Obstacle aggregation method and device, electronic equipment and storage medium

By analyzing sensor data in real time and selecting a merging strategy based on clustering conditions, the problems of low accuracy of obstacle detection and insufficient real-time performance are solved, and more accurate and efficient obstacle aggregation and merging are achieved, enhancing the environmental perception ability of smart vehicles .

CN118982810APending Publication Date: 2024-11-19FULSCIENCE AUTOMOTIVE ELECTRONICS CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411068980.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

In the prior art, obstacle detection has low accuracy under scene image interference, and deep learning-based clustering algorithms require a large amount of labeled data and have low real-time performance.

Method used

By analyzing the current measurement data and the previous measurement data in real time, the corresponding target merging strategy is selected according to different clustering conditions, and the location and type information of obstacles are processed to improve the real-time and accuracy of obstacle aggregation.

Benefits of technology

It improves the accuracy and continuity of obstacle detection results, enhances the robustness and applicability of the algorithm, simplifies the design, reduces the calculation amount of subsequent processing, and improves the processing efficiency of the overall system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118982810A_ABST
    Figure CN118982810A_ABST
Patent Text Reader

Abstract

The invention provides an obstacle aggregation method and apparatus, an electronic device and a storage medium. The method comprises the steps of obtaining current measurement data and previous measurement data collected by a target sensor; analyzing the current measurement data and the previous measurement data, and determining whether any preset clustering condition is met or not; when the preset clustering condition is satisfied, selecting a corresponding target merging strategy according to the satisfied preset clustering condition; and processing the first obstacle in the current measurement data and the second obstacle in the current measurement data according to the target merging strategy, and outputting the processed target obstacle. Therefore, the current measurement data and the previous measurement data are analyzed in real time, and the corresponding target merging strategies are selected according to different clustering conditions, so that the real-time performance of obstacle aggregation and the accuracy of the aggregation result can be effectively ensured.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of sensor detection technology, and in particular to an obstacle aggregation method, device, electronic device and storage medium. Background Art

[0002] The environmental perception technology of intelligent vehicles in outdoor environments is the key technology for realizing autonomous navigation, and obstacle detection is essential for correct and safe navigation. In the prior art, sensors are usually installed on vehicles to sense the surrounding environment, identify and track obstacles, etc. For example, millimeter wave radar, laser radar and camera. However, in real life, once the scene image is disturbed, it is easy to cause the accuracy of the detected image data to be low, which easily leads to low accuracy of the detection result of whether there is an obstacle. The accuracy of obstacle detection results can be effectively improved by obstacle fusion processing. Among them, the prior art also proposes some obstacle clustering processing methods, such as clustering algorithms based on deep learning. This method uses deep learning models such as convolutional neural networks to extract features from ultrasonic data and perform obstacle clustering; however, this method requires a large amount of labeled data and training in advance, so there is a problem of low real-time performance, and if the labeling personnel are not professional, there will also be problems with incorrect clustering results. Summary of the invention

[0003] In view of this, the purpose of the present application is to provide an obstacle aggregation method, device, electronic device and storage medium, which can effectively ensure the real-time performance of obstacle aggregation and the accuracy of the aggregation results by analyzing the current measurement data and the previous measurement data in real time and selecting the corresponding target merging strategy according to different clustering conditions.

[0004] The embodiment of the present application provides a method for aggregating obstacles, the method comprising:

[0005] Obtain the current measurement data and the previous measurement data collected by the target sensor;

[0006] Analyze the current measurement data and the previous measurement data to determine whether any preset clustering condition is met;

[0007] When the conditions are met, the corresponding target merging strategy is selected according to the preset clustering conditions that are met;

[0008] According to the target merging strategy, the first obstacle in the current measurement data and the second obstacle in the current measurement data are processed, and the processed target obstacle is output.

[0009] Optionally, analyzing the current measurement data and the previous measurement data to determine whether any preset clustering condition is satisfied includes:

[0010] Identify whether the current measurement data and the previous measurement data include obstacle information;

[0011] If both are included, obtaining and determining first attribute information of the first obstacle included in the current measurement data, and obtaining second attribute information of the second obstacle included in the previous measurement data;

[0012] Determining, according to first location information in the first attribute information and second location information in the second attribute information, whether the first obstacle and the second obstacle meet a location clustering condition;

[0013] If the location clustering condition is met, select the target merging strategy corresponding to the location clustering condition;

[0014] If the position clustering condition is not met, determining whether the first obstacle and the second obstacle meet the type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information;

[0015] If the type clustering condition is met, the target merging strategy corresponding to the type clustering condition is selected.

