A method, device, equipment and storage medium for processing false obstacles

Abstract

The present invention relates to a method, device, equipment and storage medium for processing false obstacles, and belongs to the field of radar detection technology, wherein the method comprises: establishing a multi-frame obstacle list based on multi-frame obstacle data obtained by continuously detecting obstacle data around a target vehicle; determining a suspicious obstacle in the obstacle list of a current frame based on the obstacle list; judging whether the suspicious obstacle in the current frame is a false obstacle based on the obstacle list; and deleting obstacle data corresponding to the false obstacle from the obstacle list. The present invention establishes an obstacle list based on multi-frame obstacle data continuously detected around a target vehicle, and by analyzing obstacle data in obstacle lists in different frames, judges false obstacles existing in the obstacle list, and deletes the false obstacle data, so that the actual obstacles around the target vehicle can be accurately determined, and the surrounding conditions of the target vehicle can be understood.

Description

Technical Field

[0001] The present invention relates to the field of radar detection technology, and in particular to a method, device, equipment and storage medium for processing false obstacles. Background Art

[0002] As autonomous driving technology matures, autonomous driving will become an important means of travel in the future. The functional modules of the current autonomous driving system mainly include perception module, positioning module, planning and decision module, control module, high-precision map module and cloud computing module. The core of the autonomous driving system is to accurately perceive the surrounding environment and make correct judgments and responses to detected obstacles.

[0003] Millimeter-wave radar has become an important part of the perception link of autonomous driving due to its advantages such as high distance resolution, robustness to bad weather, and ability to provide speed information. The existing autonomous driving vehicle positioning technology solution mainly uses millimeter-wave radar to judge the attributes of obstacles with the help of prior information of point cloud maps or grid maps. Millimeter-wave radar, with its high resolution, strong anti-interference ability, and unaffected by light, can effectively improve the long-term stable positioning of autonomous driving vehicles in typical urban conditions such as urban canyons and GNSS signal obstruction, and determine the surrounding conditions of the vehicle.

[0004] However, millimeter-wave radar has some limitations, such as ground reflection and clutter interference. The positioning accuracy is also limited by the quality of the laser point cloud at the current moment. When there are a large number of dynamic obstacles in the laser point cloud, false obstacles will appear. The inability to eliminate the false obstacles will lead to deviations in the detection results of the vehicle's surroundings and fail to accurately reflect the conditions around the vehicle. Summary of the invention

[0005] In view of this, it is necessary to provide a method, device, equipment and storage medium for processing false obstacles to solve the problem in the prior art that false obstacles appear when the millimeter-wave radar detects the vehicle's surrounding environment, resulting in deviations in the detection results of the vehicle's surrounding environment and failure to accurately reflect the vehicle's surrounding conditions.

[0006] In order to solve the above problems, the present invention provides a method for processing false obstacles, which is used to detect obstacles around a target vehicle based on a millimeter wave radar, comprising:

[0007] Establishing a multi-frame obstacle list according to multi-frame obstacle data obtained by continuously detecting obstacle data around the target vehicle;

[0008] Determine the suspicious obstacles in the obstacle list of the current frame based on the obstacle list;

[0009] Based on the obstacle list, determine whether the suspicious obstacle in the current frame is a false obstacle;

[0010] Delete the obstacle data corresponding to the false obstacle from the obstacle list.

[0011] In a possible implementation, a multi-frame obstacle list is established based on multi-frame obstacle data obtained by continuously detecting obstacle data around the target vehicle, including:

[0012] Set initial obstacle parameters for multiple frames of obstacle data;

[0013] Numbering multiple frames of obstacle data obtained through continuous detection to obtain obstacle numbers;

[0014] A multi-frame obstacle list is established according to the obstacle number, multi-frame obstacle data and obstacle parameters.

[0015] In a possible implementation, the obstacle parameter includes the number of loops; determining a suspicious obstacle in the obstacle list of the current frame based on the obstacle list includes:

[0016] At the beginning of the loop, increase the loop count by one;

[0017] Determine whether the obstacle is stationary relative to the target vehicle and whether the obstacle is within the suspected range of the target vehicle according to the obstacle data;

[0018] When the obstacle is stationary relative to the target vehicle and is within the preset range of the target vehicle, the obstacle is a suspicious obstacle.

