Obstacle detection method, device, computer equipment and storage medium

Abstract

The present disclosure provides an obstacle detection method, apparatus, computer device, and storage medium. The obstacle detection method includes: determining a first location point corresponding to an object to be identified in the target scene based on current frame radar scan data obtained by scanning a target scene; determining whether the object to be identified is a target object to be deleted based on historical location information corresponding to historical candidate deletion objects in historical frame radar scan data and the first location point; and determining that the object to be identified is an obstacle in the target scene in response to the object to be identified not being the target object to be deleted. The obstacle detection method provided by the embodiments of the present disclosure has high accuracy when detecting obstacles.

Description

Technical Field

[0001] The present disclosure relates to the field of radar detection technology, and in particular to an obstacle detection method, apparatus, computer equipment, and storage medium. Background Art

[0002] LiDAR typically detects a driving area by emitting laser light and receiving reflections from the ground and obstacles. However, if non-obstacle objects in the driving area reflect the laser light, such as water on the road or wet lane markings, the abnormally reflected laser light will be interpreted as an abnormal object, resulting in lower accuracy in obstacle detection. Summary of the Invention

[0003] The embodiments of the present disclosure at least provide a method, apparatus, computer device, and storage medium for detecting an obstacle.

[0004] In a first aspect, an embodiment of the present disclosure provides a method for detecting an obstacle, comprising: determining a first position point corresponding to an object to be identified in the target scene based on current frame radar scanning data obtained by scanning a target scene; determining whether the object to be identified is a target object to be deleted based on historical position information corresponding to historical candidate deletion objects in historical frame radar scanning data and the first position point; and in response to the object to be identified not being the target object to be deleted, determining the object to be identified as an obstacle in the target scene.

[0005] In this way, by combining the historical position information of the historical candidate deletion objects in the historical frame radar data, it is determined whether the object to be identified in the current frame radar scanning data needs to be deleted, thereby combining the spatial domain and the time domain to comprehensively judge whether the object to be identified is the target object to be deleted, so as to judge whether the object to be identified is an obstacle in the target scene, with higher detection accuracy.

[0006] In an optional embodiment, the determining whether the object to be identified is a target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point includes: judging whether the object to be identified is a current candidate deletion object based on the first position point corresponding to the object to be identified; in response to the object to be identified being the current candidate deletion object, determining whether the object to be identified is a target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point.

[0007] In this way, the first position point can be used to first determine whether the object to be identified is the current candidate deletion object; when it can be determined that the object to be identified is the current candidate deletion object, a more accurate judgment can be made on whether the object to be identified is the target object to be deleted based on the historical position information and the first position point.

[0008] In an optional implementation, in response to the object to be identified not being the current candidate deletion object, the object to be identified is determined to be an obstacle in the target scene.

[0009] In this way, when it is determined that the object to be identified is not the current candidate for deletion, the object to be identified can be directly determined as an obstacle, which is more efficient. Since the detection method provided by the present disclosure can more accurately determine whether the object to be identified is the current candidate for deletion, it is also more accurate in determining whether the object to be identified is an obstacle in the target scene.

[0010] In an optional embodiment, the method of determining the first position point corresponding to the object to be identified in the target scene based on the current frame radar scanning data obtained by scanning the target scene includes: for each object to be identified, determining the point cloud point corresponding to the object to be identified from the current frame radar scanning data; determining the contour information corresponding to the object to be identified based on the three-dimensional position information of the point cloud point corresponding to the object to be identified in the target scene; and determining the first position point corresponding to the object to be identified based on the contour information.

[0011] In this way, since the first position point is obtained using radar scanning data, in addition to retaining the information contained in the original point cloud point, more information about the object to be identified can also be represented; therefore, using the first position point instead of the point cloud point determined by radar scanning requires less calculation, consumes less computing power, and is more efficient.

[0012] In an optional embodiment, the determining of the contour information corresponding to the object to be identified based on the three-dimensional position information of the point cloud points corresponding to the object to be identified in the target scene includes: projecting the point cloud points corresponding to the object to be identified onto a preset plane to obtain a first projection point; and determining the contour information of the object to be identified based on the two-dimensional position information of the first projection point in the preset plane.

[0013] In this way, the method of using the first projection point to determine the contour information of the object to be identified is simpler to calculate using two-dimensional position information than the method of directly determining the contour information using point cloud points, and the contour information of the object to be identified is also more accurate.

[0014] In an optional embodiment, determining the first position point corresponding to the object to be identified based on the contour information includes: using the contour information of the object to be identified to determine the projection area of ​​the object to be identified in a preset plane; and determining the first position point corresponding to the object to be identified based on the area of ​​the projection area.

[0015] In an optional embodiment, based on the area of ​​the projection area, the first position point corresponding to the object to be identified is determined, including: comparing the area of ​​the projection area with a preset area threshold; in response to the area being greater than the area threshold, determining a minimum bounding box of the projection area based on the projection area; based on the first area corresponding to the minimum bounding box and a preset first interval step, determining multiple alternative position points within the first area; and determining the alternative position point located within the projection area as the first position point.

[0016] In this way, for the object to be identified whose area is larger than the area threshold of the projection area, the method of determining its first position point by using the first area determined by the minimum enclosing box corresponding to its projection area can better retain the position points falling in the projection area, so that when the first position point is used in the subsequent judgment of whether the object to be identified is a target object that can be deleted, the accuracy is higher.

[0017] In an optional embodiment, determining the first position point corresponding to the object to be identified based on the area of ​​the projection area also includes: comparing the area of ​​the projection area with a preset area threshold; in response to the area being less than or equal to the area threshold, determining the center point of the projection area; based on the center point and a preset radius length, determining a second area with the center point as the center and the preset radius length as the radius; based on the second area and a preset second interval step, determining the first position point within the second area.

[0018] Since an object to be identified whose projection area is smaller than the area threshold is typically smaller, directly determining the first location point for it may result in a smaller number of location points. Therefore, using this method of determining the first location point, multiple location points related to the object to be identified can be determined. By increasing the number of location points, accuracy can also be improved when subsequently using the first location point to determine whether the object to be identified is a target object that can be deleted.

[0019] In an optional embodiment, the determining whether the object to be identified is the current candidate object for deletion based on the first position point corresponding to the object to be identified includes: obtaining a current frame image obtained by scanning the target scene; projecting the first position point into the current frame image to obtain a second projection point; and determining whether the object to be identified is the current candidate object for deletion based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image.

[0020] In this way, by using the second projection point of the first position point in the current frame image, it is relatively easy to first determine whether the object to be identified can be used as the current candidate for deletion; if it cannot be used as the current candidate for deletion, no further processing or judgment is performed on it. This can further improve the efficiency of detection.

[0021] In an optional embodiment, the determining whether the object to be identified is the current candidate object for deletion based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image includes: predicting an obstacle prediction result corresponding to the second projection point based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image; the obstacle prediction result includes: whether there is an obstacle at the position corresponding to the second projection point, or whether there is no obstacle; and determining whether the object to be identified is the current candidate object for deletion based on the obstacle prediction results corresponding to each of the second projection points.

[0022] In this way, the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image are relatively easy to obtain; and the prediction result of the obstacle corresponding to the second projection point can be gradually fused with the data represented by each second projection point while retaining the position information of the second projection point, so the accuracy of the obstacle prediction result is also higher.

[0023] In an optional embodiment, determining whether the object to be identified is the current candidate for deletion based on the obstacle prediction results corresponding to each second projection point includes: determining the confidence that the object to be identified is an obstacle based on the obstacle prediction results corresponding to each second projection point; and determining whether the object to be identified is the current candidate for deletion based on the confidence and a preset confidence threshold.

[0024] In an optional embodiment, there are n second projection points; determining the confidence level that the object to be identified is an obstacle based on the obstacle prediction results corresponding to each second projection point includes: traversing the 2nd to nth second projection points; for the traversed i-th second projection point, determining a criterion function corresponding to the i-th second projection point based on the obstacle prediction result of the i-th second projection point; wherein i is a positive integer greater than 1; determining a fusion criterion result corresponding to the i-th second projection point based on the criterion function corresponding to the i-th second projection point and the fusion criterion results of the 1st to i-1th second projection points; and obtaining a confidence level that the object to be identified is an obstacle based on the fusion criterion result corresponding to the i-th second projection point.

