose corresponding detection height value has not been determined.

[0082] There may be one or more ways to determine the reference grid cell from the grid cells within the viewing angle range, which may include at least one of the following: determining the grid cell within the sensor viewing angle range and located on the ray passing through the position of the perceived light emission and a target grid cell with an undetermined corresponding detection height value as the reference grid cell. The reference grid cell may also be determined by other methods, which are not limited in this embodiment.

[0083] Step 2: determine the target grid cells whose corresponding detection height values ​​are not determined among all the grid cells passed by the vertical projection of the current light path on the target grid map, and obtain a group of grid cells to be detected.

[0084] The mobile device can vertically project the current light path onto the target grid map, thereby obtaining all grid cells that the vertical projection of the current light path on the target grid map passes through, and determine the target grid cells whose corresponding height values ​​have not been determined from all the grid cells that the vertical projection of the current light path on the target grid map passes through, thereby obtaining a set of grid cells to be detected.

[0085] For example, for Figure 3 and Figure 4The scenario shown here depicts a legged robot (e.g., a quadruped robot) equipped with a perception sensor for autonomous navigation. The robot is located on a traversable grid map (i.e., a target grid map) and can use the onboard perception sensor to identify dynamic obstacles. The perception sensor's effective field of view (i.e., effective viewing angle) can include areas with low traversable grid cells. An optical path is formed by selecting a static object within the effective field of view that reflects the light path and the location of the perception light emitted by the perception sensor.

[0086] When performing dynamic obstacle recognition on low-passable grid cells, only a traversal check is required. Figure 3 Whether there is a dynamic obstacle in the grid cell is determined by checking whether there is line of sight interference between the grid cells in the circular area (representing low passability values) and the light path. For example, the light emission point position of the sensing sensor can be used as the starting point, and the current measurement reflection point position (i.e., the aforementioned reference grid cell) can be used as the end point, and the two points can be connected to form a light path SP. The current light path is vertically projected onto the grid map, and the grid cells with low passability values ​​among all the grid cells projected by the light path in the grid map are counted to form a set T_low (i.e., the aforementioned set of grid cells to be detected).

[0087] Step three: determining the vertical distance between the current light path and each grid cell to be detected in a group of grid cells to be detected as a detection height value of each grid cell to be detected.

[0088] After obtaining a set of grid cells to be detected, the mobile device can obtain the detection height value of each grid cell to be detected. The detection height value of each grid cell to be detected can be obtained by determining the vertical distance between the current light path and each grid cell to be detected as the detection height value of each grid cell to be detected. The detection height value here is not necessarily the actual height value of each grid cell to be detected, but an estimated value. The detection height value of each grid cell to be detected is not lower than the actual height value of each grid cell to be detected (otherwise, the light path cannot pass through the grid cell to reach the reference grid cell).

[0089] Through this embodiment, the grid unit to be detected is determined by the optical path formed between the selected reference grid unit and the position of the light emission point, and then the detection height value of the grid unit to be detected is determined based on the formed optical path, which can improve the efficiency and accuracy of dynamic object recognition.

[0090] In an exemplary embodiment, performing dynamic object recognition on each target grid cell according to the detected height value of each target grid cell and the preset height value of each target grid cell includes:

[0091] S51, performing the following steps for each target grid cell, wherein each target grid cell is a current grid cell when performing the following steps:

[0092] When the detected height value of the current grid cell is less than the preset height value of the current grid cell, it is determined that a dynamic object is recognized from the current grid cell:

[0093] When the detected height value of the current grid cell is greater than or equal to the preset height value of the current grid cell, it is determined that no dynamic object is recognized from the current grid cell.

[0094] In this embodiment, when performing dynamic object recognition on each target grid cell, the mobile device may use each target grid cell as a current grid cell and perform the following steps to determine whether there is a dynamic object in each target grid cell:

[0095] Step 1: When the detected height value of the current grid cell is less than the preset height value of the current grid cell, it is determined that a dynamic object is recognized from the current grid cell.

[0096] If the detected height value of the current grid cell is less than the preset height value of the current grid cell, it means that the height value of the current grid cell has decreased, indicating that a dynamic object has been identified from the current grid cell. For example, in the above scenario where the detected height value is determined by the light path, if the detected height value of the current grid cell is less than the preset height value of the current grid cell, it means that the original height of the current grid cell does not allow the formation of the current light path. However, a light path has now been formed through the current grid cell, indicating that the originally high obstacle has disappeared and the current grid cell contains a dynamic object.