[0016] Optionally, determining whether the first obstacle and the second obstacle meet a position clustering condition according to the first position information in the first attribute information and the second position information in the second attribute information includes:

[0017] Determine a deviation distance between the first obstacle and the second obstacle according to the first position information in the first attribute information and the second position information in the second attribute information;

[0018] If the deviation distance is less than a preset distance threshold, determining that the first obstacle and the second obstacle meet a position clustering condition;

[0019] If the deviation distance is not less than a preset distance threshold, it is determined that the first obstacle and the second obstacle do not meet the position clustering condition.

[0020] Optionally, determining whether the first obstacle and the second obstacle meet a type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information includes:

[0021] determining, according to first type information in the first attribute information and second type information in the second attribute information, whether the first obstacle and the second obstacle are of the same type;

[0022] If the types are consistent, determining that the first obstacle and the second obstacle meet a type clustering condition;

[0023] If the types are inconsistent, it is determined that the first obstacle and the second obstacle do not meet the type clustering condition.

[0024] Optionally, determining whether the first obstacle and the second obstacle meet a type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information includes:

[0025] Determine a target type pair of the first obstacle and the second obstacle according to first type information in the first attribute information and second type information in the second attribute information;

[0026] Determining whether the target type pair is a preset type pair;

[0027] If yes, determining that the first obstacle and the second obstacle meet the type clustering condition;

[0028] If not, it is determined that the first obstacle and the second obstacle do not meet the type clustering condition.

[0029] Optionally, when the preset clustering condition satisfied is a type clustering condition, selecting a corresponding target merging strategy according to the preset clustering condition satisfied includes:

[0030] Determine a target type pair of the first obstacle and the second obstacle according to first type information in the first attribute information and second type information in the second attribute information;

[0031] According to the predetermined correspondence between type pairs and merging strategies, the target type pairs are used to perform a corresponding search to determine the target merging strategy.

[0032] Optionally, the processing, according to the target merging strategy, the first obstacle in the current measurement data and the second obstacle in the current measurement data, and outputting the processed target obstacle includes:

[0033] According to the target merging strategy, analyzing and processing the first attribute information of the first obstacle and the second attribute information of the second obstacle to determine whether a merging condition is met;

[0034] If the conditions are met, the first obstacle and the second obstacle are merged according to a merging method in a target merging strategy to obtain a target obstacle;

[0035] If not satisfied, the first obstacle is determined as the target obstacle and output.

[0036] The embodiment of the present application further provides an obstacle aggregation device, the aggregation device comprising:

[0037] An acquisition module is used to acquire the current measurement data and the previous measurement data collected by the target sensor;

[0038] An analysis module, used to analyze the current measurement data and the previous measurement data to determine whether any preset clustering condition is met;

[0039] A selection module, for selecting a corresponding target merging strategy according to the preset clustering condition when it is determined that any preset clustering condition is met;

[0040] The merging module is used to process the first obstacle in the current measurement data and the second obstacle in the current measurement data according to the target merging strategy, and output the processed target obstacle.

[0041] Optionally, when the analysis module is used to analyze the current measurement data and the previous measurement data to determine whether any preset clustering condition is met, the analysis module is used to:

[0042] Identify whether the current measurement data and the previous measurement data include obstacle information;

[0043] If both are included, obtaining and determining first attribute information of the first obstacle included in the current measurement data, and obtaining second attribute information of the second obstacle included in the previous measurement data;

[0044] Determining, according to first location information in the first attribute information and second location information in the second attribute information, whether the first obstacle and the second obstacle meet a location clustering condition;

[0045] If the location clustering condition is met, select the target merging strategy corresponding to the location clustering condition;

[0046] If the position clustering condition is not met, determining whether the first obstacle and the second obstacle meet the type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information;

[0047] If the type clustering condition is met, the target merging strategy corresponding to the type clustering condition is selected.

[0048] Optionally, when the analysis module is used to determine whether the first obstacle and the second obstacle meet the position clustering condition according to the first position information in the first attribute information and the second position information in the second attribute information, the analysis module is used to:

[0049] Determine a deviation distance between the first obstacle and the second obstacle according to the first position information in the first attribute information and the second position information in the second attribute information;

[0050] If the deviation distance is less than a preset distance threshold, determining that the first obstacle and the second obstacle meet a position clustering condition;

[0051] If the deviation distance is not less than a preset distance threshold, it is determined that the first obstacle and the second obstacle do not meet the position clustering condition.

[0052] Optionally, when the analysis module is used to determine whether the first obstacle and the second obstacle meet a type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information, the analysis module is used to:

[0053] determining, according to first type information in the first attribute information and second type information in the second attribute information, whether the first obstacle and the second obstacle are of the same type;

[0054] If the types are consistent, determining that the first obstacle and the second obstacle meet a type clustering condition;

[0055] If the types are inconsistent, it is determined that the first obstacle and the second obstacle do not meet the type clustering condition.