[0019] In a possible implementation, the obstacle data includes obstacle speed and obstacle coordinates; judging whether the obstacle is stationary relative to the target vehicle and the obstacle is within a suspicious range of the target vehicle according to the obstacle data includes:

[0020] When the absolute value of the difference between the speed of the obstacle and the speed of the target vehicle is less than a preset speed threshold, the obstacle is stationary relative to the target vehicle;

[0021] When the obstacle coordinates are within the preset range, the obstacle is within the suspicious range of the target vehicle.

[0022] In a possible implementation, the obstacle parameter includes a life cycle; before determining whether a suspicious obstacle in a current frame is a false obstacle based on the obstacle list, including:

[0023] When the difference between the number of cycles and the life cycle corresponding to all obstacles in the current frame exceeds the preset value, the obstacle list is cleared.

[0024] In a possible implementation, the obstacle parameter includes the accumulated number of times that the obstacle position does not change; judging whether the suspicious obstacle in the current frame is a false obstacle based on the obstacle list includes:

[0025] If the obstacle coordinates of the suspected obstacle in the current frame are the same as those in the previous frame, the accumulated number is increased by one, and the number of cycles is assigned to the life cycle;

[0026] If the accumulated number of suspicious obstacles in the current frame is greater than the preset accumulated number threshold, the suspicious obstacle is a false obstacle.

[0027] In a possible implementation, judging whether a suspicious obstacle in the current frame is a false obstacle based on the obstacle list further includes:

[0028] If the obstacle coordinates of the suspected obstacle in the current frame are different from those in the previous frame, the accumulated times are set to zero;

[0029] If the accumulated number of suspicious obstacles in the current frame is not greater than the preset accumulated number threshold, the suspicious obstacle is a normal obstacle.

[0030] The present invention also provides a device for processing false obstacles, comprising:

[0031] A table building module is used to build a multi-frame obstacle list based on multi-frame obstacle data obtained by continuously detecting obstacle data around the target vehicle;

[0032] A suspicious determination module, used to determine the suspicious obstacles in the obstacle list of the current frame based on the obstacle list;

[0033] A false judgment module is used to judge whether a suspicious obstacle in the current frame is a false obstacle based on the obstacle list;

[0034] The deletion module is used to delete the obstacle data corresponding to the false obstacle from the obstacle list.

[0035] In a third aspect, the present invention further provides a false obstacle processing device, comprising a memory and a processor, wherein:

[0036] Memory, used to store programs;

[0037] The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the lithium-ion battery health status prediction method in any of the above implementations.

[0038] In a fourth aspect, the present invention further provides a computer-readable storage medium for storing computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the lithium-ion battery health status prediction method in any of the above-mentioned implementation modes.

[0039] The beneficial effect of the present invention is as follows: a method for processing false obstacles provided by the present invention first establishes a multi-frame obstacle list according to multi-frame obstacle data obtained by continuously detecting obstacle data around a target vehicle by a millimeter-wave radar, wherein the multi-frame obstacle list contains obstacle data of different frames, and then analyzes the obstacles with the obstacle data in the obstacle lists of different frames to determine suspicious obstacles, and then further analyzes the suspicious obstacles to find out false obstacles, and removes the obstacle data corresponding to the detected false obstacles, retains the normal obstacle data, thereby confirming the true detection result of the vehicle's surrounding environment, so as to accurately reflect the driving conditions around the target vehicle and reduce safety hazards during driving. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic diagram of a flow chart of an embodiment of a method for processing false obstacles provided by the present invention;

[0041] Figure 2 The present invention provides Figure 1 A schematic flow chart of an embodiment of step S101;

[0042] Figure 3 The present invention provides Figure 1 A schematic diagram of a flow chart of an embodiment of step S102;

[0043] Figure 4 A schematic diagram of a scene of an embodiment of an obstacle and a target vehicle provided by the present invention;

[0044] Figure 5 A schematic structural diagram of an embodiment of a device for processing false obstacles provided by the present invention;

[0045] Figure 6 A schematic diagram of the structure of a false obstacle processing device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0047] Possible reasons why the mmWave radar on a vehicle may detect false obstacles are as follows:

[0048] First of all, the working principle of millimeter wave radar is to measure the position, direction of movement and speed of an object based on the transmission and reception of millimeter waves. When the beam emitted by the millimeter wave radar hits an object, a reflected signal is generated. After the reflected signal is received by the radar, the relevant information of the object is obtained through calculation. However, in some cases, other factors in the environment may cause the radar to receive reflected signals from non-target objects, thereby misjudging them as obstacles.