[0025] In an optional embodiment, the determining whether the object to be identified is the target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point includes: determining the target candidate object from the historical candidate objects based on the historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point; determining whether the object to be identified is the target object to be deleted based on the first projection area of ​​the target candidate object on the preset plane and the second projection area of ​​the object to be identified on the preset plane.

[0026] In this way, the method of using the first projection area and the second projection area to judge whether the object to be identified is the target object to be deleted is relatively simple, and thus the detection efficiency can be improved.

[0027] In an optional embodiment, the determining of the target candidate object from the historical candidate objects based on the historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point includes: determining the distance between each historical candidate object and the object to be identified based on the historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point; and determining the target candidate object that is closest to the object to be identified from the historical candidate objects based on the distance between each historical candidate object and the object to be identified.

[0028] In this way, by using the distances between the historical candidate objects and the object to be identified, the target candidate object can be determined more quickly and accurately, and whether the object to be identified is the target object to be deleted can be further determined based on the determined target candidate object.

[0029] In an optional embodiment, determining whether the object to be identified is a target object to be deleted based on the first projection area of ​​the target candidate object on a preset plane and the second projection area of ​​the object to be identified on the preset plane includes: determining whether there is an overlapping area between the first projection area and the second projection area; and in response to the fact that there is no overlapping area between the first projection area and the second projection area, determining that the object to be identified is a target object to be deleted.

[0030] In an optional embodiment, the determining whether the object to be identified is the target object to be deleted based on the first projection area of ​​the target candidate object on the preset plane and the second projection area of ​​the object to be identified on the preset plane also includes: in response to the existence of an overlapping area between the first projection area and the second projection area in the current frame radar scanning data, taking the object to be identified as a new historical candidate object, until there is an overlapping area between the first projection area and the second projection area in N consecutive frames of radar scanning data after the current frame radar scanning data, it is determined that the candidate deletion object is not the target object to be deleted, and the new historical candidate object is deleted; N is a positive integer.

[0031] This allows for faster and more efficient determination of whether an object to be identified is a target object to be deleted from a limited number of frames of scanned data. This makes it suitable for scenarios requiring a prompt and accurate response to obstacles. Furthermore, obstacle detection can be made more timely, further ensuring driving safety.

[0032] In an optional implementation, it also includes: for each of the historical candidate deletion objects, detecting the storage time of the historical candidate deletion object and the time difference between the current time; if the time difference is greater than or equal to a preset time difference threshold, deleting the historical candidate deletion object.

[0033] In this way, it can also ensure that the historical candidate deletion objects are used to accurately filter out the objects to be deleted and identified, and to a certain extent reduce the data that needs to be stored. At the same time, it also reduces the matching calculation between the historical candidate deletion objects and the current objects to be identified, thereby improving the detection efficiency of obstacles.

[0034] In a second aspect, an embodiment of the present disclosure further provides an obstacle detection device, comprising:

[0035] A first determining module is configured to determine a first position point corresponding to an object to be identified in the target scene based on current frame radar scanning data obtained by scanning the target scene;

[0036] A second determination module is configured to determine whether the object to be identified is a target object to be deleted based on historical position information corresponding to historical candidate deletion objects in the historical frame radar scanning data and the first position point;

[0037] The third determining module is configured to determine the object to be identified as an obstacle in the target scene in response to the object to be identified not being the target object to be deleted.

[0038] In an optional embodiment, when the second determination module determines whether the object to be identified is a target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point, it is used to: judge whether the object to be identified is a current candidate deletion object based on the first position point corresponding to the object to be identified; in response to the object to be identified being the current candidate deletion object, determine whether the object to be identified is a target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point.

[0039] In an optional implementation, the second determination module is further configured to: in response to the object to be identified not being the current candidate deletion object, determine the object to be identified as an obstacle in the target scene.

[0040] In an optional embodiment, when the first determination module determines the first position point corresponding to the object to be identified in the target scene based on the current frame radar scanning data obtained by scanning the target scene, it is used to: determine, for each object to be identified, the point cloud point corresponding to the object to be identified from the current frame radar scanning data; determine the contour information corresponding to the object to be identified based on the three-dimensional position information of the point cloud point corresponding to the object to be identified in the target scene; and determine the first position point corresponding to the object to be identified based on the contour information.

[0041] In an optional embodiment, when determining the contour information corresponding to the object to be identified based on the three-dimensional position information of the point cloud points corresponding to the object to be identified in the target scene, the first determination module is used to: project the point cloud points corresponding to the object to be identified onto a preset plane to obtain a first projection point; and determine the contour information of the object to be identified based on the two-dimensional position information of the first projection point in the preset plane.

[0042] In an optional embodiment, when the first determination module determines the first position point corresponding to the object to be identified based on the contour information, it is used to: use the contour information of the object to be identified to determine the projection area of ​​the object to be identified in a preset plane; and determine the first position point corresponding to the object to be identified based on the area of ​​the projection area.

[0043] In an optional embodiment, when the first determination module determines the first position point corresponding to the object to be identified based on the area of ​​the projection area, it is used to: compare the area of ​​the projection area with a preset area threshold; in response to the area being greater than the area threshold, determine the minimum enclosing box of the projection area based on the projection area; based on the first area corresponding to the minimum enclosing box and a preset first interval step, determine multiple alternative position points within the first area; and determine the alternative position point located within the projection area as the first position point.

[0044] In an optional embodiment, when the first determination module determines the first position point corresponding to the object to be identified based on the area of ​​the projection area, it is also used to: compare the area of ​​the projection area with a preset area threshold; in response to the area being less than or equal to the area threshold, determine the center point of the projection area; based on the center point and a preset radius length, determine a second area with the center point as the center and the preset radius length as the radius; based on the second area and a preset second interval step, determine the first position point within the second area.

[0045] In an optional embodiment, when the second determination module determines whether the object to be identified is the current candidate object for deletion based on the first position point corresponding to the object to be identified, it is used to: obtain the current frame image obtained by capturing the target scene; project the first position point into the current frame image to obtain a second projection point; and determine whether the object to be identified is the current candidate object for deletion based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image.

[0046] In an optional embodiment, when the second determination module determines whether the object to be identified is the current candidate object for deletion based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image, it is used to: predict an obstacle prediction result corresponding to the second projection point based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image; the obstacle prediction result includes: whether there is an obstacle at the position corresponding to the second projection point, or whether there is no obstacle; based on the obstacle prediction results corresponding to each second projection point, determine whether the object to be identified is the current candidate object for deletion.

[0047] In an optional embodiment, when the second determination module determines whether the object to be identified is the current candidate for deletion based on the obstacle prediction results corresponding to each second projection point, it is used to: determine the confidence level that the object to be identified is an obstacle based on the obstacle prediction results corresponding to each second projection point; and determine whether the object to be identified is the current candidate for deletion based on the confidence level and a preset confidence threshold.

[0048] In an optional embodiment, there are n second projection points; when the second determination module determines the confidence level that the object to be identified is an obstacle based on the obstacle prediction results corresponding to each second projection point, it is used to: traverse the 2nd to nth second projection points; for the traversed i-th second projection point, determine the criterion function corresponding to the i-th second projection point based on the obstacle prediction result of the i-th second projection point; wherein i is a positive integer greater than 1; based on the criterion function corresponding to the i-th second projection point and the fusion criterion results of the 1st to i-1th second projection points, determine the fusion criterion result corresponding to the i-th second projection point; and based on the fusion criterion result corresponding to the i-th second projection point, obtain the confidence level that the object to be identified is an obstacle.

[0049] In an optional embodiment, when the second determination module determines whether the object to be identified is the target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point, it is used to: determine the target candidate object from the historical candidate objects based on the historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point; determine whether the object to be identified is the target object to be deleted based on the first projection area of ​​the target candidate object on the preset plane and the second projection area of ​​the object to be identified on the preset plane.

[0050] In an optional embodiment, when the second determination module determines the target candidate object from the historical candidate objects based on the historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point, it is used to: determine the distance between each historical candidate object and the object to be identified based on the historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point; and determine, from the historical candidate objects, the historical candidate object that is closest to the object to be identified as the target candidate object based on the distance between each historical candidate object and the object to be identified.

[0051] In an optional embodiment, when the second determination module determines whether the object to be identified is a target object to be deleted based on the first projection area of ​​the target candidate object on the preset plane and the second projection area of ​​the object to be identified on the preset plane, it is used to: determine whether there is an overlapping area between the first projection area and the second projection area; in response to the fact that there is no overlapping area between the first projection area and the second projection area, determine that the object to be identified is a target object to be deleted.