[0097] Step 2: When the detected height value of the current grid cell is greater than or equal to the preset height value of the current grid cell, it is determined that no dynamic object is recognized from the current grid cell.

[0098] If the detection height value of the current grid cell is greater than or equal to the preset height value of the current grid cell, it at least indicates that no object message in the current grid cell is detected, and it can be determined that no dynamic object is identified from the current grid cell.

[0099] For example, Figure 5 As shown, to sense the position of the light emission point of the sensor ( Figure 5 The point S in the figure is the starting point, and the current reflection point position ( Figure 5 The point P in the image is the end point, and the two points are connected to form a light path SP. The current grid unit to the corresponding point of the current light path ( Figure 5 The vertical height value H_max of point C in the figure is H_max. The height value H_max is calculated based on the principle of similar triangles, and the calculation formula is shown in formula (1):

[0100] H_max=CP / SP*(H_sensor-H_lowest)+H_lowest (1)

[0101] Among them, CP is the distance from the corresponding point C on the optical path to the reflection point P, Figure 5 The triangles SPA and CPB in the figure are a pair of similar triangles. The lengths of AP and BP can be derived from the distances between grid cells on the traversable grid map. H_sensor is the height from the sensor to the ground, and H_lowest is the height of the static object. The calculation formula for CP can then be shown in formula (2):

[0102] CP=SP*(BP / AP) (2)

[0103] After obtaining the detection height value H_max, the legged robot can determine whether line of sight interference occurs by calculating the vertical distance value from the grid cell in the low passable value area to the corresponding point of the light path and comparing it with the grid cell height value. That is, the legged robot can compare H_max with the current grid cell height value elevation. When the elevation is less than H_max (such as Figure 5 In the case shown in Figure (a), the legged robot can determine that there is no line of sight interference, and then determine that there is no dynamic obstacle in the vertical direction of this grid unit; and when the elevation is greater than H_max (such as Figure 5 In the case shown in Figure (b) in the figure, the legged robot can determine that there is interference in the line of sight, and then determine that a dynamic object appears in the vertical direction of this grid unit.

[0104] See also Figure 6 , the elevation of the grid cell where the pedestrian is located is the height of the pedestrian. When the pedestrian leaves, the sensing light emitted by the sensing sensor can pass through these grid cells. At this time, the elevation is greater than H_max, indicating that the original dynamic obstacle on the grid cell is no longer on the grid.

[0105] According to this embodiment, by comparing the preset height value and the detected height value of the grid cell to determine whether a dynamic object is detected on the grid cell, the efficiency and accuracy of dynamic object recognition can be improved.

[0106] In an exemplary embodiment, after performing dynamic object recognition on each target grid cell based on the detected height value of each target grid cell and the preset height value of each target grid cell, the method further includes:

[0107] S61, resetting the preset height values ​​of target grid cells in a group of target grid cells where no dynamic object is recognized to the reference height value, to obtain an updated preset height value of each grid cell;

[0108] S62, determining a passable parameter of each grid cell according to the updated preset height value of each grid cell;

[0109] S63, determining a moving path for the mobile device based on the passable parameters of each grid cell to obtain a target moving path;

[0110] S64: Control the mobile device to move within the target area corresponding to the target grid map according to the target moving path.

[0111] In this embodiment, after dynamic object recognition is performed for each target grid cell, the mobile device may reset the preset height values ​​of the grid cells in a group of target grid cells where no dynamic objects are recognized to the reference height value, thereby obtaining an updated preset height value for each grid cell. The reference height value is used to indicate that the height of the grid cell has been reset to the reference height, or in other words, the initial height without dynamic objects.

[0112] After updating the reference for each grid cell, the mobile device can determine the passability parameter of each grid cell based on the preset height value of each grid cell. The manner in which the mobile device determines the passability parameter of each grid cell based on the preset height value of each grid cell is similar to the manner in which the passability parameter of each grid cell is determined based on the preset height value of each grid cell in the aforementioned embodiment, and will not be further described here.

[0113] For example, combined with Figure 5 If elevation is greater than H_max, a dynamic obstacle is determined to exist in the vertical direction of this grid cell, and the grid cell value is cleared and set to the initial value. Otherwise, no dynamic obstacle is determined to exist, and the grid cell value remains unchanged. After evaluating one light path, the next light path within the effective field of view can be evaluated until every grid cell in the T_low set is evaluated, resulting in an updated traversable map.