[0049] Secondly, the parameter settings and calibration of the radar system may also affect its detection of obstacles. If the radar sensitivity is set too high or the calibration is inaccurate, it may misjudge normal objects or noise in the environment as obstacles.

[0050] In addition, the structure and material of the vehicle itself may also affect the detection of millimeter-wave radar. For example, the front bumper or other metal parts of the vehicle may refract and reflect the electromagnetic waves emitted and received by the radar, causing the radar to receive incorrect reflected signals and misjudge them as obstacles.

[0051] Finally, the radar data processing algorithm may also have the possibility of misjudgment. If the algorithm cannot accurately distinguish between real obstacles and false obstacles, then misjudgment will occur.

[0052] See also Figure 1 , Figure 1 A flowchart of an embodiment of a method for processing false obstacles provided by the present invention is provided. A specific embodiment of the present invention discloses a method for processing false obstacles, which is used to detect obstacles around a target vehicle based on a millimeter-wave radar, including:

[0053] S101, establishing a multi-frame obstacle list according to multi-frame obstacle data obtained by continuously detecting obstacle data around a target vehicle;

[0054] S102, determining a suspicious obstacle in the obstacle list of the current frame based on the obstacle list;

[0055] S103, judging whether the suspicious obstacle in the current frame is a false obstacle based on the obstacle list;

[0056] S104: Delete obstacle data corresponding to the false obstacle from the obstacle list.

[0057] In the above embodiment, the obstacle data around the target vehicle is continuously detected by the millimeter wave radar, and multiple frames of different obstacle data can be obtained. The obstacle data may include the position (i.e., coordinates), speed, size and other information of the obstacle. The established obstacle list contains all the information of the obstacle data corresponding to each frame, and sets the initial obstacle parameters for each obstacle.

[0058] Specifically, millimeter wave radar mainly detects and images targets by emitting millimeter wave signals and receiving echoes. Millimeter wave radar uses the Doppler effect to extract information about target motion, and can obtain the target's position information and image through signal processing algorithms. Millimeter wave radar has the characteristics of weak penetration, strong reflection ability, and short wavelength. Therefore, it is widely used in many fields, such as human security inspection and respiratory monitoring. The wavelength range of millimeter wave radar is 1 to 10 mm. The corresponding frequency range is 30 to 300 GHz. Millimeter wave radar is a radar that works in the millimeter wave band (millimeter wave) for detection. Its ranging principle is the same as that of ordinary radar, that is, radio waves are sent out, and then echoes are received, and the target's position data is measured based on the time difference between sending and receiving.

[0059] In addition, millimeter-wave radar can distinguish and identify very small targets, and can identify multiple targets at the same time, has imaging capabilities, and is small in size, has good mobility and concealment, and has strong survivability on the battlefield. Commonly used millimeter-wave radar frequency bands are 24GHz, 60GHz, and 77GHz. Among them, the wavelength of 24GHz is 1.25 cm, but the industry still calls it millimeter wave; the wavelength of 60GHz is 5 mm; the wavelength of 77GHz is 3.9 mm.

[0060] The detected obstacles may be other cars, motorcycles, bicycles, pedestrians, animals, construction obstacles, temporary roadblocks, etc. on the road. In addition to the obstacle information detected above, the shape, size and other attributes of the obstacle can also be determined.

[0061] In the millimeter wave radar detection system, the detection results are expressed in frames. Each frame refers to a complete data set, which contains all obstacle information collected by the radar in a specific time period. Each frame of data represents the radar scanning result at a moment, which contains the position, speed and other relevant information of all obstacles detected in the surrounding environment at that moment.

[0062] Millimeter wave radar continuously transmits and receives millimeter waves at a very high frequency. By processing these reflected signals, the radar can generate a series of data frames. Each frame of data includes all obstacle data detected in the current scanning cycle. These data are usually digital and can be easily stored, transmitted and processed.