[0052] In an optional embodiment, the second determination module, when determining whether the object to be identified is the target object to be deleted based on the first projection area of ​​the target candidate object on the preset plane and the second projection area of ​​the object to be identified on the preset plane, is also used to: in response to the existence of an overlapping area between the first projection area and the second projection area in the current frame radar scanning data, take the object to be identified as a new historical candidate object, until there is an overlapping area between the first projection area and the second projection area in N consecutive frames of radar scanning data after the current frame radar scanning data, it is determined that the candidate deletion object is not the target object to be deleted, and the new historical candidate object is deleted; N is a positive integer.

[0053] In an optional embodiment, the detection device also includes: a processing module, which is used to detect the storage time of each historical candidate deletion object and the time difference between the storage time of the historical candidate deletion object and the current time; if the time difference is greater than or equal to a preset time difference threshold, the historical candidate deletion object is deleted.

[0054] In a third aspect, an optional implementation of the present disclosure further provides a computer device, a processor, and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the processor is used to execute the machine-readable instructions stored in the memory. When the machine-readable instructions are executed by the processor, the machine-readable instructions perform the steps of the above-mentioned first aspect, or any possible implementation of the first aspect.

[0055] In a fourth aspect, an optional implementation of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, it executes the steps of the above-mentioned first aspect or any possible implementation of the first aspect.

[0056] For a description of the effects of the above-mentioned obstacle detection device, computer equipment, and computer-readable storage medium, please refer to the description of the above-mentioned obstacle detection method, which will not be repeated here.

[0057] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following briefly introduces the drawings required for use in the embodiments. The drawings herein are incorporated into and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and, together with the specification, are used to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without inventive effort.

[0059] Figure 1 A flowchart of an obstacle detection method provided by an embodiment of the present disclosure is shown;

[0060] Figure 2 A schematic diagram of determining a first position point provided by an embodiment of the present disclosure is shown;

[0061] Figure 3 A schematic diagram showing another method for determining a first position point provided by an embodiment of the present disclosure is shown;

[0062] Figure 4 A schematic diagram showing a projection area provided by an embodiment of the present disclosure;

[0063] Figure 5 A specific flow chart of detecting an object to be identified provided by an embodiment of the present disclosure is shown;

[0064] Figure 6 A schematic diagram of an obstacle detection device provided by an embodiment of the present disclosure is shown;

[0065] Figure 7 A schematic diagram of a computer device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solutions and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure is not intended to limit the scope of the present disclosure for protection, but merely represents the selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present disclosure.

[0067] Research has found that lidar can be used to scan a vehicle's driving area to identify potential obstacles. Specifically, lidar emits laser light into the driving area and receives the reflected laser light, determining whether there are obstacles based on the received laser light. This can easily lead to an inability to properly detect obstacles if the laser light reflected back after scanning an object is abnormal. For example, if there is water on the road or wet paint on lane markings, the laser light emitted by the lidar will not be reflected properly due to mirror reflection, thus misjudging these non-obstacle objects as obstacles, resulting in low accuracy in obstacle detection using laser light.

[0068] Based on the above research, the present disclosure provides an obstacle detection method, which determines whether to delete the object to be identified in the current frame radar scanning data by combining the historical position information of the historical candidate deletion objects in the historical frame radar data, thereby combining the spatial domain and the time domain to comprehensively judge whether the object to be identified is the target object to be deleted, so as to judge whether the object to be identified is an obstacle in the target scene, with higher detection accuracy.

[0069] The defects in the above solutions are the results obtained by the inventors after practice and careful research. Therefore, the process of discovering the above problems and the solutions proposed by this disclosure for the above problems below should be the contributions made by the inventors to this disclosure during the disclosure process.

[0070] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0071] To facilitate understanding of this embodiment, a detailed introduction to an obstacle detection method disclosed in an embodiment of the present disclosure is first provided. The obstacle detection method provided in this embodiment of the present disclosure is generally executed by a computer device with certain computing capabilities. This computer device may include, for example, a terminal device, a server, or other processing device. The terminal device may be a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, an in-vehicle device, a wearable device, or the like. In some possible implementations, the obstacle detection method may be implemented by a processor invoking computer-readable instructions stored in a memory.

[0072] The obstacle detection method provided by the embodiment of the present disclosure is described below.

[0073] See also Figure 1 FIG. 1 is a flow chart of an obstacle detection method provided by an embodiment of the present disclosure, wherein the method includes steps S101 to S103, wherein:

[0074] S101: Determine a first position point corresponding to an object to be identified in the target scene based on current frame radar scanning data obtained by scanning the target scene;

[0075] S102: Determining whether the object to be identified is a target object to be deleted based on historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point;

[0076] S103: In response to the object to be identified not being the target object to be deleted, determining the object to be identified as an obstacle in the target scene.

[0077] The disclosed embodiment uses radar scan data from the current frame of the target scene to determine the first location point corresponding to the object to be identified, as well as the historical location information corresponding to historical candidate deletion objects in the historical frame scan data, to determine whether the object to be identified is the target object to be deleted. If the object to be identified is not the target object to be deleted, the object to be identified is determined to be an obstacle in the target scene. This method combines the spatial and temporal domains to comprehensively determine whether the object to be identified is the target object to be deleted, thereby determining whether the object to be identified is an obstacle in the target scene, achieving higher detection accuracy.

[0078] The above S101 to S103 are described in detail below.

[0079] With respect to S101 above, the obstacle detection method provided by the embodiments of the present disclosure can be applied to different scenarios. For example, in an autonomous driving scenario, the target scene may include, for example, the space in which the autonomous driving vehicle is traveling, which may include, for example, other traveling vehicles, lane markings, signboards, green belts, etc. In an intelligent warehousing scenario, the target scene may include, for example, the space in which an intelligent robot is traveling, which may include, for example, other robots, staff, shelves, cargo boxes, positioning markers, etc.

[0080] The obstacle detection method provided by the embodiment of the present disclosure is described below using an autonomous driving scenario as an example.

[0081] In an autonomous driving scenario, a lidar (lidar) radar may be installed on the vehicle to scan and detect the area in which the vehicle is traveling. As the vehicle moves, the lidar radar may scan the target scene at 0.2-second intervals, generating radar scan data. Here, the radar scan data with the closest scan time to the current moment is used as the current frame of radar scan data for the target scene.

[0082] After obtaining the current frame radar scan data, the current frame radar scan data can be used to determine multiple objects identified in the target scene. Exemplarily, the current frame radar scan data can be processed using object detection. Since radar scan data can reflect the size, shape, etc. of an object, the object detection method can be used to identify objects in some target scenes, such as vehicles and signboards; these objects can be directly identified as obstacles because they are objects that need to be avoided. At the same time, there may be objects that cannot be identified using the object detection method. After detecting them using the object recognition method, these unidentifiable objects are treated as objects to be identified in the current frame radar scan data, that is, objects with a classification label of "unknown objects".

[0083] When determining whether the object to be identified is a target object to be deleted, for example, the first position point determined for the object to be identified by the radar scanning data of the current frame can be used for implementation.

[0084] Here, the first position point is used as a criterion or basis for determining whether the object to be identified can be a candidate for deletion. A candidate for deletion is an object that may be an obstacle. The subsequent process of determining the target object further determines whether the candidate object is indeed an obstacle. If the candidate object is the target object, the corresponding object to be identified is not an obstacle, but rather a false detection.

[0085] Specifically, when using the current frame radar scanning data to determine the first position point corresponding to the object to be identified in the target scene, the following method can be adopted: for each object to be identified, determine the point cloud point corresponding to the object to be identified from the current frame radar scanning data; based on the three-dimensional position information of the point cloud point corresponding to the object to be identified in the target scene, determine the contour information corresponding to the object to be identified; based on the contour information, determine the first position point corresponding to the object to be identified.

[0086] Taking any one of the identified objects as an example, when the object to be identified is determined in the current frame radar scanning data by object recognition, the point cloud points corresponding to the object to be identified can be determined using the current frame radar scanning data.

