Obstacle recognition method and device, electronic equipment and storage medium

Through the pre-established obstacle pattern library and dynamic adjustment of window size, the problem of low efficiency of obstacle recognition algorithm in automobile assisted driving is solved, and efficient and accurate obstacle detection is achieved to meet the system's real-time needs.

CN118981018APending Publication Date: 2024-11-19FULSCIENCE AUTOMOTIVE ELECTRONICS CO LTD
View PDF 0 Cites 5 Cited by

Patent Information

Application Number
CN202411070087.X
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 existing automotive assisted driving technology, obstacle recognition algorithms require a large amount of training data in complex environments, resulting in high complexity and low efficiency in real-time computing.

Method used

Through a pre-established obstacle pattern library and dynamic adjustment of window size, the real-time calculation complexity is reduced, the execution efficiency of obstacle recognition algorithms is improved, and the detection accuracy and system real-timeness are ensured.

Benefits of technology

It realizes that, while ensuring detection accuracy, reduce unnecessary computing overhead, improve the execution efficiency of obstacle recognition algorithm, meet the real-time requirements of the system, and enhances the robustness and adaptability of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118981018A_ABST
    Figure CN118981018A_ABST
Patent Text Reader

Abstract

The invention provides an obstacle recognition method and device, electronic equipment and a storage medium, and the method comprises the steps: collecting an echo signal of a target vehicle in a detection range in real time according to an ultrasonic sensor, and determining whether a stable obstacle exists or not based on the stability and duration of the echo signal; if yes, determining a target obstacle corresponding to the echo signal based on the echo signal and a preset obstacle mode library; and dynamically adjusting the size of a window based on the motion characteristics of the target obstacle, and performing motion state analysis processing, scene correlation analysis processing and redundant data management on the target obstacle in the window so as to complete identification of the target obstacle. The complexity of real-time calculation is greatly reduced through a pre-established obstacle mode library, the execution efficiency of an obstacle recognition algorithm is improved, and unnecessary calculation overhead can be reduced and the real-time requirement of a system can be met on the premise of ensuring the detection accuracy by dynamically adjusting the window size.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

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

[0002] With the improvement of people's living standards, cars, as a common means of transportation, have become one of the necessities of life. With the development and exploration of the information industry and the intelligent field, the intelligence of cars has developed rapidly. Among them, the assisted driving technology of cars is a key research direction in this field. In the assisted driving technology of cars, detecting and identifying obstacles in the direction of the car's advance is a key technical problem faced by assisted driving of cars. The existing technology uses deep neural networks to extract and match features of ultrasonic echo data. The extracted features are matched with the pre-trained environmental map to achieve positioning. Although this method can improve the positioning accuracy in complex environments, it requires a large amount of training data. Therefore, how to reduce the complexity of real-time calculations and improve the execution efficiency of obstacle recognition algorithms has become a technical problem that cannot be underestimated. Summary of the invention

[0003] In view of this, the purpose of the present application is to provide an obstacle recognition method, device, electronic device and storage medium, which can greatly reduce the complexity of real-time calculations through a pre-established obstacle pattern library, improve the execution efficiency of the obstacle recognition algorithm, and dynamically adjust the window size to reduce unnecessary computing overhead while ensuring detection accuracy to meet the real-time requirements of the system.

[0004] The present application embodiment provides an obstacle recognition method, the obstacle recognition method comprising:

[0005] The ultrasonic sensor collects echo signals of the target vehicle in the detection range in real time, and determines whether there is a stable obstacle based on the stability and duration of the echo signals;

[0006] If yes, determining the position information of the target obstacle corresponding to the echo signal based on the echo signal and a preset obstacle pattern library;

[0007] The size of the window is dynamically adjusted based on the motion characteristics of the target obstacle, and the motion state analysis processing, scene correlation analysis processing and redundant data management of the target obstacle are performed within the window to complete the identification of the target obstacle.

[0008] In a possible implementation manner, determining whether there is a stable obstacle based on the stability and duration of the echo signal:

[0009] Detecting whether the echo signal is collected multiple times and whether the duration of the echo signal exceeds a preset time threshold;

[0010] If both are true, then there is a stable target obstacle at the echo signal;

[0011] If any one of them is no, then there is noise or interference in the echo signal.

[0012] In a possible implementation manner, determining the position information of the target obstacle corresponding to the echo signal based on the echo signal and a preset obstacle pattern library includes:

[0013] Matching the characteristic information of the echo signal with the characteristic information of obstacles at each position in a preset obstacle pattern library, and determining a similarity value between the characteristic information of the echo signal and the characteristic information of each obstacle;

[0014] Detecting whether the similarity value is greater than or equal to a preset similarity threshold;

[0015] If not, there is no obstacle feature information at the echo signal; if so, the obstacle feature information corresponding to the highest similarity value is used as the target obstacle, and the position information of the obstacle feature information corresponding to the highest similarity value is used as the position information of the target obstacle; wherein the obstacle feature information includes obstacle category information, obstacle size information, obstacle shape information and obstacle position coordinate information.