[0056] Optionally, when the analysis module is used to determine whether the first obstacle and the second obstacle meet a type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information, the analysis module is used to:

[0057] Determine a target type pair of the first obstacle and the second obstacle according to first type information in the first attribute information and second type information in the second attribute information;

[0058] Determining whether the target type pair is a preset type pair;

[0059] If yes, determining that the first obstacle and the second obstacle meet the type clustering condition;

[0060] If not, it is determined that the first obstacle and the second obstacle do not meet the type clustering condition.

[0061] Optionally, when the selection module is used to select a corresponding target merging strategy according to the preset clustering condition that is satisfied, the selection module is used to:

[0062] When the preset clustering condition satisfied is a type clustering condition, determining a target type pair of the first obstacle and the second obstacle according to first type information in the first attribute information and second type information in the second attribute information;

[0063] According to the predetermined correspondence between type pairs and merging strategies, the target type pairs are used to perform a corresponding search to determine the target merging strategy.

[0064] Optionally, when the merging module is used to process the first obstacle in the current measurement data and the second obstacle in the current measurement data according to the target merging strategy and output the processed target obstacle, the merging module is used to:

[0065] According to the target merging strategy, analyzing and processing the first attribute information of the first obstacle and the second attribute information of the second obstacle to determine whether a merging condition is met;

[0066] If the conditions are met, the first obstacle and the second obstacle are merged according to a merging method in a target merging strategy to obtain a target obstacle;

[0067] If not satisfied, the first obstacle is determined as the target obstacle and output.

[0068] An embodiment of the present application also provides an electronic device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the aggregation method as described above are performed.

[0069] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned aggregation method are executed.

[0070] An embodiment of the present application provides an obstacle aggregation method, device, electronic device and storage medium, the aggregation method comprising: obtaining current measurement data and previous measurement data collected by a target sensor; analyzing the current measurement data and the previous measurement data to determine whether any preset clustering condition is satisfied; when satisfied, selecting a corresponding target merging strategy according to the satisfied preset clustering condition; processing a first obstacle in the current measurement data and a second obstacle in the current measurement data according to the target merging strategy, and outputting the processed target obstacle.

[0071] In this way, the technical solution provided by this application can achieve the following technical effects:

[0072] 1. Better handle the clustering relationship between different types of obstacles. For different types of obstacle objects, corresponding merging strategies are designed to more accurately identify and merge similar obstacles, improving the accuracy and continuity of the aggregation results.

[0073] 2. Enhanced the robustness and applicability of the algorithm. Corresponding processing measures are proposed for some special obstacle scenarios to ensure that the algorithm can cope with various complex obstacle environments and greatly improve its scope of application.

[0074] 3. The algorithm design is simple and efficient. The algorithm realizes the effective aggregation and correction of obstacles through the analysis of sensor data and the design of clustering logic. The overall design is simple and efficient, easy to implement and deploy.

[0075] 4. Improved efficiency of subsequent processing. By clustering and merging obstacles, a more continuous and complete object model can be generated, which greatly reduces the amount of calculation and complexity of subsequent processing and improves the processing efficiency of the overall system.

[0076] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0078] Figure 1 A flowchart of a method xx provided in an embodiment of the present application is shown;

[0079] Figure 2 A schematic diagram of a judgment process for satisfying preset clustering conditions provided in this application;

[0080] Figure 3 A schematic diagram of the structure of an obstacle aggregation device provided in an embodiment of the present application;

[0081] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0082] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application usually described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, each other embodiment obtained by those skilled in the art without making creative work belongs to the scope of protection of the present application.

[0083] The environmental perception technology of intelligent vehicles in outdoor environments is the key technology for realizing autonomous navigation, and obstacle detection is essential for correct and safe navigation. In the prior art, sensors are usually installed on vehicles to sense the surrounding environment, identify and track obstacles, etc. For example, millimeter wave radar, laser radar and camera. However, in real life, once the scene image is disturbed, it is easy to cause the accuracy of the detected image data to be low, which easily leads to low accuracy of the detection result of whether there is an obstacle. The accuracy of obstacle detection results can be effectively improved by obstacle fusion processing. Among them, the prior art also proposes some obstacle clustering processing methods, such as clustering algorithms based on deep learning. This method uses deep learning models such as convolutional neural networks to extract features from ultrasonic data and perform obstacle clustering; however, this method requires a large amount of labeled data and training in advance, so there is a problem of low real-time performance, and if the labeling personnel are not professional, there will also be problems with incorrect clustering results.