[0087] Here, since the number of point cloud points corresponding to the object to be identified may be large, and the point cloud points use a large amount of three-dimensional position information data when representing the position of the object to be identified, for the object to be identified, the method of re-modeling the object to be identified using the point cloud points corresponding to the object to be identified and then determining whether it is an obstacle, or the method of re-encoding the point cloud points corresponding to the object to be identified using a neural network and then performing a similarity judgment with the point cloud points of a specific object determined historically, is computationally intensive, requires a lot of computing power, and is also inefficient.

[0088] Therefore, after determining the point cloud points corresponding to the object to be identified, the contour information corresponding to the object to be identified can be determined based on the three-dimensional position information of the point cloud points corresponding to the object to be identified in the target scene, and then the first position point corresponding to the object to be identified can be determined using the contour information, thereby using the first position point with less data to determine whether the object to be identified is the target object to be deleted.

[0089] Specifically, when using the three-dimensional position information of the point cloud points corresponding to the object to be identified in the target scene to determine the contour information corresponding to the object to be identified, for example, the following method can be adopted: project the point cloud points corresponding to the object to be identified into a preset plane to obtain a first projection point; based on the two-dimensional position information of the first projection point in the preset plane, determine the contour information of the object to be identified.

[0090] The preset plane can be, for example, the plane of the ground on which the autonomous vehicle is traveling. By projecting the point cloud points corresponding to the object to be identified onto the preset plane, a first projection point can be obtained. At this point, the three-dimensional position information corresponding to the point cloud points is converted into two-dimensional position information corresponding to the first projection point, reducing the amount of data. Using the two-dimensional position information of the first projection point in the preset plane, projection points located at the edge of the first projection point can be determined. Using these edge projection points and their corresponding two-dimensional position information, the contour information corresponding to the object to be identified can be determined.

[0091] When using contour information to determine the first position point corresponding to the object to be identified, for example, the contour information corresponding to the object to be identified can be used to determine the projection area to be identified in the preset plane; then, based on the area of ​​the projection area, the first position point corresponding to the object to be identified is determined.

[0092] Specifically, using the contour information corresponding to the object to be identified, multiple corner points corresponding to the contour surrounding the object to be identified can be determined; using the two-dimensional position information of multiple corner points, the area occupied by the projection area to be identified on the preset plane can be determined, thereby determining the first position point corresponding to the object to be identified.

[0093] In a specific implementation, when determining the first position point corresponding to the object to be identified based on the area of ​​the projection area, for example, the area of ​​the projection area can be compared with a preset area threshold, and the first position point corresponding to the object to be identified can be determined based on the comparison result.

[0094] The preset area threshold may include, for example, 0.3 square meters, 0.5 square meters, etc. The preset area threshold may be determined, for example, based on experience or actual conditions, and will not be described in detail here.

[0095] Taking the preset area threshold of 0.3 square meters as an example, when it is determined that the area of ​​the projection area is greater than the area threshold, the minimum enclosing box of the projection area is determined based on the projection area; based on the first area corresponding to the minimum enclosing box and the preset first interval step, multiple alternative position points are determined within the first area; and the alternative position point located within the projection area is determined as the first position point.

[0096] See also Figure 2 FIG. 1 is a schematic diagram of determining a first position point according to an embodiment of the present disclosure. Figure 2(a) shows a projection area 21 determined for the object to be identified. This projection area is, for example, an irregular polygon. Using this projection area 21, a corresponding minimum bounding box 22 can be determined. In one possible embodiment, this minimum bounding box 22 comprises a rectangle. After determining the minimum bounding box 22, the first area 23 occupied by the minimum bounding box 22 can be determined.

[0097] Then, using the preset first interval step, for example, multiple candidate position points 24 can be determined in the first area 23. Here, the preset first interval step can include, for example, 0.2 meters. Figure 2 As shown in (b), it shows that a plurality of alternative position points 24 are determined in the first area 23 using a preset first interval step; wherein, for position points corresponding to the alternative position points 24 in the four directions of up, down, left and right, respectively, the interval between them and the alternative position points 24 corresponding to the four directions of up, down, left and right is the preset first interval step.

[0098] After determining multiple candidate position points 24, the projection area 21 can be used to select the candidate position point 24 corresponding to the object to be identified as the first position point 25. Figure 2 As shown in (c), using Figure 2 The multiple candidate location points 24 determined in (b), and Figure 2 The projection area 21 determined in (a) can be framed to select a candidate position point 24 located within the projection area 21 as the first position point 25 .

[0099] Here, since the projection area 21 is usually an irregular shape, when the projection area 21 is used to determine the first position point 25, an alternative position point 24 may appear on the boundary of the projection area 21, for example, Figure 2 Points a and b are shown in (c).

[0100] In view of this situation, in a possible implementation, in order to retain the number of position points to a maximum extent, the candidate position points 24 corresponding to point a and point b respectively may both be used as the first position point 25 .

[0101] In another possible embodiment, since the projection area of ​​the object to be identified is relatively large, a sufficient number of first position points can be determined. To reduce the amount of data processed and improve processing accuracy, the candidate position points 24 corresponding to points a and b can be selectively screened. For example, a larger portion of the candidate position point 24 corresponding to point a falls within the projection area 21 than the candidate position point 24 corresponding to point b; therefore, the candidate position point 24 corresponding to point a is retained in the first position points 25, while the candidate position point 24 corresponding to point b is screened out.

[0102] The above two implementation processes can be determined according to actual conditions and are not limited here.

[0103] In addition, when it is determined that the area of ​​the projection area is less than or equal to the area threshold, the center point of the projection area can also be determined; based on the center point and the preset radius length, a second area with the center point as the center and the preset radius length as the radius is determined; based on the second area and the preset second interval step, the first position point is determined within the second area.

[0104] See also Figure 3 FIG. 1 is a schematic diagram of another embodiment of the present disclosure for determining a first position point. Figure 3 (a) shows a projection area 31 (represented by the area outlined by the dashed lines in the figure) determined for the object to be identified. This projection area is, for example, an irregular polygon. Using this projection area 31, the center point 32 of the projection area can be determined. Then, using the center point 32 and a preset radius length, for example, a second area 33 (represented by the area outlined by the solid lines in the figure) can be determined. Specifically, for example, a circular second area 33 can be determined with the center point 32 as the center and a preset radius length as the radius. The preset radius length can be, for example, 0.2 meters, 0.3 meters, etc.

[0105] In addition, in a possible implementation, for example, when the determined center point 32 radiates outward to the boundary of the projection area 31, the determined maximum length can be used as the preset radius length. Figure 3 (a) shows the maximum length r from the center point 32 to the edge of the projection area 31. This maximum length r is used as the preset radius to determine the second area 33. In this way, for an object to be identified with a small projection area, a large number of first position points can be determined, and a sufficient number of first position points can be used to further determine whether the object to be identified is a target object to be deleted.

[0106] Then, using the preset second interval step, for example, multiple position points can be determined within the second area 33. In order to determine more position points for the object to be identified to a greater extent, position points that fall on the edge of the second area 33 are retained when determining the position points. Here, the preset second interval step can be, for example, the same as the preset first interval step, such as 0.2 meters; or a value different from the preset first interval step, such as 0.15 meters, can be determined based on actual conditions. The specific method can be determined based on actual conditions and will not be described in detail here. Figure 3 As shown in (b), a plurality of position points can be determined in the second area 33 using the preset second interval step.

[0107] Here, since the projection area 31 corresponding to the object to be identified is relatively small, in order to ensure that there are a sufficient number of position points for the object to be identified, all the position points within the second area can be determined as the first position points 34. Figure 3 As shown in (b), the number of the obtained first position points 34 is greater than the number of position points falling into the projection area 31.

[0108] Regarding the above S102, after determining the first position point corresponding to the object to be identified in the target scene, it is also possible to determine whether the object to be identified is the target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data.

[0109] In a specific implementation, for example, the following method can be adopted: based on the first position point corresponding to the object to be identified, determine whether the object to be identified is the current candidate deletion object; in response to the object to be identified being the current candidate deletion object, determine whether the object to be identified is the target object to be deleted based on the historical position information corresponding to the historical candidate deletion objects in the historical frame radar scanning data, and the first position point.

[0110] Specifically, when judging whether the object to be identified is the current candidate for deletion based on the first position point of the object to be identified, the following method can be specifically adopted: obtain the current frame image obtained by scanning the target scene; project the first position point into the current frame image to obtain a second projection point; based on the position information of the second projection point in the current frame image, and the position of the obstacle included in the current frame image in the current frame image, determine whether the object to be identified is the current candidate for deletion.