[0016] In a possible implementation manner, dynamically adjusting the size of the window based on the motion characteristics of the target obstacle includes:

[0017] Predicting the position information of the target obstacle at a next moment based on the position information and speed information of the target obstacle at a previous moment;

[0018] Based on the position information at the next moment, determining the center position of the window;

[0019] Determine a tracking period of the target obstacle based on the motion characteristics of the target obstacle; wherein the faster the motion characteristics, the shorter the tracking period, and the more stable the motion characteristics, the longer the tracking period, and dynamically adjust the tracking period factor based on the change trend of the tracking period;

[0020] The size of the window is determined based on the center position of the window, the tracking period, the tracking distance, the tracking period factor, and the tracking distance factor.

[0021] In a possible implementation manner, the performing motion state analysis processing, scene correlation analysis processing, and redundant data management on the target obstacle within the window to complete the identification of the target obstacle includes:

[0022] Performing motion state analysis on the target obstacle to determine whether the target obstacle is a new obstacle;

[0023] Performing scene relevance analysis on the target obstacle to determine whether the scene relevance of the target obstacle exceeds the observation range of the ultrasonic sensor or exceeds the moving track of the target vehicle, and if either is true, not tracking the target obstacle;

[0024] Redundant data management is performed on the target obstacle to determine whether there is redundant data for the target obstacle, and if so, the redundant data is deleted.

[0025] In a possible implementation manner, the performing motion state analysis processing on the target obstacle to determine whether the target obstacle is a new obstacle includes:

[0026] Matching the position information of the target obstacle and the motion information of the target obstacle with the position information of the historical obstacles and the motion information of the historical obstacles;

[0027] If the match is successful, the target obstacle and the historical obstacle are the same obstacle;

[0028] If the matching is unsuccessful, the target obstacle is a new obstacle.

[0029] In a possible implementation manner, after dynamically adjusting the size of the window based on the motion characteristics of the target obstacle, and performing motion state analysis processing, scene correlation analysis processing, and redundant data management on the target obstacle within the window to complete the identification of the target obstacle, the obstacle identification method further includes:

[0030] Obtaining a height detection result of the target obstacle by an adjacent ultrasonic sensor;

[0031] Detecting whether the height detection result exceeds a passable height threshold of the target vehicle;

[0032] If yes, the target vehicle needs to perform evasive action on the target obstacle;

[0033] If not, the target obstacle is marked as a traversable target.

[0034] The embodiment of the present application further provides an obstacle recognition device, the obstacle recognition device comprising:

[0035] A preliminary detection module is used to collect echo signals of the target vehicle within the detection range in real time according to the ultrasonic sensor, and determine whether there is a stable obstacle based on the stability and duration of the echo signals;

[0036] an obstacle position determination module, configured to determine, if yes, the position information of the target obstacle corresponding to the echo signal based on the echo signal and a preset obstacle pattern library;

[0037] The obstacle processing module is used to dynamically adjust the size of the window based on the motion characteristics of the target obstacle, and perform motion state analysis processing, scene correlation analysis processing and redundant data management on the target obstacle within the window to complete the identification of the target obstacle.

[0038] An embodiment of the present application also provides an electronic device, including: 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 obstacle recognition method as described above are performed.

[0039] 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 obstacle identification method as described above are executed.

[0040] The embodiment of the present application provides an obstacle identification method, device, electronic device and storage medium, the obstacle identification method includes: collecting echo signals of target vehicles in the detection range in real time according to ultrasonic sensors, determining whether there is a stable obstacle based on the stability and duration of the echo signals; if so, determining the position information of the target obstacle corresponding to the echo signal based on the echo signals and the preset obstacle pattern library; dynamically adjusting the size of the window based on the motion characteristics of the target obstacle, and performing motion state analysis processing, scene correlation analysis processing and redundant data management on the target obstacle within the window to complete the identification of the target obstacle. By pre-establishing an obstacle pattern library to greatly reduce the complexity of real-time calculations and improve the execution efficiency of the obstacle identification algorithm, the dynamic adjustment of the window size can reduce unnecessary calculation overhead while ensuring detection accuracy to meet the real-time requirements of the system.

[0041] 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

[0042] 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.

[0043] Figure 1 A flowchart of an obstacle identification method provided in an embodiment of the present application;

[0044] Figure 2 One of the structural schematic diagrams of an obstacle identification device provided in an embodiment of the present application;

[0045] Figure 3 The second structural diagram of an obstacle identification device provided in an embodiment of the present application;

[0046] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0047] 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.