[0084] Based on this, the embodiments of the present application provide an obstacle aggregation method, device, electronic device and storage medium, which effectively ensures the real-time performance of obstacle aggregation and the accuracy of the aggregation results by analyzing the current measurement data and the previous measurement data in real time and selecting the corresponding target merging strategy according to different clustering conditions.

[0085] Here, the aggregation method is a cluster merging method.

[0086] See also Figure 1 , Figure 1 This is a flow chart of an obstacle aggregation method provided in an embodiment of the present application. Figure 1 As shown in, the polymerization method provided in the embodiment of the present application includes:

[0087] S101. Acquire current measurement data and previous measurement data collected by a target sensor.

[0088] S102: Analyze the current measurement data and the previous measurement data to determine whether any preset clustering condition is satisfied.

[0089] S103: When the conditions are met, select a corresponding target merging strategy according to the preset clustering conditions met.

[0090] S104: Process the first obstacle in the current measurement data and the second obstacle in the current measurement data according to the target merging strategy, and output the processed target obstacle.

[0091] Regarding step S101, in this step, the target sensor may be a sensor installed on the vehicle for sensing the surrounding environment. For example, the target sensor may be a laser radar, a radar, a camera, an ultrasonic sensor, etc. The target sensor in this solution may be an ultrasonic sensor.

[0092] Here, it should be noted that the existing ultrasonic obstacle clustering algorithm has some objective defects. For example, the convergence speed of the existing ultrasonic obstacle clustering algorithm is slow. This is because the existing clustering algorithm needs to perform complex clustering calculations on a large amount of obstacle data, resulting in low overall processing efficiency. Secondly, the algorithm has weak noise resistance. Specifically, in practical applications, ultrasonic sensors are easily affected by environmental interference and generate more noise data, which will have a greater impact on the clustering results.

[0093] Continuing with step S101 , the current measurement data is data determined after processing the raw data currently collected by the target sensor; the previous measurement data is data determined after processing the raw data collected by the target sensor at the previous moment.

[0094] Here, data processing is obstacle analysis processing, which may include, for example, denoising, coordinate conversion and other processing means.

[0095] With respect to step S102, the preset clustering condition may include a location clustering condition and a type clustering condition.

[0096] According to different preset clustering conditions, different analysis and processing methods are used for the current measurement data and the previous measurement data.

[0097] Continuing with step S102, please refer to Figure 2 , Figure 2 A schematic diagram of a judgment process for satisfying preset clustering conditions provided by this application. Figure 2 As shown, in an implementation manner provided in the present application, analyzing the current measurement data and the previous measurement data to determine whether any preset clustering condition is met includes:

[0098] S1021. Identify whether the current measurement data and the previous measurement data include obstacle information.

[0099] S1022: If both are included, obtain and determine first attribute information of the first obstacle included in the current measurement data, and obtain second attribute information of the second obstacle included in the previous measurement data.

[0100] S1023. Determine whether the first obstacle and the second obstacle meet a position clustering condition according to the first position information in the first attribute information and the second position information in the second attribute information.

[0101] S1024: If the location clustering condition is met, select a target merging strategy corresponding to the location clustering condition.

[0102] S1025: If the position clustering condition is not met, determine whether the first obstacle and the second obstacle meet the type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information.

[0103] S1026: If the type clustering condition is met, select a target merging strategy corresponding to the type clustering condition.

[0104] With respect to step S1021, in this step, it is identified whether the current measurement data includes a detection result indicating the existence of an obstacle, and it is identified whether the previous measurement data includes a detection result indicating the existence of an obstacle.

[0105] If it is determined that the current measurement data or the previous measurement data includes obstacle information, the included obstacle information may include basic information such as the position, direction, and size of the obstacle, and may also include obstacle type information.

[0106] For example, please refer to Table 1, which is a table of obstacle information provided:

[0107] Table 1:

[0108]

[0109]

[0110] The obstacle type information may be further specifically divided into point obstacles, line obstacles and surface obstacles.

[0111] Continuing with step S1021, if it is identified that the current measurement data includes obstacle information, and the previous measurement data includes obstacle information, step S1022 is executed.

[0112] If it is determined that the current measurement data does not include obstacle information and the previous measurement data includes obstacle information, the result is output, and the result is empty at this time.

[0113] If it is determined that the current measurement data includes obstacle information and the previous measurement data does not include obstacle information, the result is output, and the result is the obstacle information in the current measurement data.

[0114] If it is determined that the current measurement data does not include obstacle information and the previous measurement data does not include obstacle information, the result is output, and the result is empty at this time.