[0111] The current frame image may be acquired, for example, using an image acquisition device mounted on an autonomous vehicle. Specifically, the image acquisition device may include a color camera. Upon acquiring the current frame image, the first position point may be projected onto the current frame image to obtain a second projection point. While projecting the second projection point, position information of the obtained second projection point may also be determined.

[0112] Here, since the current frame image can be used to determine the drivable area (freespace) of the autonomous driving vehicle, the second projection point obtained by projecting the first position point onto the current frame image can be used to further determine whether the object to be identified is an object in the drivable area, and then determine whether the object to be identified needs to be deleted.

[0113] When determining whether the object to be identified is the current candidate deletion object based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image, for example, the following method can be used:

[0114] Based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image, an obstacle prediction result corresponding to the second projection point is predicted; the obstacle prediction result includes: whether there is an obstacle at the position corresponding to the second projection point or no obstacle exists; based on the obstacle prediction results corresponding to each of the second projection points, it is determined whether the object to be identified is the current candidate deletion object.

[0115] Specifically, when predicting the obstacle prediction result corresponding to the second projection point, it can be determined, for example, based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image.

[0116] In one possible implementation, when the second projection point is located at the position where an obstacle is included in the current frame image, since it can be considered with a higher degree of confidence that the second projection point indicates the presence of an obstacle at the corresponding position, the corresponding obstacle prediction result is determined as follows: an obstacle exists at the position corresponding to the second projection point. In another possible implementation, when the second projection point is not located at the position where the obstacle is included in the current frame image, since it can be considered with a higher degree of confidence that the second projection point indicates the absence of an obstacle at the corresponding position, the corresponding obstacle prediction result is determined as follows: no obstacle exists at the position corresponding to the second projection point.

[0117] In this way, for multiple second projection points, the obstacle prediction result corresponding to each second projection point can be determined.

[0118] When determining whether the object to be identified is the current candidate for deletion using the obstacle prediction results corresponding to each of the second projection points, for example, the confidence level that the object to be identified is an obstacle can be determined based on the obstacle prediction results corresponding to each of the second projection points; and based on the confidence level and a preset confidence threshold, it is determined whether the object to be identified is the current candidate for deletion.

[0119] In a specific implementation, the determined second projection points may include, for example, n; wherein n is a positive integer. When determining the confidence level that the object to be identified is an obstacle based on the obstacle prediction results corresponding to the second projection points, for example, the following method may be used:

[0120] Traversing the 2nd to nth second projection points; for the traversed i-th second projection point, determining the criterion function corresponding to the i-th second projection point based on the obstacle prediction result of the i-th second projection point; wherein i is a positive integer greater than 1; based on the criterion function corresponding to the i-th second projection point and the fusion criterion results of the 1st to i-1th second projection points, determining the fusion criterion result corresponding to the i-th second projection point; based on the fusion criterion result corresponding to the i-th second projection point, obtaining the confidence level for determining that the object to be identified is an obstacle.

[0121] For example, for the first second projection point, the corresponding obstacle prediction result may include, for example, whether an obstacle exists at the location corresponding to the second projection point, or whether an obstacle does not exist. Different obstacle prediction results may result in different criterion functions. For example, if the obstacle prediction result includes the presence of an obstacle at the location corresponding to the second projection point, the corresponding criterion function M_1(·) may, for example, represent the likelihood function (mass) of determining the confidence level of the hypothesis that an obstacle exists at the second projection point. If the obstacle prediction result includes the absence of an obstacle at the location corresponding to the second projection point, the corresponding criterion function M_2(·) may, for example, represent the likelihood function of determining the confidence level of the hypothesis that an obstacle does not exist at the second projection point.

[0122] Here, for the sake of convenience, for the second projection point, its corresponding criterion function is represented by M2(·), and the second projection point corresponding to different obstacle prediction results can be specifically the above-mentioned M_1(·) and M_2(·).

[0123] When traversing the 2nd to nth second projection points, the fusion criterion results of the 1st to (i-1)th second projection points may also be determined to determine the fusion criterion result corresponding to the i-th second projection point.

[0124] Taking the second second projection point as an example, when determining the fusion criterion result corresponding to the second second projection point, for example, the following formula (1) can be used:

[0125]

[0126] Where a1 represents the first second projection point, and a2 represents the second second projection point. Here, the criterion function corresponding to a2 can be expressed as M2(a2). Since the previous second projection points only include a1 when traversing to a2, the fusion criterion result M1(·) can be directly expressed as M1(a1); alternatively, it can be expressed as M2(a1).

[0127] In addition, K represents a normalization coefficient, which satisfies the following formula (2):

[0128]

[0129] Using formula (1), we can determine the fusion criterion result M of a1 and a2 when traversing to the second projection point a2: 2 12 (a1; a1). Here, the superscript i of M indicates traversal to the i-th second projection point.

[0130] Then, the third second projection point a3 is traversed. Similarly, the criterion function corresponding to a3 can be expressed as M2(a3). At this time, the fusion criterion result corresponding to the third second projection point determined when traversing to a3 can be determined to satisfy the following formula (3):

[0131]

[0132] The normalization coefficient K satisfies the following formula (4):

[0133]

[0134] …

[0135] The above process is continued in this way to continue traversing other second projection points, determine the criterion function corresponding to the second projection point, and determine the fusion criterion results of the second projection points traversed from the first to the currently traversed point, until all the second projection points are traversed to obtain the fusion criterion result corresponding to the nth second projection point, thereby obtaining the confidence level for determining that the object to be identified is an obstacle.

[0136] The confidence level that the object to be identified is an obstacle may include, for example, a probability value, such as 0.61, 0.70, or 0.86.

[0137] After determining the confidence level that the object to be identified is an obstacle, a preset confidence threshold may be used to determine whether the object to be identified is a current candidate for deletion.

[0138] The preset confidence threshold may also include a probability value, such as 0.75. Specifically, when determining the preset confidence threshold, it can be determined based on experience or multiple experiments. The specific method for determining the preset confidence threshold is not described in detail here.

[0139] Furthermore, if the confidence level that the object to be identified is an obstacle is numerically greater than the preset confidence level threshold, it is considered that the object to be identified can be determined as the current candidate object to be deleted.

[0140] For example, when the confidence level for determining that the object to be identified is an obstacle is 0.6 and the preset confidence threshold is 0.55, the object to be identified is determined not to be a current candidate for deletion; and when the confidence level for determining that the object to be identified is an obstacle is 0.80, the object to be identified is determined to be a current candidate for deletion.

[0141] After determining the current candidate deletion object, it is also possible to determine whether the object to be identified is a target object to be deleted based on historical position information corresponding to the historical candidate deletion objects in the historical frame radar scanning data and the first position point.

[0142] In a specific implementation, the following method can be used to determine whether the object to be identified is a target object to be deleted: based on the historical position information corresponding to the historical candidate object in the historical frame radar scanning data, and the first position point, determine the target candidate object from the historical candidate objects; based on the first projection area of ​​the target candidate object on the preset plane, and the second projection area of ​​the object to be identified on the preset plane, determine whether the object to be identified is a target object to be deleted.

[0143] Specifically, when the radar scans the target scene, for example, it also includes obtaining historical radar scan data determined before the current frame of radar scan data. Similarly, using this historical radar scan data, the corresponding historical candidate deletion object can be determined, along with its historical location information. The details are not repeated here.

[0144] After determining the historical location information corresponding to the candidate historical objects, the first location point can be used to determine the distance between each candidate historical object and the object to be identified. The historical location information used to determine the distance can, for example, be the coordinates of the center point of the candidate historical object. Alternatively, the location of the object to be identified can be represented by determining a first location point within the first location point that represents the center point of the object to be identified. The determined historical location information and the first location point can then be used to determine the distance between each candidate historical object and the object to be identified.

[0145] In this way, the target candidate object closest to the object to be identified can be determined from the historical candidate objects based on the distances between each historical candidate object and the object to be identified. In one possible implementation, the distance with the smallest value can be determined from the multiple determined distances, and the corresponding historical candidate object can be used as the target candidate object; alternatively, the historical candidate objects corresponding to multiple distances with smaller values ​​can be used as the target candidate object. The specific method for determining the target candidate object can be determined based on actual circumstances and will not be elaborated here.