[0048] First, the application scenarios to which the present application is applicable are introduced. The present application can be applied to the field of obstacle detection technology.

[0049] Research has found that with the improvement of people's living standards, cars, as a common means of transportation, have become one of the necessities of life. With the development and exploration of the information industry and the intelligent field, the intelligence of cars has developed rapidly. Among them, the assisted driving technology of cars is a key research direction in this field. In the assisted driving technology of cars, detecting and identifying obstacles in the direction of the car's advance is a key technical problem faced by assisted driving of cars. The existing technology uses deep neural networks to extract and match features of ultrasonic echo data. The extracted features are matched with the pre-trained environmental map to achieve positioning. Although this method can improve the positioning accuracy in complex environments, it requires a large amount of training data. Therefore, how to reduce the complexity of real-time calculations and improve the execution efficiency of obstacle recognition algorithms has become a technical problem that cannot be underestimated.

[0050] Based on this, an embodiment of the present application provides an obstacle recognition method, which greatly reduces the complexity of real-time calculations through a pre-established obstacle pattern library, improves the execution efficiency of the obstacle recognition algorithm, and dynamically adjusts the window size to reduce unnecessary computing overhead while ensuring detection accuracy to meet the real-time requirements of the system.

[0051] See also Figure 1 , Figure 1 This is a flow chart of an obstacle identification method provided in an embodiment of the present application. Figure 1 As shown in , the obstacle recognition method provided by the embodiment of the present application includes:

[0052] S101: Collecting echo signals of a target vehicle within a detection range in real time using an ultrasonic sensor, and determining whether there is a stable obstacle based on the stability and duration of the echo signals.

[0053] In this step, the ultrasonic sensor collects the echo signal of the target vehicle within the detection range in real time, and determines whether there is a stable obstacle based on the stability and duration of the echo signal.

[0054] Here, the stability and duration of the echo signal also need to be considered. If similar echo signals are detected multiple times in a row and last for a long time, it can be determined that there is a stable obstacle. On the contrary, if the detection results fluctuate greatly or last for a short time, it may be noise or other interference.

[0055] In a possible implementation manner, determining whether there is a stable obstacle based on the stability and duration of the echo signal:

[0056] (1): Detect whether the echo signal is collected multiple times and whether the duration of the echo signal exceeds a preset time threshold.

[0057] Here, it is detected whether the echo signal is collected multiple times and whether the duration of the echo signal exceeds a preset time threshold.

[0058] (2): If both are yes, there is a stable target obstacle at the echo signal; if any one is no, there is noise or interference at the echo signal.

[0059] Here, if the echo signal is collected multiple times and the duration of the echo signal exceeds the preset time threshold, there is a stable target obstacle at the echo signal; if either is not the case, there is noise or interference at the echo signal.

[0060] S102: If yes, determining the position information of the target obstacle corresponding to the echo signal based on the echo signal and a preset obstacle pattern library.

[0061] In this step, if yes, the position information of the target obstacle corresponding to the echo signal is determined according to the echo signal and the preset obstacle pattern library.

[0062] Among them, the obstacle pattern library is a pattern library that covers the entire detection area in advance according to the installation position and detection range of the sensor. Each pattern in the obstacle pattern library represents the obstacle characteristics at a certain position. The obstacle characteristics include the location coordinates, size, shape, category and other information of the obstacle. The obstacle pattern library includes point obstacles, open walls, single-side sealed walls and double-side sealed walls. Point obstacles have only one coordinate point information, open walls have two end point coordinates but are not sealed, single-side sealed walls have one end point coordinate plus the sealing point coordinate, and double-side sealed walls have complete two end point coordinate information.

[0063] In a possible implementation manner, determining the position information of the target obstacle corresponding to the echo signal based on the echo signal and a preset obstacle pattern library includes:

[0064] A: Match the characteristic information of the echo signal with the characteristic information of obstacles at each position in the preset obstacle pattern library, and determine the similarity value between the characteristic information of the echo signal and the characteristic information of each obstacle.

[0065] Here, the characteristic information of the echo signal is matched with the characteristic information of the obstacle at each position in the preset obstacle pattern library, and the similarity value between the characteristic information of the echo signal and the characteristic information of each obstacle is determined.

[0066] B: Detect whether the similarity value is greater than or equal to a preset similarity threshold.

[0067] Here, it is detected whether there is a similarity value greater than or equal to a preset similarity threshold.

[0068] C: If not, there is no obstacle feature information at the echo signal; if so, the obstacle feature information corresponding to the highest similarity value is used as the target obstacle, and the position information of the obstacle feature information corresponding to the highest similarity value is used as the position information of the target obstacle.