[0115] Continuing with step S1022, in this step, first attribute information of the first obstacle included in the current measurement data is obtained and determined, and second attribute information of the second obstacle included in the previous measurement data is obtained.

[0116] The first attribute information may include at least one of the following: the location, direction, size and type of the obstacle; the second attribute information may include at least one of the following: the location, direction, size and type of the obstacle.

[0117] With respect to step S1023, in an implementation manner provided in the present application, determining whether the first obstacle and the second obstacle meet the position clustering condition according to the first position information in the first attribute information and the second position information in the second attribute information includes:

[0118] S10231. Determine a deviation distance between the first obstacle and the second obstacle based on the first position information in the first attribute information and the second position information in the second attribute information.

[0119] S10232: If the deviation distance is less than a preset distance threshold, determine that the first obstacle and the second obstacle meet a position clustering condition.

[0120] S10233: If the deviation distance is not less than a preset distance threshold, determine that the first obstacle and the second obstacle do not meet the position clustering condition.

[0121] For step S10231, in this step, when determining the deviation distance between the first obstacle and the second obstacle based on the first position information in the first attribute information and the second position information in the second attribute information, it can be specifically as follows: according to the center position point of the first obstacle in the first attribute information and the center position point of the second obstacle in the second attribute information, determine the deviation distance between the first obstacle and the second obstacle.

[0122] With respect to step S10232, if the deviation distance is less than the preset distance threshold, it is determined that the first obstacle and the second obstacle meet the position clustering condition, and then step S1024 is executed.

[0123] With respect to step S10233, if the deviation distance is not less than the preset distance threshold, it is determined that the first obstacle and the second obstacle do not meet the position clustering condition, and then step S1025 is executed.

[0124] For step S1024, when the position clustering condition is met, when the target merging strategy corresponding to the position clustering condition is selected, for example, a bounding box merging strategy, a point cloud merging strategy, a convex hull merging strategy, a polygon merging strategy, a grid-based merging strategy, etc. can be used. The specific merging strategy to be used can be determined according to the actual situation.

[0125] With respect to step S1025 , after determining that the position clustering condition is not satisfied, it is determined whether the first obstacle and the second obstacle satisfy the type clustering condition.

[0126] In an implementation manner provided in the present application, determining whether the first obstacle and the second obstacle meet the type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information includes:

[0127] S11. Determine whether the first obstacle and the second obstacle are of the same type based on the first type information in the first attribute information and the second type information in the second attribute information.

[0128] S12: If the types are consistent, determine whether the first obstacle and the second obstacle meet a type clustering condition.

[0129] S13: If the types are inconsistent, determine that the first obstacle and the second obstacle do not meet the type clustering condition.

[0130] Here, if it is determined that the first obstacle and the second obstacle are of the same type, step S1025 is executed. If it is determined that the first obstacle and the second obstacle are of different types, the result is output, and the result is the obstacle information in the current measurement data, that is, the first obstacle is output as a target obstacle.

[0131] In another implementation manner provided in the present application, determining whether the first obstacle and the second obstacle meet the type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information includes:

[0132] S21. Determine a target type pair of the first obstacle and the second obstacle according to first type information in the first attribute information and second type information in the second attribute information.

[0133] S22: Determine whether the target type pair is a preset type pair.

[0134] S23: If yes, determine whether the first obstacle and the second obstacle meet the type clustering condition.

[0135] S24: If no, determine that the first obstacle and the second obstacle do not meet the type clustering condition.

[0136] With respect to step S21, when the first type of information is in point form, and when the second type of information is in surface form, the target type pair formed is (point, surface).

[0137] For step S22, a plurality of preset type pairs may be preset.

[0138] Here, if it is determined that the target type pair is a preset type pair, it is determined that the first obstacle and the second obstacle meet the type clustering condition, and then step S1025 is performed. If it is determined that the target type pair is not a preset type pair, it is determined that the first obstacle and the second obstacle do not meet the type clustering condition, and the result is output. In this case, the result is the obstacle information in the current measurement data, that is, the first obstacle is output as the target obstacle.

[0139] With respect to step S103, in an implementation manner provided in the present application, when the preset clustering condition satisfied is a type clustering condition, selecting a corresponding target merging strategy according to the preset clustering condition satisfied includes:

[0140] S1031. Determine a target type pair of the first obstacle and the second obstacle according to first type information in the first attribute information and second type information in the second attribute information.

[0141] S1032: According to the predetermined correspondence between type pairs and merging strategies, use the target type pairs to perform a corresponding search and determine the target merging strategy.