[0146] After determining the candidate target object, a first projection area of ​​the candidate target object within a preset plane and a second projection area of ​​the object to be identified within the preset plane may also be determined. Here, the preset plane may include, for example, the plane of the road on which the autonomous driving vehicle is traveling.

[0147] For example, see Figure 4 FIG. 4 is a schematic diagram of a projection area provided by an embodiment of the present disclosure. It includes a first projection area 41 of a target candidate object on a preset plane, and a second projection area 42 of an object to be identified on a preset plane (for ease of distinction, the boundary of the first projection area 41 is shown in dotted lines). Figure 4 In the case shown in (a), the first projection area 41 and the second projection area 42 have an overlapping area; Figure 4 In the case shown in (b), there is no overlapping area between the first projection area 41 and the second projection area 42.

[0148] Here, for the case where there is no overlapping area between the first projection area and the second projection area, it can be considered that there is no target candidate object corresponding to the object to be identified in the most recent frame of historical frame scanning data, indicating that the object to be identified is a newly appeared object different from the target candidate object and has not appeared in the recent period of time; therefore, when detecting the object to be identified determined in the current frame scanning data, the object to be identified can be judged as a target object to be deleted in the current frame.

[0149] Furthermore, since the object to be identified is newly present, it can also be treated as a new historical candidate object, and its location information can be saved. Specifically, this location information can be used to perform obstacle detection on the next frame of radar scan data. The method for performing obstacle detection on the next frame of radar scan data using the location information of the object to be identified is similar to the aforementioned method for performing obstacle detection using the location information corresponding to the target candidate object in the historical frame scan data and the object to be identified in the current frame scan data, and will not be further described here.

[0150] Here, since the object to be identified is judged as the target object to be deleted only for the current frame; if the object to be identified appears in the same area on the preset plane in multiple consecutive frames of radar scanning data, it can be determined that the object to be identified is not the target object to be deleted after determining the multiple frames of radar scanning data.

[0151] Specifically, in response to the existence of an overlapping area between the first projection area and the second projection area in the current frame radar scanning data, the object to be identified is taken as a new historical candidate object, until there is an overlapping area between the first projection area and the second projection area in N consecutive frames of radar scanning data after the current frame radar scanning data, then the candidate deletion object is determined to be the target object to be deleted, and the new historical candidate object is deleted; N is a positive integer.

[0152] The value of N may include, for example, 5, 6, 8, 10, etc. Specifically, it may be determined based on the shooting interval when acquiring radar scanning data, or based on experiments, etc., and the details are not repeated here. The following description takes N as an example, where it is set to 10.

[0153] In one possible scenario, if it is determined that the first and second projection areas overlap in N consecutive frames of historical radar scan data, the object to be identified can be determined to be an object that actually exists in the target scene. For example, if a warning cone is placed on the road surface in the target scene, after obtaining the first frame of radar scan data containing the warning cone, since the warning cone actually exists, the warning cone is included in multiple subsequent frames of radar scan data. For example, after 15 frames of radar scan data, the warning cone is no longer captured in the radar scan data. In other words, if the object to be identified is an object that actually exists in the target scene, then multiple consecutive frames of radar scan data will contain the object to be identified. Since objects that actually exist in the target scene may interfere with the autonomous vehicle's operation, they need to be treated as non-target objects to be deleted. Furthermore, since the object to be identified can be determined to be non-target objects to be deleted, there is no need to continue detecting it, so the object to be identified as a new historical candidate can be directly deleted.

[0154] In another possible scenario, if the first projection area and the second projection area do not overlap in all N consecutive frames of historical radar scan data for the object to be identified, for example, if the object to be identified appears and is only included in no more than N consecutive frames of radar scan data, then the object to be identified may be considered a target object to be deleted.

[0155] For example, when the object to be identified includes water, due to the influence of mirror reflection and reflection angle, when the autonomous driving vehicle continuously acquires multiple frames of scanning data during driving, it is possible that multiple frames of continuous radar scanning data contain the water. However, after multiple frames of radar scanning data, for example, after three frames of radar scanning data, due to changes in angle, distance, etc. when scanning the object to be identified, the radar will no longer detect the water through scanning. Therefore, the radar scanning data of the object to be identified will no longer collect radar scanning data of the object to be identified at a position close to the object to be identified on the preset plane. In this case, the object to be identified can be determined to be the target object to be deleted.

[0156] In another embodiment of the present disclosure, since the radar scan data obtained by the autonomous vehicle is constantly changing during its driving process, if a historical candidate for deletion object has not appeared for a period of time, it can be assumed that the historical candidate for deletion object will not appear again during continued driving.

[0157] In a specific implementation, for each historical candidate deletion object, the storage time of the historical candidate deletion object and the time difference with the current time can be detected; if the time difference is greater than or equal to a preset time difference threshold, the historical candidate deletion object is deleted.

[0158] The preset time difference threshold may include, for example, 3 seconds, 4 seconds, etc.; it may be specifically determined based on actual conditions or experiments, and will not be elaborated here.

[0159] In this way, it can also ensure that the objects to be deleted are gradually screened out using historical candidate deletion objects, and to a certain extent reduce the data that needs to be stored. At the same time, it also reduces the matching calculation between historical candidate deletion objects and current objects to be identified, thereby improving the detection efficiency of obstacles.

[0160] With respect to the above S103, when it is determined according to the above S102 that the object to be identified is not the target object to be deleted, the object to be identified can be determined as an obstacle in the target scene. At this time, for example, the autonomous driving vehicle can be controlled to perform an evasive action. Since the obstacle detection method provided by the embodiment of the present disclosure has a high detection accuracy when detecting and judging whether the object to be identified is the target object to be deleted, when the obstacle detection method provided by the present disclosure is applied to the autonomous driving of the autonomous driving vehicle, it is possible to more accurately judge the obstacles that actually need to be avoided in the driving area and complete effective obstacle avoidance. In this way, the person driving the autonomous driving vehicle can effectively reduce the sudden turns or sudden brakes caused by the identification of non-obstacles during the process of riding the autonomous driving vehicle, and the riding experience is better.

[0161] In another embodiment of the present disclosure, a specific embodiment for detecting objects to be identified in a target scene is also provided. Figure 5 As shown, it is a specific flow chart of detecting an object to be identified provided by an embodiment of the present disclosure; wherein,

[0162] S501: Determine a first position point corresponding to an object to be identified based on current frame radar scanning data obtained by scanning a target scene.

[0163] S502: Based on the first position point, determine whether the object to be identified is a current candidate deletion object; if so, jump to step S503; if not, jump to step S507.

[0164] S503: Determine a target candidate object from the historical candidate objects based on the historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point.

[0165] S504: Determine whether there is an overlapping area between the first projection area of ​​the candidate target object on the preset plane and the second projection area of ​​the object to be identified on the preset plane; if so, jump to S505; if not, jump to S508.

[0166] S505: The object to be identified is taken as a new historical candidate object.

[0167] S506: Determine whether there is an overlapping area between the first projection area and the second projection area in the continuous N frames of radar scanning data from the current frame of radar scanning data frame to the next frame; if so, jump to S507; if not, jump to S508.

[0168] S507: Determine that the object to be identified is an obstacle in the target scene.

[0169] S508: Determine that the object to be identified is a target object to be deleted.

[0170] Those skilled in the art will understand that in the above-mentioned method of the specific implementation method, the writing order of each step does not mean a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0171] Based on the same inventive concept, an obstacle detection device corresponding to the obstacle detection method is also provided in the embodiment of the present disclosure. Since the principle of solving the problem by the device in the embodiment of the present disclosure is similar to the above-mentioned obstacle detection method in the embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0172] Reference Figure 6FIG. 1 is a schematic diagram of an obstacle detection device provided by an embodiment of the present disclosure, wherein the device includes: a first determination module 61, a second determination module 62, and a third determination module 63; wherein,

[0173] A first determining module 61 is configured to determine a first position point corresponding to an object to be identified in the target scene based on current frame radar scanning data obtained by scanning the target scene;

[0174] A second determination module 62 is configured to determine whether the object to be identified is a target object to be deleted based on historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point;

[0175] The third determining module 63 is configured to determine the object to be identified as an obstacle in the target scene in response to the object to be identified not being the target object to be deleted.

[0176] In an optional embodiment, when the second determination module 62 determines whether the object to be identified is a target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point, it is used to: judge whether the object to be identified is a current candidate deletion object based on the first position point corresponding to the object to be identified; in response to the object to be identified being the current candidate deletion object, determine whether the object to be identified is a target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point.