[0114] After determining the passability parameters of each grid cell, the mobile device can determine a movement path for the mobile device based on the passability parameters of each grid cell to obtain a target movement path. For example, the mobile device can plan a movement path for the mobile device based on the starting point and end point of the movement, selecting grid cells with high passability parameters, and thus determine the target movement path. The passability parameters of all grid cells passed by the target movement path are greater than a set parameter threshold. The mobile device can move along the target movement path, for example, the mobile device can move along the target movement path at a preset speed.

[0115] Through this embodiment, by updating the constructed grid map and performing mobile path planning based on the updated grid map, the accuracy of autonomous navigation of the mobile device can be improved.

[0116] The dynamic object recognition method in this embodiment is explained below with reference to an optional example. In this optional example, the mobile device is a legged robot, the target grid map is a traversable map, and the dynamic object is a dynamic obstacle.

[0117] To address the related art problem of removing dynamic obstacles from a map, which requires traversing and checking all map grid cells within the entire sensor's field of view. This is time-consuming, inefficient, and prone to mistakenly deleting non-dynamic obstacle map portions, this optional example provides a solution for efficiently removing dynamic obstacles from a map. The provided dynamic obstacle detection and removal algorithm is based on ray tracing technology, using existing traversable values ​​in the map as prior knowledge to reduce the number of map grid cells that need to be traversed. Compared with the aforementioned method that requires traversing all map grid cells, this can significantly reduce the amount of traversal calculations, thereby accelerating the frequency of map updates.

[0118] like Figure 7 As shown, the process of the dynamic object recognition method in this optional example may include the following steps:

[0119] Step S702: construct a traversable map.

[0120] Step S704: Count the grid cells with lower passability values ​​within the effective range of the current perception sensor's viewing angle.

[0121] Step S706: Connect the grid cells from the light emission position of the sensing sensor to the current measurement reflection point to form a light path.

[0122] Step S708 , counting the line grid cells formed by the vertical projection of the current light path on the map, and recording all grid cells on this line with lower passability values.

[0123] Step S710 , calculating the vertical height from one of the grid cells to the current light path, which is recorded as H_max.

[0124] Step S712: compare H_max with the current grid cell height value elevation.

[0125] Step S714, determine whether the elevation is greater than H_max, if so, execute step S716, otherwise, execute step S722.

[0126] Step S716 is a dynamic obstacle.

[0127] Step S718: Clear the occupied grid cell height from the map and set it to the initial value.

[0128] Step S720: traverse the next grid cell.

[0129] Step S722: Not a dynamic obstacle.

[0130] Step S724: Keep the height value of the grid unit unchanged.

[0131] Step S726, traverse the next light path.

[0132] Combine Figure 6 During the autonomous navigation process of the legged robot, a pedestrian passes by within the visual range in front. Through the above-mentioned dynamic obstacle detection and removal method, the pedestrian can be quickly detected as a dynamic obstacle and removed from the grid map.

[0133] This example uses the existing traversability values ​​in the map as prior knowledge to reduce the number of map grid cells that need to be traversed. This can greatly reduce the amount of calculation, thereby increasing the frequency of map updates and reducing the probability of non-dynamic obstacles being mistakenly cleared.

[0134] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0135] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it 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 the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM (Read-Only Memory, Read-Only Memory) / RAM (Random Access Memory, Random Access Memory), a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0136] According to another aspect of the embodiments of the present application, a dynamic object recognition device for implementing the above-mentioned dynamic object recognition method is also provided. Figure 8 is a structural block diagram of an optional dynamic object recognition device according to an embodiment of the present application, such as Figure 8 As shown, the device may include:

[0137] A first determining unit 802 is configured to determine a group of target grid cells from the target grid map based on the passable parameters of each grid cell in the target grid map, wherein the passable parameters of each target grid cell in the group of target grid cells meet a preset parameter condition;

[0138] a detection unit 804 connected to the first determination unit 802, configured to perform height detection on each target grid cell according to a unit position of the group of target grid cells in the target grid map, and obtain a detected height value of each target grid cell;

[0139] The identification unit 806 is connected to the detection unit 804 and is used to perform dynamic object recognition on each target grid cell according to the detected height value of each target grid cell and the preset height value of each target grid cell.

[0140] It should be noted that the first determining unit 802 in this embodiment can be used to execute the above step S202, the detecting unit 804 in this embodiment can be used to execute the above step S204, and the identifying unit 806 in this embodiment can be used to execute the above step S206.