[0142] Here, when the preset clustering condition satisfied is a type clustering condition, the target merging strategy determined may include, for example: two-point merging, three-point merging, wall-type merging, point-surface merging, and the like.

[0143] The merging of two points includes: firstly, determining whether the two points are uniformly moving points or reliably stationary points. Then, determining the existence probability of the two points, and retaining the one with a larger probability. Otherwise, calculating the midpoint of the two points, merging the tracking data and existence probability output of the two points.

[0144] Merge three points: Determine whether the third point is on the straight line where the first two points are located, and determine whether the distance from the third point to the straight line is within the tolerance range based on the distance between the points and the distance to the straight line. If it meets the clustering requirements, merge the three points data output.

[0145] Wall merging: For wall-like obstacles, it is necessary to consider features such as openings and closures, and divide them into open walls, single-closure walls, double-closure walls, and intersecting walls. The principle is that if one end point of two lines (walls) is very close, the lines are merged and spliced. For wall obstacle splicing, it is necessary to consider whether the angles are similar, whether the heights are similar, whether the spaces are in the same space, whether the distances are close enough, and whether they do not block each other.

[0146] For merging points with wall obstacle objects (surfaces), the principle is that if a point is very close to a line (wall), the point is merged into the line.

[0147] For example, point to single-end wall aggregation: first calculate the distance from the point to the wall line, determine whether the distance is within the tolerance range, then determine whether the point is on the wall line, and if so, calculate the distance from the point to the two end points of the wall. If the point is not on the wall line, but on its left or right side, update the wall point information with the point as the new end point of the wall point. Finally, merge the information, delete the point object, and output the information.

[0148] With respect to step S104, in an implementation manner provided in the present application, processing the first obstacle in the current measurement data and the second obstacle in the current measurement data according to the target merging strategy, and outputting the processed target obstacle, includes:

[0149] S1041. Analyze and process the first attribute information of the first obstacle and the second attribute information of the second obstacle according to the target merging strategy to determine whether a merging condition is met.

[0150] S1042: If the conditions are met, merge the first obstacle and the second obstacle according to a merging method in a target merging strategy to obtain a target obstacle.

[0151] S1043: If not satisfied, determine the first obstacle as a target obstacle and output it.

[0152] For step S1041, the merging condition examples may include:

[0153] Proximity: Two obstacles need to be close in space or have overlapping parts. If they are too far apart, they are not suitable for merging.

[0154] Shape and size: The shapes and sizes of the two obstacles need to be compatible. For example, two long obstacles are more likely to merge into one long obstacle.

[0155] Functionality: In some applications, such as robot obstacle avoidance or path planning, the fusion of obstacles needs to consider practical functions. If the fused obstacles are more suitable for navigation or path planning, then they are suitable for merging.

[0156] Dynamic characteristics: If the obstacles are dynamic (such as moving obstacles), their movement trajectories and speeds need to be considered. If they cross or approach each other in motion, they may need to be fused in real time in motion.

[0157] With respect to step S1042, the first obstacle and the second obstacle are fused in a merging manner in a target merging strategy, and geometric fusion and regional fusion processing may be used as examples.

[0158] Geometric fusion: Combine the geometric features of two obstacles. This can be achieved by calculating their bounding boxes, convex hulls, or other geometric representations.

[0159] Region Fusion: Determine the overlapping region of two obstacles and merge them into a unified obstacle. This can be achieved using Boolean operations such as union.

[0160] In this way, the technical solution provided by this application can achieve the following technical effects:

[0161] 1. Better handle the clustering relationship between different types of obstacles. For different types of obstacle objects, corresponding merging strategies are designed to more accurately identify and merge similar obstacles, improving the accuracy and continuity of the aggregation results.

[0162] 2. Enhanced the robustness and applicability of the algorithm. Corresponding processing measures are proposed for some special obstacle scenarios to ensure that the algorithm can cope with various complex obstacle environments and greatly improve its scope of application.

[0163] 3. The algorithm design is simple and efficient. The algorithm realizes the effective aggregation and correction of obstacles through the analysis of sensor data and the design of clustering logic. The overall design is simple and efficient, easy to implement and deploy.

[0164] 4. Improved efficiency of subsequent processing. By clustering and merging obstacles, a more continuous and complete object model can be generated, which greatly reduces the amount of calculation and complexity of subsequent processing and improves the processing efficiency of the overall system.