[0177] In an optional implementation, the second determining module 62 is further configured to: in response to the object to be identified not being the current candidate deletion object, determine the object to be identified as an obstacle in the target scene.

[0178] In an optional embodiment, when the first determination module 61 determines the first position point corresponding to the object to be identified in the target scene based on the current frame radar scanning data obtained by scanning the target scene, it is used to: determine, for each object to be identified, the point cloud point corresponding to the object to be identified from the current frame radar scanning data; determine the contour information corresponding to the object to be identified based on the three-dimensional position information of the point cloud point corresponding to the object to be identified in the target scene; and determine the first position point corresponding to the object to be identified based on the contour information.

[0179] In an optional embodiment, when determining the contour information corresponding to the object to be identified based on the three-dimensional position information of the point cloud points corresponding to the object to be identified in the target scene, the first determination module 61 is used to: project the point cloud points corresponding to the object to be identified onto a preset plane to obtain a first projection point; and determine the contour information of the object to be identified based on the two-dimensional position information of the first projection point in the preset plane.

[0180] In an optional embodiment, when the first determination module 61 determines the first position point corresponding to the object to be identified based on the contour information, it is used to: use the contour information of the object to be identified to determine the projection area of ​​the object to be identified in a preset plane; and determine the first position point corresponding to the object to be identified based on the area of ​​the projection area.

[0181] In an optional embodiment, when determining the first position point corresponding to the object to be identified based on the area of ​​the projection area, the first determination module 61 is used to: compare the area of ​​the projection area with a preset area threshold; in response to the area being greater than the area threshold, determine the minimum enclosing box of the projection area based on the projection area; determine multiple alternative position points within the first area based on the first area corresponding to the minimum enclosing box and a preset first interval step; and determine the alternative position point located within the projection area as the first position point.

[0182] In an optional embodiment, when the first determination module 61 determines the first position point corresponding to the object to be identified based on the area of ​​the projection area, it is also used to: compare the area of ​​the projection area with a preset area threshold; in response to the area being less than or equal to the area threshold, determine the center point of the projection area; based on the center point and a preset radius length, determine a second area with the center point as the center and the preset radius length as the radius; based on the second area and a preset second interval step, determine the first position point within the second area.

[0183] In an optional embodiment, when the second determination module 62 determines whether the object to be identified is the current candidate deletion object based on the first position point corresponding to the object to be identified, it is used to: obtain the current frame image obtained by capturing the target scene; project the first position point into the current frame image to obtain a second projection point; and determine whether the object to be identified is the current candidate deletion object based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image.

[0184] In an optional embodiment, when the second determination module 62 determines whether the object to be identified is the current candidate object for deletion based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image, it is used to: predict the obstacle prediction result corresponding to the second projection point based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image; the obstacle prediction result includes: whether there is an obstacle at the position corresponding to the second projection point, or whether there is no obstacle; based on the obstacle prediction results corresponding to each second projection point, determine whether the object to be identified is the current candidate object for deletion.

[0185] In an optional embodiment, when the second determination module 62 determines whether the object to be identified is the current candidate for deletion based on the obstacle prediction results corresponding to each second projection point, it is used to: determine the confidence level that the object to be identified is an obstacle based on the obstacle prediction results corresponding to each second projection point; and determine whether the object to be identified is the current candidate for deletion based on the confidence level and a preset confidence threshold.

[0186] In an optional embodiment, there are n second projection points; when determining the confidence level that the object to be identified is an obstacle based on the obstacle prediction results corresponding to each second projection point, the second determination module 62 is used to: traverse the 2nd to nth second projection points; for the traversed i-th second projection point, determine the criterion function corresponding to the i-th second projection point based on the obstacle prediction result of the i-th second projection point; wherein i is a positive integer greater than 1; based on the criterion function corresponding to the i-th second projection point and the fusion criterion results of the 1st to i-1th second projection points, determine the fusion criterion result corresponding to the i-th second projection point; and based on the fusion criterion result corresponding to the i-th second projection point, obtain the confidence level that the object to be identified is an obstacle.

[0187] In an optional embodiment, when the second determination module 62 determines whether the object to be identified is the target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point, it is used to: determine the target candidate object from the historical candidate objects based on the historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point; determine whether the object to be identified is the target object to be deleted based on the first projection area of ​​the target candidate object on the preset plane and the second projection area of ​​the object to be identified on the preset plane.

[0188] In an optional embodiment, when determining the target candidate object from the historical candidate objects based on the historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point, the second determination module 62 is used to: determine the distance between each historical candidate object and the object to be identified based on the historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point; and determine, from the historical candidate objects, the historical candidate object that is closest to the object to be identified as the target candidate object based on the distance between each historical candidate object and the object to be identified.

[0189] In an optional embodiment, when the second determination module 62 determines whether the object to be identified is a target object to be deleted based on the first projection area of ​​the target candidate object on the preset plane and the second projection area of ​​the object to be identified on the preset plane, it is used to: determine whether there is an overlapping area between the first projection area and the second projection area; and in response to the fact that there is no overlapping area between the first projection area and the second projection area, determine that the object to be identified is a target object to be deleted.

[0190] In an optional embodiment, the second determination module 62, when determining whether the object to be identified is the target object to be deleted based on the first projection area of ​​the target candidate object on the preset plane and the second projection area of ​​the object to be identified on the preset plane, is also used to: in response to the existence of an overlapping area between the first projection area and the second projection area in the current frame radar scanning data, take the object to be identified as a new historical candidate object, until there is an overlapping area between the first projection area and the second projection area in N consecutive frames of radar scanning data after the current frame radar scanning data, it is determined that the candidate deletion object is not the target object to be deleted, and the new historical candidate object is deleted; N is a positive integer.

[0191] In an optional embodiment, the detection device also includes: a processing module 64, which is used to detect the storage time of each historical candidate deletion object and the time difference between the storage time of the historical candidate deletion object and the current time; if the time difference is greater than or equal to a preset time difference threshold, the historical candidate deletion object is deleted.

[0192] For descriptions of the processing flow of each module in the device and the interaction flow between each module, reference can be made to the relevant descriptions in the above method embodiment, which will not be described in detail here.

[0193] The present disclosure also provides a computer device, such as Figure 7 FIG. 1 is a schematic diagram of a computer device structure provided by an embodiment of the present disclosure, including:

[0194] A processor 10 and a memory 20; the memory 20 stores machine-readable instructions executable by the processor 10, and the processor 10 is configured to execute the machine-readable instructions stored in the memory 20. When the machine-readable instructions are executed by the processor 10, the processor 10 performs the following steps:

[0195] Based on the current frame radar scanning data obtained by scanning the target scene, a first position point corresponding to the object to be identified in the target scene is determined; based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point, it is determined whether the object to be identified is the target object to be deleted; in response to the object to be identified not being the target object to be deleted, the object to be identified is determined to be an obstacle in the target scene.

[0196] The above-mentioned memory 20 includes internal memory 210 and external memory 220; the memory 210 here is also called internal memory, which is used to temporarily store the calculation data in the processor 10, as well as the data exchanged with the external memory 220 such as the hard disk. The processor 10 exchanges data with the external memory 220 through the internal memory 210.

[0197] The specific execution process of the above instructions can refer to the steps of the obstacle detection method described in the embodiment of the present disclosure, and will not be repeated here.

[0198] The present disclosure also provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program executes the steps of the obstacle detection method described in the above method embodiment. The storage medium may be a volatile or non-volatile computer-readable storage medium.

[0199] The embodiments of the present disclosure further provide a computer program product, which carries program code. The instructions included in the program code can be used to execute the steps of the obstacle detection method described in the above method embodiment. For details, please refer to the above method embodiment and will not be repeated here.

[0200] The computer program product may be implemented in hardware, software, or a combination thereof. In one embodiment, the computer program product is implemented as a computer storage medium. In another embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).

[0201] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0202] 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 to achieve the purpose of this embodiment according to actual needs.

[0203] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0204] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art or the 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 a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0205] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The scope of protection of the present disclosure is not limited thereto. Although the present disclosure has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed in the present disclosure, or replace some of the technical features therein with equivalents. Such modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure shall be subject to the scope of protection of the claims.