[0141] Through the above module, a group of target grid cells are determined from the target grid map based on the passable parameters of each grid cell in the target grid map, wherein the passable parameters of each target grid cell in the group of target grid cells meet preset parameter conditions; based on the unit positions of the group of target grid cells in the target grid map, each target grid cell is height-detected to obtain a detected height value of each target grid cell; and based on the detected height value of each target grid cell and the preset height value of each target grid cell, dynamic object recognition is performed on each target grid cell. This solves the problem of low dynamic object recognition efficiency in dynamic object recognition methods in related arts due to the need to traverse and check all map grid cells in the map, thereby improving the efficiency of dynamic object recognition.

[0142] In an exemplary embodiment, the apparatus further comprises:

[0143] The second determination unit is used to determine the passable parameter corresponding to the preset height value of each grid cell according to the mapping relationship between the height value of the grid cell and the passable value of the grid cell before determining a group of target grid cells from the target grid map according to the passable parameter of each grid cell in the target grid map, wherein the passable parameter of each grid cell is the passable value of each grid cell.

[0144] In an exemplary embodiment, the passable parameter of each grid cell is a passable value of each grid cell; and the first determining unit includes:

[0145] The first determination module is configured to determine each grid cell in the target grid map whose passability value is less than or equal to the target passability threshold as a target grid cell, thereby obtaining a group of target grid cells.

[0146] In an exemplary embodiment, the first determining unit includes:

[0147] An acquisition module is used to acquire grid cells within the viewing range of the perception sensor in the target grid map to obtain a set of candidate grid cells;

[0148] The second determining module is configured to determine a group of target grid cells from a group of candidate grid cells according to the accessible parameters of each candidate grid cell in the group of candidate grid cells.

[0149] In an exemplary embodiment, the second determining module includes:

[0150] An execution submodule is used to loop through the following steps until the detection height value of each target grid cell is determined;

[0151] Determining a reference grid cell from grid cells within a viewing angle of the perception sensor, wherein a current light path formed by the reference grid cell and a sensing light emission position on the perception sensor passes through a target grid cell for which no corresponding detection height value is determined in a group of target grid cells;

[0152] Determine the target grid cells whose corresponding detection height values ​​are not determined among all grid cells passed by the vertical projection of the current light path on the target grid map, and obtain a group of grid cells to be detected;

[0153] The vertical distance between the current light path and each grid cell to be detected in a group of grid cells to be detected is determined as a detection height value of each grid cell to be detected.

[0154] In an exemplary embodiment, the first identification unit includes:

[0155] The execution module is configured to execute the following steps for each target grid cell, wherein each target grid cell is a current grid cell when executing the following steps:

[0156] When the detected height value of the current grid cell is less than the preset height value of the current grid cell, determining that a dynamic object is recognized from the current grid cell;

[0157] When the detected height value of the current grid cell is greater than or equal to the preset height value of the current grid cell, it is determined that no dynamic object is recognized from the current grid cell.

[0158] In an exemplary embodiment, the apparatus further comprises:

[0159] a reset unit for resetting the preset height values ​​of target grid cells in a group of target grid cells where no dynamic object is identified to a reference height value after performing dynamic object identification on each target grid cell based on the detected height value of each target grid cell and the preset height value of each target grid cell, thereby obtaining an updated preset height value of each grid cell;

[0160] a third determining unit, configured to determine a passable parameter of each grid cell according to the updated preset height value of each grid cell;

[0161] a fourth determining unit, configured to determine a moving path of the mobile device according to the passable parameters of each grid unit to obtain a target moving path;

[0162] The control unit is used to control the mobile device to move within a target area corresponding to the target grid map according to the target movement path.

[0163] 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 contents disclosed in the above embodiments. Figure 1 The hardware environment shown can be implemented through software or hardware, wherein the hardware environment includes a network environment.

[0164] According to another aspect of the embodiments of the present application, a storage medium is further provided. Optionally, in this embodiment, the storage medium can be used to execute the program code of any of the above-mentioned dynamic object recognition methods in the embodiments of the present application.

[0165] Optionally, in this embodiment, the above-mentioned storage medium may be located on at least one network device among the multiple network devices in the network shown in the above-mentioned embodiment.