[0165] See also Figure 3 , Figure 3This is a schematic diagram of the structure of an obstacle aggregation device provided in an embodiment of the present application. Figure 3 As shown in , the polymerization device 300 includes:

[0166] An acquisition module 310 is used to acquire current measurement data and previous measurement data collected by a target sensor;

[0167] An analysis module 320 is used to analyze the current measurement data and the previous measurement data to determine whether any preset clustering condition is met;

[0168] A selection module 330, configured to select a corresponding target merging strategy according to the preset clustering condition when it is determined that any preset clustering condition is met;

[0169] The merging module 340 is used to process the first obstacle in the current measurement data and the second obstacle in the current measurement data according to the target merging strategy, and output the processed target obstacle.

[0170] Optionally, when the analysis module 320 is used to analyze the current measurement data and the previous measurement data to determine whether any preset clustering condition is met, the analysis module 320 is used to:

[0171] Identify whether the current measurement data and the previous measurement data include obstacle information;

[0172] If both are included, obtaining and determining first attribute information of the first obstacle included in the current measurement data, and obtaining second attribute information of the second obstacle included in the previous measurement data;

[0173] Determining, according to first location information in the first attribute information and second location information in the second attribute information, whether the first obstacle and the second obstacle meet a location clustering condition;

[0174] If the location clustering condition is met, select the target merging strategy corresponding to the location clustering condition;

[0175] If the position clustering condition is not met, determining whether the first obstacle and the second obstacle meet the type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information;

[0176] If the type clustering condition is met, the target merging strategy corresponding to the type clustering condition is selected.

[0177] Optionally, when the analysis module 320 is used to determine whether the first obstacle and the second obstacle meet the position clustering condition according to the first position information in the first attribute information and the second position information in the second attribute information, the analysis module 320 is used to:

[0178] Determine a deviation distance between the first obstacle and the second obstacle according to the first position information in the first attribute information and the second position information in the second attribute information;

[0179] If the deviation distance is less than a preset distance threshold, determining that the first obstacle and the second obstacle meet a position clustering condition;

[0180] If the deviation distance is not less than a preset distance threshold, it is determined that the first obstacle and the second obstacle do not meet the position clustering condition.

[0181] Optionally, when the analysis module 320 is used to determine whether the first obstacle and the second obstacle meet the type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information, the analysis module 320 is used to:

[0182] determining, according to first type information in the first attribute information and second type information in the second attribute information, whether the first obstacle and the second obstacle are of the same type;

[0183] If the types are consistent, determining that the first obstacle and the second obstacle meet a type clustering condition;

[0184] If the types are inconsistent, it is determined that the first obstacle and the second obstacle do not meet the type clustering condition.

[0185] Optionally, when the analysis module 320 is used to determine whether the first obstacle and the second obstacle meet the type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information, the analysis module 320 is used to:

[0186] Determine a target type pair of the first obstacle and the second obstacle according to first type information in the first attribute information and second type information in the second attribute information;

[0187] Determining whether the target type pair is a preset type pair;

[0188] If yes, determining that the first obstacle and the second obstacle meet the type clustering condition;

[0189] If not, it is determined that the first obstacle and the second obstacle do not meet the type clustering condition.

[0190] Optionally, when the selection module 330 is used to select a corresponding target merging strategy according to the preset clustering condition that is satisfied, the selection module 330 is used to:

[0191] When the preset clustering condition satisfied is a type clustering condition, determining a target type pair of the first obstacle and the second obstacle according to first type information in the first attribute information and second type information in the second attribute information;

[0192] According to the predetermined correspondence between type pairs and merging strategies, the target type pairs are used to perform a corresponding search to determine the target merging strategy.

[0193] Optionally, when the merging module 340 is used to process the first obstacle in the current measurement data and the second obstacle in the current measurement data according to the target merging strategy and output the processed target obstacle, the merging module 340 is used to:

[0194] According to the target merging strategy, analyzing and processing the first attribute information of the first obstacle and the second attribute information of the second obstacle to determine whether a merging condition is met;

[0195] If the conditions are met, the first obstacle and the second obstacle are merged according to a merging method in a target merging strategy to obtain a target obstacle;

[0196] If not satisfied, the first obstacle is determined as the target obstacle and output.

[0197] See also Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 4 As shown in , the electronic device 400 includes a processor 410 , a memory 420 and a bus 430 .

[0198] The memory 420 stores machine-readable instructions executable by the processor 410. When the electronic device 400 is running, the processor 410 communicates with the memory 420 via the bus 430. When the machine-readable instructions are executed by the processor 410, the above-mentioned Figure 1 as well as Figure 2 The specific implementation of the steps in the method embodiment shown can be found in the method embodiment, and will not be repeated here.

[0199] The present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the computer program can execute the above-mentioned Figure 1 as well as Figure 2The specific implementation of the steps in the method embodiment shown can be found in the method embodiment, and will not be repeated here.