Claims

1. A method for detecting an obstacle, characterized in that: include: Determining a first position point corresponding to an object to be identified in the target scene based on current frame radar scanning data obtained by scanning the target scene; The first position point is determined by projecting contour information obtained by the point cloud points of the object to be identified; Determining whether the object to be identified is a target object to be deleted based on historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point; In response to the object to be identified not being the target object to be deleted, determining the object to be identified as an obstacle in the target scene; The determining whether the object to be identified is a target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point includes: judging whether the object to be identified is a current candidate deletion object based on the first position point corresponding to the object to be identified; in response to the object to be identified being the current candidate deletion object, determining whether the object to be identified is a target object to be deleted based on the historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point; The determining, based on the first position point corresponding to the object to be identified, whether the object to be identified is a current candidate for deletion includes: Acquire a current frame image obtained by capturing the target scene; Projecting the first position point onto the current frame image to obtain a second projection point; Based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image, it is determined whether the object to be identified is the current candidate deletion object.

2. The detection method according to claim 1, wherein Also includes: In response to the object to be identified not being the current candidate deletion object, the object to be identified is determined to be an obstacle in the target scene.

3. The detection method according to claim 1 or 2, characterized in that The determining, based on the current frame radar scanning data obtained by scanning the target scene, a first position point corresponding to the object to be identified in the target scene includes: For each of the objects to be identified, determining a point cloud point corresponding to the object to be identified from the radar scanning data of the current frame; Determining contour information corresponding to the object to be identified based on three-dimensional position information of the point cloud points corresponding to the object to be identified in the target scene; Based on the contour information, a first position point corresponding to the object to be identified is determined.

4. The detection method according to claim 3, characterized in that The determining, based on the three-dimensional position information of the point cloud points corresponding to the object to be identified in the target scene, the contour information corresponding to the object to be identified includes: Projecting the point cloud points corresponding to the object to be identified onto a preset plane to obtain a first projection point; Based on the two-dimensional position information of the first projection point in the preset plane, the contour information of the object to be identified is determined.

5. The detection method according to claim 3 or 4, characterized in that The determining, based on the contour information, a first position point corresponding to the object to be identified includes: Determining a projection area of ​​the object to be identified in a preset plane using the contour information of the object to be identified; Based on the area of ​​the projection region, a first position point corresponding to the object to be identified is determined.

6. The detection method according to claim 5, characterized in that Determining a first position point corresponding to the object to be identified based on the area of ​​the projection area includes: Comparing the area of ​​the projection region with a preset area threshold; In response to the area being greater than the area threshold, determining a minimum bounding box of the projection area based on the projection area; Determining a plurality of candidate position points within a first area corresponding to the minimum bounding box and a preset first interval step size; A candidate position point located within the projection area is determined as the first position point.

7. The detection method according to claim 5, characterized in that The determining, based on the area of ​​the projection region, a first position point corresponding to the object to be identified further includes: Comparing the area of ​​the projection region with a preset area threshold; In response to the area being less than or equal to the area threshold, determining a center point of the projection area; Based on the center point and the preset radius length, determining a second area with the center point as the center and the preset radius length as the radius; Based on the second area and a preset second interval step, the first position point is determined in the second area.

8. The detection method according to claim 1, wherein The determining whether the object to be identified is the current candidate deletion object based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image includes: Based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image, predicting an obstacle prediction result corresponding to the second projection point; the obstacle prediction result includes: whether an obstacle exists at the position corresponding to the second projection point, or whether an obstacle does not exist; Based on the obstacle prediction results respectively corresponding to each of the second projection points, it is determined whether the object to be identified is the current candidate deletion object.

9. The detection method according to claim 8, characterized in that The determining, based on the obstacle prediction results corresponding to each of the second projection points, whether the object to be identified is the current candidate deletion object includes: Determining a confidence level that the object to be identified is an obstacle based on obstacle prediction results corresponding to each of the second projection points; Based on the confidence level and a preset confidence level threshold, it is determined whether the object to be identified is the current candidate deletion object.

10. The detection method according to claim 9, characterized in that: There are n second projection points; The determining, based on the obstacle prediction results corresponding to each second projection point, the confidence level that the object to be identified is an obstacle includes: Traverse the second projection points from the 2nd to the nth; For the traversed i-th second projection point, determine a criterion function corresponding to the i-th second projection point based on the obstacle prediction result of the i-th second projection point; where i is a positive integer greater than 1; Determine the fusion criterion result corresponding to the i-th second projection point based on the criterion function corresponding to the i-th second projection point and the fusion criterion results of the 1st to (i-1)-th second projection points; Based on the fusion criterion result corresponding to the i-th second projection point, a confidence level for determining that the object to be identified is an obstacle is obtained.

11. The detection method according to any one of claims 1 to 10, characterized in that The determining whether the object to be identified is a target object to be deleted based on historical position information corresponding to the historical candidate deletion object in the historical frame radar scanning data and the first position point includes: Determine a target candidate object from the historical candidate objects based on historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point; Based on a first projection area of ​​the target candidate object on a preset plane and a second projection area of ​​the object to be identified on the preset plane, it is determined whether the object to be identified is a target object to be deleted.

12. The detection method according to claim 11, characterized in that The determining of a target candidate object from the historical candidate objects based on historical position information corresponding to the historical candidate objects in the historical frame radar scanning data and the first position point includes: Determining the distance between each historical candidate object and the object to be identified based on historical position information corresponding to the historical candidate object in the historical frame radar scanning data and the first position point; Based on the distances between each historical candidate object and the object to be identified, the historical candidate object with the closest distance to the object to be identified is determined from the historical candidate objects as the target candidate object.

13. The detection method according to claim 11 or 12, characterized in that: The determining whether the object to be identified is a target object to be deleted based on the first projection area of ​​the target candidate object on the preset plane and the second projection area of ​​the object to be identified on the preset plane includes: determining whether there is an overlapping area between the first projection area and the second projection area; In response to the fact that there is no overlapping area between the first projection area and the second projection area, it is determined that the object to be identified is a target object to be deleted.

14. The detection method according to claim 13, characterized in that The determining whether the object to be identified is a target object to be deleted based on the first projection area of ​​the target candidate object on the preset plane and the second projection area of ​​the object to be identified on the preset plane further includes: In response to the presence of an overlapping area between the first projection area and the second projection area in the current frame radar scan data, the object to be identified is taken as a new historical candidate object. Until there is an overlapping area between the first projection area and the second projection area in N consecutive frames of radar scan data after the current frame radar scan data, it is determined that the new historical candidate object is not the target object to be deleted, and the new historical candidate object is deleted; N is a positive integer.

15. The detection method according to any one of claims 1 to 14, characterized in that Also includes: For each of the historical candidate deletion objects, detecting the storage time of the historical candidate deletion object and the time difference between the storage time and the current time; If the time difference is greater than or equal to a preset time difference threshold, the historical candidate deletion object is deleted.

16. An obstacle detection device, characterized in that: include: A first determining module is configured to determine a first position point corresponding to an object to be identified in the target scene based on current frame radar scanning data obtained by scanning the target scene; The first position point is determined by projecting contour information obtained by the point cloud points of the object to be identified; A second determination module is configured to determine whether the object to be identified is a target object to be deleted based on historical position information corresponding to historical candidate deletion objects in the historical frame radar scanning data and the first position point; a third determining module, configured to, in response to the object to be identified not being the target object to be deleted, determine the object to be identified as an obstacle in the target scene; The second determination module is specifically configured to: determine whether the object to be identified is a current candidate for deletion based on a first position point corresponding to the object to be identified; in response to the object to be identified being the current candidate for deletion, determine whether the object to be identified is a target object to be deleted based on historical position information corresponding to historical candidate for deletion objects in the historical frame radar scanning data and the first position point; The second determining module, when determining whether the object to be identified is a current candidate deletion object based on the first position point corresponding to the object to be identified, is specifically configured to: Acquire a current frame image obtained by capturing the target scene; Projecting the first position point onto the current frame image to obtain a second projection point; Based on the position information of the second projection point in the current frame image and the position of the obstacle included in the current frame image in the current frame image, it is determined whether the object to be identified is the current candidate deletion object.

17. A computer device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the processor is configured to execute the machine-readable instructions stored in the memory. When the machine-readable instructions are executed by the processor, the processor performs the steps of the obstacle detection method according to any one of claims 1 to 15.

18. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program. When the computer program is executed by a computer device, the computer device performs the steps of the obstacle detection method according to any one of claims 1 to 15.

Images