[0069] Here, if no, there is no obstacle feature information at the echo signal. If yes, the obstacle feature information corresponding to the highest similarity value is used as the target obstacle, and the position information of the obstacle feature information corresponding to the highest similarity value is used as the position information of the target obstacle.

[0070] The obstacle feature information includes obstacle category information, obstacle size information, obstacle shape information and obstacle position coordinate information.

[0071] In this application, the currently observed echo signal is compared with each pattern in the obstacle pattern library to determine the similarity between them. By finding the best matching pattern through this pattern matching, the current obstacle position can be determined. In a continuous time series, the position of the obstacle is tracked using the results of pattern matching to achieve continuous monitoring of the obstacle. The position of the obstacle can be quickly determined through the pre-established obstacle pattern library, avoiding complex mathematical calculations. At the same time, the obstacle pattern library can be dynamically adjusted according to actual conditions to improve the adaptability of the algorithm.

[0072] S103: dynamically adjusting the size of the window based on the motion characteristics of the target obstacle, and performing motion state analysis processing, scene correlation analysis processing, and redundant data management on the target obstacle within the window, so as to complete the identification of the target obstacle.

[0073] In this step, the size of the window is dynamically adjusted according to the motion characteristics of the target obstacle, and the motion state analysis and processing, scene correlation analysis and processing, and redundant data management of the target obstacle are performed within the window to complete the identification of the target obstacle.

[0074] In a possible implementation manner, dynamically adjusting the size of the window based on the motion characteristics of the target obstacle includes:

[0075] a: Predicting the position information of the target obstacle at the next moment based on the position information and speed information of the target obstacle at the previous moment.

[0076] Here, the position information of the target obstacle at the next moment is predicted based on the position information and speed information of the target obstacle at the previous moment.

[0077] b: Based on the position information at the next moment, determine the center position of the window.

[0078] Here, the position information at the next moment is used as the center position of the window.

[0079] c: Determine the tracking period of the target obstacle based on the motion characteristics of the target obstacle; wherein, the faster the motion characteristics, the shorter the tracking period, and the more stable the motion characteristics, the longer the tracking period, and dynamically adjust the tracking period factor based on the changing trend of the tracking period.

[0080] Here, the tracking period of the target obstacle is determined according to the motion characteristics of the target obstacle.

[0081] First, the position and speed information of the target obstacle at the current moment can be used to predict the position at the next moment to determine the center position of the matching window. Secondly, the determination of the window size can consider two factors: the motion characteristics of the target obstacle and the real-time requirements of the system. For targets that move faster, a larger window needs to be set to ensure that the target does not exceed the window range, while for targets that move slower, the window can be appropriately reduced to reduce the amount of calculation. At the same time, the real-time requirements of the system should also be considered, and a larger matching window should be given to more urgent targets. By dynamically adjusting the matching window size, the accuracy of the algorithm can be improved and the real-time requirements of the system can be met, thereby achieving efficient tracking of obstacles. The dynamic configuration of the tracking cycle is as follows: Different tracking strategies need to be adopted for different types of obstacle targets. For targets of types such as points (obstacles), (cylinders) or (walls), their motion characteristics and prediction difficulties are different. Therefore, it is necessary to dynamically adjust the tracking cycle according to the target type to improve the accuracy and stability of tracking. For targets that move faster and are difficult to predict, the tracking cycle should be appropriately shortened and the tracking frequency should be increased to capture the changes of the target in time. For relatively stable targets, the tracking cycle can be appropriately extended to reduce unnecessary tracking calculations and save system resources. At the same time, it is also necessary to dynamically adjust the value range of the tracking cycle factor according to the changing trend of the current tracking cycle to ensure the stability of the tracking effect. The tracking cycle factor is a parameter used to calculate the size of the matching window. The tracking cycle factor represents a coefficient that increases proportionally with the increase of the tracking cycle. The tracking cycle factor can be set in the following ways: The tracking cycle factor is initially set by configuration parameters and obtained by reading configuration parameters. According to different obstacle objects, the tracking cycle factor can be configured with different values ​​to adapt to the movement rules of different types of obstacles. The tracking cycle factor can be configured to change linearly or exponentially as the tracking cycle increases, and the increasing trend is configurable. The tracking cycle factor can set upper and lower limits to constrain its dynamic change range to avoid too fast or too slow growth. The tracking cycle factor can be calculated in real time according to the tracking cycle of the current object at runtime instead of a pre-set value. In summary, the tracking cycle factor supports multiple dynamic configurations to meet different needs.

[0082] d: Determine the size of the window based on the center position of the window, the tracking period, the tracking distance, the tracking period factor, and the tracking distance factor.

[0083] Here, the matching size = the center position of the window + the tracking period factor * the tracking period + the tracking distance factor * the tracking distance.