[0166] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps:

[0167] S1, determining a group of target grid cells from the target grid map according to the passable parameters of each grid cell in the target grid map, wherein the passable parameters of each target grid cell in the group of target grid cells meet a preset parameter condition;

[0168] S2, performing height detection on each target grid cell according to the unit positions of a group of target grid cells in the target grid map, and obtaining a detection height value of each target grid cell;

[0169] S3, performing dynamic object recognition on each target grid cell according to the detected height value of each target grid cell and the preset height value of each target grid cell.

[0170] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, which will not be described in detail in this embodiment.

[0171] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media that can store program codes, such as a USB flash drive, a ROM, a RAM, a mobile hard disk, a magnetic disk, or an optical disk.

[0172] According to another aspect of the embodiments of the present application, an electronic device for implementing the above-mentioned dynamic object recognition method is also provided. The electronic device can be a server, a terminal, or a combination thereof.

[0173] Figure 9 is a structural block diagram of an optional electronic device according to an embodiment of the present application, such as Figure 9As shown, it includes a processor 902, a communication interface 904, a memory 906 and a communication bus 908, wherein the processor 902, the communication interface 904 and the memory 906 communicate with each other through the communication bus 908, wherein,

[0174] Memory 906, for storing computer programs;

[0175] The processor 902 is configured to execute the computer program stored in the memory 906 to implement the following steps:

[0176] S1, determining a group of target grid cells from the target grid map according to the passable parameters of each grid cell in the target grid map, wherein the passable parameters of each target grid cell in the group of target grid cells meet a preset parameter condition;

[0177] S2, performing height detection on each target grid cell according to the unit positions of a group of target grid cells in the target grid map, and obtaining a detection height value of each target grid cell;

[0178] S3, performing dynamic object recognition on each target grid cell according to the detected height value of each target grid cell and the preset height value of each target grid cell.

[0179] Optionally, in this embodiment, the communication bus may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 9 The communication interface is used for communication between the electronic device and other devices.

[0180] The aforementioned memory may include RAM, or may include non-volatile memory (non-volatile memory), for example, at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0181] As an example, the memory 906 may include, but is not limited to, the first determining unit 802, the detecting unit 804, and the identifying unit 806 in the control device of the device. In addition, it may also include, but is not limited to, other module units in the control device of the device, which will not be repeated in this example.

[0182] The above-mentioned processor 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 Processing), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0183] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, and this embodiment will not be described in detail here.

[0184] It can be understood by those skilled in the art that Figure 9 The structure shown is for illustration only. The device for implementing the above-mentioned dynamic object recognition method may be a terminal device, which may be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (Mobile Internet Devices, MID), a PAD, and other terminal devices. Figure 9 It does not limit the structure of the above electronic device. For example, the electronic device may also include Figure 9 More or fewer components (such as network interfaces, display devices, etc.) shown in, or with Figure 9 Different configurations shown.

[0185] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, which can include: a flash drive, ROM, RAM, a magnetic disk or an optical disk, etc.

[0186] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0187] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above-mentioned computer-readable storage medium. Based on this understanding, the technical solution of the present application, 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, which is stored in a storage medium and includes several instructions for enabling one or more computer devices (which can be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application.

[0188] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0189] In the several embodiments provided in this application, it should be understood that the disclosed client can be implemented in other ways. Among them, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, and can be electrical or other forms.

[0190] The units described as separate components may or may not be physically separate, and 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 these units may be selected based on actual needs to achieve the purpose of the solution provided in this embodiment.

[0191] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0192] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for identifying dynamic objects, characterized in that: include: Determining a group of target grid cells from the target grid map according to the passable parameters of each grid cell in the target grid map, wherein the passable parameters of each target grid cell in the group of target grid cells meet a preset parameter condition; Performing height detection on each target grid cell according to a unit position of the group of target grid cells in the target grid map to obtain a detection height value of each target grid cell; Performing dynamic object recognition on each target grid cell according to the detected height value of each target grid cell and the preset height value of each target grid cell; Wherein, the dynamic object recognition is performed on each target grid cell according to the detection height value of each target grid cell and the preset height value of each target grid cell, comprising: performing the following steps on each target grid cell respectively, wherein, when performing the following steps, each target grid cell is a current grid cell: when the detection height value of the current grid cell is less than the preset height value of the current grid cell, determining that a dynamic object is recognized from the current grid cell; when the detection height value of the current grid cell is greater than or equal to the preset height value of the current grid cell, determining that no dynamic object is recognized from the current grid cell; the target grid map also includes an elevation map layer, the elevation map layer is used to store the height value data of the environment, and the preset height value refers to the height value of the grid cell in the elevation map layer; Wherein, performing height detection on each target grid cell according to the unit position of the group of target grid cells in the target grid map to obtain the detection height value of each target grid cell includes: looping through the following steps until the detection height value of each target grid cell is determined; determining a reference grid cell from the grid cells within the field of view of the perception sensor, wherein a current light path formed by the reference grid cell and the sensing light emission position on the perception sensor passes through a target grid cell for which a corresponding detection height value has not been determined in the group of target grid cells; determining target grid cells for which a corresponding detection height value has not been determined among all grid cells passed by the vertical projection of the current light path on the target grid map, to obtain a group of grid cells to be detected; and determining the vertical distance between the current light path and each grid cell to be detected in the group of grid cells to be detected as the detection height value of each grid cell to be detected.