[0200] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0201] In the several embodiments provided in the present application, 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, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0202] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0203] In addition, each functional unit in each embodiment of the present application 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 application can essentially be embodied in the form of a software product, or in other words, the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. 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 application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the above-mentioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-mentioned embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for aggregating obstacles, It is characterized in that The polymerization method comprises: Obtain the current measurement data and the previous measurement data collected by the target sensor; Analyze the current measurement data and the previous measurement data to determine whether any preset clustering condition is met; When the conditions are met, the corresponding target merging strategy is selected according to the preset clustering conditions that are met; According to the target merging strategy, the first obstacle in the current measurement data and the second obstacle in the current measurement data are processed, and the processed target obstacle is output.

2. The polymerization method according to claim 1, It is characterized in that The analyzing the current measurement data and the previous measurement data to determine whether any preset clustering condition is satisfied includes: Identify whether the current measurement data and the previous measurement data include obstacle information; If both are included, obtaining and determining first attribute information of the first obstacle included in the current measurement data, and obtaining second attribute information of the second obstacle included in the previous measurement data; Determining, according to first location information in the first attribute information and second location information in the second attribute information, whether the first obstacle and the second obstacle meet a location clustering condition; If the location clustering condition is met, select the target merging strategy corresponding to the location clustering condition; If the position clustering condition is not met, determining whether the first obstacle and the second obstacle meet the type clustering condition according to the first type information in the first attribute information and the second type information in the second attribute information; If the type clustering condition is met, the target merging strategy corresponding to the type clustering condition is selected.

3. The polymerization method according to claim 2, It is characterized in that The determining, according to the first location information in the first attribute information and the second location information in the second attribute information, whether the first obstacle and the second obstacle meet the location clustering condition includes: Determine a deviation distance between the first obstacle and the second obstacle according to the first position information in the first attribute information and the second position information in the second attribute information; If the deviation distance is less than a preset distance threshold, determining that the first obstacle and the second obstacle meet a position clustering condition; If the deviation distance is not less than a preset distance threshold, it is determined that the first obstacle and the second obstacle do not meet the position clustering condition.

4. The polymerization method according to claim 2, It is characterized in that The determining, according to the first type information in the first attribute information and the second type information in the second attribute information, whether the first obstacle and the second obstacle meet the type clustering condition includes: determining, according to first type information in the first attribute information and second type information in the second attribute information, whether the first obstacle and the second obstacle are of the same type; If the types are consistent, determining that the first obstacle and the second obstacle meet a type clustering condition; If the types are inconsistent, it is determined that the first obstacle and the second obstacle do not meet the type clustering condition.

5. The polymerization method according to claim 2, It is characterized in that The determining, according to the first type information in the first attribute information and the second type information in the second attribute information, whether the first obstacle and the second obstacle meet the type clustering condition includes: Determine a target type pair of the first obstacle and the second obstacle according to first type information in the first attribute information and second type information in the second attribute information; Determining whether the target type pair is a preset type pair; If yes, determining that the first obstacle and the second obstacle meet the type clustering condition; If not, it is determined that the first obstacle and the second obstacle do not meet the type clustering condition.

6. The polymerization method according to claim 2, It is characterized in that When the preset clustering condition satisfied is a type clustering condition, the corresponding target merging strategy is selected according to the preset clustering condition satisfied, including: Determine a target type pair of the first obstacle and the second obstacle according to first type information in the first attribute information and second type information in the second attribute information; According to the predetermined correspondence between type pairs and merging strategies, the target type pairs are used to perform a corresponding search to determine the target merging strategy.

7. The polymerization method according to claim 2, It is characterized in that The processing of the first obstacle in the current measurement data and the second obstacle in the current measurement data according to the target merging strategy, and outputting the processed target obstacle, includes: According to the target merging strategy, analyzing and processing the first attribute information of the first obstacle and the second attribute information of the second obstacle to determine whether a merging condition is met; If the conditions are met, the first obstacle and the second obstacle are merged according to a merging method in a target merging strategy to obtain a target obstacle; If not satisfied, the first obstacle is determined as the target obstacle and output.

8. An obstacle aggregation device, It is characterized in that The polymerization device comprises: An acquisition module is used to acquire the current measurement data and the previous measurement data collected by the target sensor; An analysis module, used to analyze the current measurement data and the previous measurement data to determine whether any preset clustering condition is met; A selection module, for selecting a corresponding target merging strategy according to the preset clustering condition when it is determined that any preset clustering condition is met; The merging module is used to process the first obstacle in the current measurement data and the second obstacle in the current measurement data according to the target merging strategy, and output the processed target obstacle.

9. An electronic device, It is characterized in that include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to execute the steps of the aggregation method as described in any one of claims 1 to 7.

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