[0084] Among them, the tracking distance factor and the tracking distance are determined according to the motion characteristics of the target obstacle.

[0085] In a possible implementation manner, the performing motion state analysis processing, scene correlation analysis processing, and redundant data management on the target obstacle within the window to complete the identification of the target obstacle includes:

[0086] I: Perform motion state analysis on the target obstacle to determine whether the target obstacle is a new obstacle.

[0087] Here, the motion state of the target obstacle is analyzed and processed to determine whether the target obstacle is a new obstacle.

[0088] In a possible implementation manner, the performing motion state analysis processing on the target obstacle to determine whether the target obstacle is a new obstacle includes:

[0089] The position information and the motion information of the target obstacle are matched with the position information and the motion information of the historical obstacles; if the match is successful, the target obstacle and the historical obstacle are the same obstacle; if the match is unsuccessful, the target obstacle is a new obstacle.

[0090] Here, the position information and motion information of the target obstacle are matched with the position information and motion information of the historical obstacles; if the match is successful, the target obstacle and the historical obstacle are the same obstacle; if the match is unsuccessful, the target obstacle is a new obstacle.

[0091] In a specific embodiment, the detailed algorithm for achieving object matching using motion and position matching is as follows: Calculate the motion trend of the target obstacle (new OLO point) relative to the most recent historical obstacle (OLO point). Its main function is to calculate the motion trend of the new OLO point relative to a certain OLO point. The specific process of determining whether the motion of the new point matches is: (1) Calculate the displacement between the two points: Use the input two-point coordinates (new point coordinates and OLO point coordinates) to calculate the displacement between the two points in the X and Y directions; (2) Calculate the displacement size: Take the square root of the square sum of the displacement to get the distance between the two points; (3) Calculate the motion angle: Use the inverse tangent function to find the angle of the line connecting the two points; (4) Compare the motion trend: Compare the calculated motion trend (size and angle) of the new point with the historical motion record of the OLO point object. If the motion trend of the new point is similar to that of the OLO point, it is determined that the two points come from the same moving object and the motion of the new point matches the OLO point object. If the difference is large, it is determined to be mismatched. Therefore, the function of calculating the relative motion of the new point is realized in this way. (5): New and old motion matching: Motion matching is achieved by comparing the motion vector matching between the new and old objects. The matching of the motion of the new and old objects can be achieved through the following aspects: Motion direction: Calculate the motion direction of the new and old objects. If the directions are basically the same, the motion is matched. For example, if the motion direction of the old object is 30° and the motion direction of the new object is 40°, it can be seen that they are moving in the same direction. Motion speed: The motion speed of the new and old objects can differ within a certain range. If the speed is basically the same, it can be seen as matching. Motion acceleration: Compare the changes in acceleration of the new and old objects during the motion process to see if they show similar changing patterns. Motion stability: If the old object shows a stable motion state and the motion of the new object is also relatively stable, it can be considered that the motion is matched. Motion time correlation: Whether the motion of the new and old objects has temporal continuity and correlation. Motion mode judgment: Establish models of different motion modes, such as uniform motion, accelerated motion, etc., to determine whether the new and old objects conform to the same motion mode. Combining these comparison aspects, the matching of the motion of the new and old objects can be compared as a reference for determining whether they are the same obstacle.

[0092] II: Perform scene relevance analysis on the target obstacle to determine whether the scene relevance of the target obstacle exceeds the observation range of the ultrasonic sensor or exceeds the moving track of the target vehicle. If either is true, the target obstacle is not tracked.

[0093] Here, the target obstacle is subjected to scene relevance analysis to determine whether the scene relevance of the target obstacle exceeds the observation range of the ultrasonic sensor or exceeds the moving track of the target vehicle. If either is true, the target obstacle is not tracked.

[0094] Among them, for obstacles that have exceeded the sensor's observation range or have left the vehicle's trajectory, their data is no longer relevant to the current scene and should be deleted from the tracking system to release limited computing resources.

[0095] III: performing redundant data management on the target obstacle to determine whether the target obstacle has redundant data, and if so, deleting the redundant data.

[0096] Here, redundant data management is performed on the target obstacle to determine whether there is redundant data for the target obstacle. If so, the redundant data is deleted.

[0097] Among them, repeated detection data of the same obstacle at different times should be identified and deleted to avoid interference in the matching process.

[0098] In this application, through preliminary analysis of detection data, scene relevance judgment and management of redundant data, a good foundation can be laid for subsequent precise matching.