2. The method according to claim 1, wherein Before determining a group of target grid cells from the target grid map based on the accessible parameters of each grid cell in the target grid map, the method further includes: According to the mapping relationship between the height value of the grid cell and the passable value of the grid cell, the passable parameter corresponding to the preset height value of each grid cell is determined, wherein the passable parameter of each grid cell is the passable value of each grid cell.

3. The method according to claim 1, wherein The passable parameter of each grid cell is the passable value of each grid cell; The step of determining a group of target grid cells from the target grid map based on the accessible parameters of each grid cell in the target grid map comprises: Each grid cell in the target grid map whose passability value is less than or equal to the target pass threshold is determined as a target grid cell to obtain the group of target grid cells.

4. The method according to claim 1, wherein The step of determining a group of target grid cells from the target grid map based on the accessible parameters of each grid cell in the target grid map comprises: Obtaining grid cells within the viewing angle of the perception sensor in the target grid map to obtain a set of candidate grid cells; The group of target grid cells is determined from the group of candidate grid cells according to the accessible parameters of each candidate grid cell in the group of candidate grid cells.

5. The method according to any one of claims 1 to 4, characterized in that After performing dynamic object recognition on each target grid cell according to the detected height value of each target grid cell and the preset height value of each target grid cell, the method further includes: Resetting the preset height values ​​of the target grid cells in the group of target grid cells where no dynamic object is identified to the reference height value to obtain an updated preset height value of each of the grid cells; Determining a passable parameter of each grid cell according to the updated preset height value of each grid cell; Determine a moving path for the mobile device according to the passable parameters of each grid unit to obtain a target moving path; The mobile device is controlled to move within a target area corresponding to the target grid map according to the target movement path.

6. A dynamic object recognition device, characterized in that: include: A first determining unit is configured to determine a group of target grid cells from the target grid map according to the passable parameter of each grid cell in the target grid map, wherein the passable parameter of each target grid cell in the group of target grid cells satisfies a preset parameter condition; a detection unit, configured to perform height detection on each target grid cell according to a unit position of the group of target grid cells in the target grid map, and obtain a detected height value of each target grid cell; an identification unit, configured to perform dynamic object identification on each target grid cell according to a detected height value of each target grid cell and a preset height value of each target grid cell; The identification unit includes: an execution module, configured to respectively perform the following steps on each target grid cell, wherein when performing the following steps, each target grid cell is a current grid cell: when the detected height value of the current grid cell is less than the preset height value of the current grid cell, determining that a dynamic object is identified from the current grid cell; when the detected height value of the current grid cell is greater than or equal to the preset height value of the current grid cell, determining that no dynamic object is identified from the current grid cell; the target grid map also includes an elevation map layer, the elevation map layer is used to store the height value data of the environment, and the preset height value refers to the height value of the grid cell in the elevation map layer; The device further includes: an execution submodule, configured to loop through the following steps until the detection height value of each target grid cell is determined; determining a reference grid cell from the grid cells within the viewing angle of the perception sensor, wherein a current light path formed by the reference grid cell and the sensing light emission position on the perception sensor passes through a target grid cell for which a corresponding detection height value has not been determined in the group of target grid cells; determining target grid cells for which a corresponding detection height value has not been determined among all grid cells passed by the vertical projection of the current light path on the target grid map, to obtain a group of grid cells to be detected; and determining the vertical distance between the current light path and each grid cell to be detected in the group of grid cells to be detected as the detection height value of each grid cell to be detected.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 5 when executed.

8. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 5 through the computer program.

Images