[0099] In a possible implementation manner, after dynamically adjusting the size of the window based on the motion characteristics of the target obstacle, and performing motion state analysis processing, scene correlation analysis processing, and redundant data management on the target obstacle within the window to complete the identification of the target obstacle, the obstacle identification method further includes:

[0100] Obtain the height detection result of the target obstacle by the adjacent ultrasonic sensor; detect whether the height detection result exceeds the passable height threshold of the target vehicle; if so, the target vehicle needs to perform evasive action on the target obstacle; if not, mark the target obstacle as a crossable target.

[0101] Here, in the obstacle detection process based on ultrasonic sensors, it is necessary to make full use of the detection results of surrounding neighboring sensors to improve the accuracy and reliability of detection. First, by comparing the detection results of different sensors, the repeatedly detected obstacle targets can be identified. These repeated targets may be redundant information generated by the overlapping fields of view of the sensors, and need to be deleted to avoid affecting the subsequent target tracking and prediction. Secondly, the height detection results of neighboring sensors can also be used to determine whether the currently detected obstacle can be crossed. If the obstacle height exceeds the passable height of the vehicle, it can be marked as an insurmountable target to avoid unnecessary evasive actions. In general, making full use of the detection information of surrounding sensors can effectively identify and delete redundant targets and improve the overall detection performance.

[0102] The present application has the following beneficial effects: (1) Improving detection accuracy and reliability: The location of obstacles can be quickly determined through a pre-established pattern library to avoid complex mathematical calculations. Combined with the detection data analysis of multiple sensors, duplicate targets can be identified and deleted, thereby improving detection accuracy. (2) Enhancing system robustness and adaptability: The pattern library can be dynamically adjusted according to actual conditions, thereby improving the adaptability of the algorithm. By pre-matching new and old targets, false matches can be effectively reduced and matching reliability can be improved. Dynamic adjustment of the matching window size and tracking cycle factor can adapt to the motion characteristics of different types of targets and improve the robustness of tracking. (3) System real-time and efficiency: By pre-establishing a pattern library, the complexity of real-time calculations can be greatly reduced and the execution efficiency of the algorithm can be improved. Dynamic adjustment of the matching window size can reduce unnecessary computing overhead while ensuring detection accuracy and meet the real-time requirements of the system. Differentiated tracking strategies for different target types can also help save system resources and improve overall performance.

[0103] An obstacle identification method provided by an embodiment of the present application includes: collecting echo signals of a target vehicle within a detection range in real time according to an ultrasonic sensor, determining whether there is a stable obstacle based on the stability and duration of the echo signals; if so, determining the position information of the target obstacle corresponding to the echo signal based on the echo signals and a preset obstacle pattern library; dynamically adjusting the size of the window based on the motion characteristics of the target obstacle, and performing motion state analysis processing, scene correlation analysis processing, and redundant data management on the target obstacle within the window to complete the identification of the target obstacle. By pre-establishing an obstacle pattern library to greatly reduce the complexity of real-time calculations and improve the execution efficiency of the obstacle identification algorithm, the dynamic adjustment of the window size can reduce unnecessary calculation overhead while ensuring detection accuracy to meet the real-time requirements of the system.

[0104] See also Figure 2 , Figure 3 , Figure 2 One of the structural schematic diagrams of an obstacle identification device provided in an embodiment of the present application; Figure 3 This is a second structural diagram of an obstacle identification device provided in an embodiment of the present application. Figure 2 As shown in , the obstacle identification device 200 includes:

[0105] The preliminary detection module 210 is used to collect the echo signal of the target vehicle within the detection range in real time according to the ultrasonic sensor, and determine whether there is a stable obstacle based on the stability and duration of the echo signal;

[0106] The obstacle position determination module 220 is used to determine the position information of the target obstacle corresponding to the echo signal based on the echo signal and a preset obstacle pattern library;

[0107] The obstacle processing module 230 is used to dynamically adjust the size of the window based on the motion characteristics of the target obstacle, and perform motion state analysis processing, scene correlation analysis processing and redundant data management on the target obstacle within the window to complete the identification of the target obstacle.

[0108] Further, when the preliminary detection module 210 is used to determine whether there is a stable obstacle based on the stability and duration of the echo signal, the preliminary detection module 210 is specifically used to:

[0109] Detecting whether the echo signal is collected multiple times and whether the duration of the echo signal exceeds a preset time threshold;

[0110] If both are true, then there is a stable target obstacle at the echo signal;

[0111] If any one of them is no, then there is noise or interference in the echo signal.

[0112] Further, when the obstacle position determination module 220 is used to determine the position information of the target obstacle corresponding to the echo signal based on the echo signal and the preset obstacle pattern library, the obstacle position determination module 220 is specifically used to:

[0113] Matching the characteristic information of the echo signal with the characteristic information of obstacles at each position in a preset obstacle pattern library, and determining a similarity value between the characteristic information of the echo signal and the characteristic information of each obstacle;

[0114] Detecting whether the similarity value is greater than or equal to a preset similarity threshold;

[0115] If not, there is no obstacle feature information at the echo signal; if so, the obstacle feature information corresponding to the highest similarity value is used as the target obstacle, and the position information of the obstacle feature information corresponding to the highest similarity value is used as the position information of the target obstacle; wherein the obstacle feature information includes obstacle category information, obstacle size information and obstacle shape information.

[0116] Further, when the obstacle processing module 230 is used to dynamically adjust the size of the window based on the motion characteristics of the target obstacle, the obstacle processing module 230 is specifically used to:

[0117] Predicting the position information of the target obstacle at the next moment based on the position information and speed information of the target obstacle at the previous moment;

[0118] Based on the position information at the next moment, determining the center position of the window;

[0119] Determine a tracking period of the target obstacle based on the motion characteristics of the target obstacle; wherein the faster the motion characteristics, the shorter the tracking period, and the more stable the motion characteristics, the longer the tracking period, and dynamically adjust the tracking period factor based on the change trend of the tracking period;

[0120] The size of the window is determined based on the center position of the window, the tracking period, the tracking distance, the tracking period factor, and the tracking distance factor.

[0121] Further, when the obstacle processing module 230 is used to perform motion state analysis processing, scene correlation analysis processing, and redundant data management on the target obstacle within the window so as to complete the identification of the target obstacle, the obstacle processing module 230 is specifically used to:

[0122] Performing motion state analysis on the target obstacle to determine whether the target obstacle is a new obstacle;

[0123] Performing scene relevance analysis on the target obstacle to determine whether the scene relevance of the target obstacle exceeds the observation range of the ultrasonic sensor or exceeds the moving track of the target vehicle, and if either is true, not tracking the target obstacle;

[0124] Redundant data management is performed on the target obstacle to determine whether there is redundant data for the target obstacle, and if so, the redundant data is deleted.

[0125] Furthermore, when the obstacle processing module 230 is used to perform motion state analysis on the target obstacle to determine whether the target obstacle is a new obstacle, the obstacle processing module 230 is specifically used to:

[0126] Matching the position information of the target obstacle and the motion information of the target obstacle with the position information of the historical obstacles and the motion information of the historical obstacles;

[0127] If the match is successful, the target obstacle and the historical obstacle are the same obstacle;

[0128] If the matching is unsuccessful, the target obstacle is a new obstacle.

[0129] Further, such as Figure 3As shown, the obstacle recognition device 200 further includes a proximity sensor detection module 240, and the proximity sensor detection module 240 is used to:

[0130] Obtaining a height detection result of the target obstacle by an adjacent ultrasonic sensor;

[0131] Detecting whether the height detection result exceeds a passable height threshold of the target vehicle;

[0132] If yes, the target vehicle needs to perform evasive action on the target obstacle;

[0133] If not, the target obstacle is marked as a traversable target.

[0134] The embodiment of the present application provides an obstacle identification device, which includes: a preliminary detection module, which is used to collect the echo signal of the target vehicle within the detection range in real time according to the ultrasonic sensor, and determine whether there is a stable obstacle based on the stability and duration of the echo signal; an obstacle position determination module, which is used to determine the position information of the target obstacle corresponding to the echo signal based on the echo signal and the preset obstacle pattern library; an obstacle processing module, which is used to dynamically adjust the size of the window based on the motion characteristics of the target obstacle, and perform motion state analysis processing, scene correlation analysis processing and redundant data management on the target obstacle in the window, so as to complete the identification of the target obstacle. Through the pre-established obstacle pattern library to greatly reduce the complexity of real-time calculation, improve the execution efficiency of the obstacle identification algorithm, and dynamically adjust the window size to reduce unnecessary calculation overhead while ensuring the detection accuracy to meet the real-time requirements of the system.

[0135] 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 .

[0136] 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 For the steps of the obstacle identification method in the illustrated method embodiment, the specific implementation method can be found in the method embodiment, which will not be described in detail here.

[0137] 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 The specific implementation of the steps of the obstacle identification method in the illustrated method embodiment can be found in the method embodiment, which will not be described in detail here.

[0138] 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.

[0139] 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.

[0140] 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.

[0141] 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.

[0142] 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.

[0143] 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. An obstacle recognition method, characterized in that: The obstacle recognition method comprises: The ultrasonic sensor collects echo signals of the target vehicle in the detection range in real time, and determines whether there is a stable obstacle based on the stability and duration of the echo signals; If yes, determining the position information of the target obstacle corresponding to the echo signal based on the echo signal and a preset obstacle pattern library; The size of the window is dynamically adjusted based on the motion characteristics of the target obstacle, and the motion state analysis processing, scene correlation analysis processing and redundant data management of the target obstacle are performed within the window to complete the identification of the target obstacle.

2. The obstacle recognition method according to claim 1, characterized in that: Determining whether there is a stable obstacle based on the stability and duration of the echo signal: Detecting whether the echo signal is collected multiple times and whether the duration of the echo signal exceeds a preset time threshold; If both are true, then there is a stable target obstacle at the echo signal; If any one of them is no, then there is noise or interference in the echo signal.

3. The obstacle recognition method according to claim 1, characterized in that: The determining, based on the echo signal and a preset obstacle pattern library, the position information of the target obstacle corresponding to the echo signal comprises: Matching the characteristic information of the echo signal with the characteristic information of obstacles at each position in a preset obstacle pattern library, and determining a similarity value between the characteristic information of the echo signal and the characteristic information of each obstacle; Detecting whether the similarity value is greater than or equal to a preset similarity threshold; If not, there is no obstacle feature information at the echo signal; if so, the obstacle feature information corresponding to the highest similarity value is used as the target obstacle, and the position information of the obstacle feature information corresponding to the highest similarity value is used as the position information of the target obstacle; wherein the obstacle feature information includes obstacle category information, obstacle size information, obstacle shape information and obstacle position coordinate information.

4. The obstacle recognition method according to claim 1, characterized in that: The dynamically adjusting the size of the window based on the motion characteristics of the target obstacle includes: Predicting the position information of the target obstacle at a next moment based on the position information and speed information of the target obstacle at a previous moment; Based on the position information at the next moment, determining the center position of the window; Determine a tracking period of the target obstacle based on the motion characteristics of the target obstacle; wherein the faster the motion characteristics, the shorter the tracking period, and the more stable the motion characteristics, the longer the tracking period, and dynamically adjust the tracking period factor based on the change trend of the tracking period; The size of the window is determined based on the center position of the window, the tracking period, the tracking distance, the tracking period factor, and the tracking distance factor.

5. The obstacle recognition method according to claim 1, characterized in that: The performing motion state analysis processing, scene correlation analysis processing and redundant data management on the target obstacle in the window so as to complete the identification of the target obstacle includes: Performing motion state analysis on the target obstacle to determine whether the target obstacle is a new obstacle; Performing scene relevance analysis on the target obstacle to determine whether the scene relevance of the target obstacle exceeds the observation range of the ultrasonic sensor or exceeds the moving track of the target vehicle, and if either is true, not tracking the target obstacle; Redundant data management is performed on the target obstacle to determine whether there is redundant data for the target obstacle, and if so, the redundant data is deleted.

6. The obstacle recognition method according to claim 4, characterized in that: The performing motion state analysis processing on the target obstacle to determine whether the target obstacle is a new obstacle includes: Matching the position information of the target obstacle and the motion information of the target obstacle with the position information of the historical obstacles and the motion information of the historical obstacles; If the match is successful, the target obstacle and the historical obstacle are the same obstacle; If the matching is unsuccessful, the target obstacle is a new obstacle.

7. The obstacle recognition method according to claim 1, characterized in that: After dynamically adjusting the size of the window based on the motion characteristics of the target obstacle, and performing motion state analysis processing, scene correlation analysis processing, and redundant data management on the target obstacle within the window to complete the identification of the target obstacle, the obstacle identification method further includes: Obtaining a height detection result of the target obstacle by an adjacent ultrasonic sensor; Detecting whether the height detection result exceeds a passable height threshold of the target vehicle; If yes, the target vehicle needs to perform evasive action on the target obstacle; If not, the target obstacle is marked as a traversable target.

8. An obstacle recognition device, characterized in that: The obstacle recognition device comprises: A preliminary detection module is used to collect echo signals of the target vehicle within the detection range in real time according to the ultrasonic sensor, and determine whether there is a stable obstacle based on the stability and duration of the echo signals; an obstacle position determination module, configured to determine, if yes, the position information of the target obstacle corresponding to the echo signal based on the echo signal and a preset obstacle pattern library; The obstacle processing module is used to dynamically adjust the size of the window based on the motion characteristics of the target obstacle, and perform motion state analysis processing, scene correlation analysis processing and redundant data management on the target obstacle within the window to complete the identification of the target obstacle.

9. An electronic device, 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 obstacle recognition method as described in any one of claims 1 to 7.

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

Citation Information

Cited By

  • Living body motion judgment and feature extraction method based on ultrasonic echoes

    CN120254822A

  • Obstacle avoidance method and device during water line flat scanning, computer equipment and storage medium

    CN120447536A

  • Vehicle braking control method and device, electronic equipment and vehicle

    CN120503748A

  • Vehicle braking control methods, devices, electronic equipment and vehicles

    CN120503748B

  • Obstacle laser detection method and system for unmanned mine card

    CN120742348A