[0063] The obstacle data detected by millimeter-wave radar can be stored in a variety of ways, such as memory cache, database storage, file storage and cloud storage. Among them, the memory cache method is suitable for real-time processing because the data can be read directly from the memory without reading from the disk or other slow storage devices; the database storage method is for data that needs to be stored for a long time or for post-analysis, which can be stored in the database. The database can provide a structured data storage method and support efficient query and analysis operations. The obstacle data can be stored in the database in the form of records, and each record contains the obstacle ID, location, speed, timestamp and other fields; the file storage method is to store the obstacle Object data is saved on disk in the form of files, which can be achieved by converting each frame of data into a specific file format (such as CSV, XML, JSON or a custom binary format). The advantage of file storage is its high flexibility, which allows for easy data exchange and sharing with other systems. At the same time, by using compression algorithms, it can effectively reduce storage space usage. Cloud storage is for data that requires remote access or distributed processing. You can consider using a cloud storage solution. By uploading obstacle data to the cloud, data backup, sharing and cross-platform access can be achieved. Cloud storage also provides elastic expansion capabilities, and can adjust the size and performance of storage space as needed.

[0064] A false obstacle is an obstacle that does not exist due to the error of millimeter-wave radar detection. That is, the obstacle does not actually exist but exists in the millimeter-wave radar detection result. The false obstacle will affect the normal driving of the target vehicle. Therefore, it is necessary to perform a cyclic analysis based on the obstacle list. If an obstacle meets one or more suspicious conditions, it will be marked as a suspicious obstacle. The suspicious obstacles, that is, the obstacles suspected to be false obstacles, are first determined from the obstacle list.

[0065] After the suspicious obstacles are identified, further analysis is performed on the obstacle data corresponding to the suspicious obstacles in different frames to determine which obstacles are truly false obstacles. The corresponding obstacle parameters are adjusted according to the changes in the obstacle data, and the obstacles are analyzed and judged based on the obstacle parameters to determine the false obstacles.

[0066] Once false obstacles are identified, they need to be marked in the obstacle list for subsequent removal. This can be done by adding a "real / false" flag to each obstacle, or by creating a separate list to store the IDs of false obstacles.

[0067] Next, traverse the entire obstacle list, and for each obstacle in the list, check whether it has been marked as false. If an obstacle is marked as false, then delete it.

[0068] After deleting the false obstacles, update the obstacle list to ensure that it only contains data for real obstacles. After deleting the false obstacles, verify that the obstacle list is accurate. Make sure that no real obstacles are deleted by mistake and no false obstacles are left. The performance of the system can be verified through simulation tests or actual scene tests. Record the process and results of deleting false obstacles for subsequent analysis and improvement. If the system frequently encounters false obstacles, it may be necessary to further optimize the obstacle detection algorithm or improve the performance of the sensor.

[0069] Compared with the prior art, the present embodiment provides a method for processing false obstacles. First, a multi-frame obstacle list is established according to multi-frame obstacle data obtained by continuously detecting obstacle data around a target vehicle by a millimeter-wave radar. The multi-frame obstacle list includes obstacle data of different frames. Then, the obstacle data in the obstacle lists of different frames are used to analyze the obstacles to determine suspicious obstacles. Then, the suspicious obstacles are further analyzed to find out false obstacles. The obstacle data corresponding to the detected false obstacles are removed, and the normal obstacle data is retained, thereby confirming the true detection result of the vehicle's surrounding environment, so as to accurately reflect the driving conditions around the target vehicle and reduce safety hazards during driving.

[0070] The obstacle list can be used as part of the data record for post-analysis and system optimization. By analyzing a large amount of historical data, the perception algorithm can be improved, the recognition accuracy can be increased, and the system performance can be optimized.

[0071] See also Figure 2 , Figure 2 The present invention provides Figure 1 In some embodiments of the present invention, a multi-frame obstacle list is established based on multi-frame obstacle data obtained by continuously detecting obstacle data around the target vehicle, including:

[0072] S201, setting initialization obstacle parameters for multiple frames of obstacle data;

[0073] S202, numbering multiple frames of obstacle data obtained by continuous detection to obtain obstacle numbers;

[0074] S203: Create a multi-frame obstacle list according to the obstacle number, multi-frame obstacle data and obstacle parameters.

[0075] In the above embodiment, the obstacle parameters include the life cycle (age) of the obstacle; the number of cycles (duration) of the obstacle; the cumulative number of obstacles whose positions do not change (err_count). Initializing the obstacle parameters means setting all the obstacle parameters to zero.

[0076] The obstacle number is obtained by numbering the multiple frames of obstacle data obtained by continuous detection in order to determine the unique identification of each obstacle, that is, the obstacle ID, so as to track the same obstacle in the multiple frames of obstacle data.

[0077] Different obstacle lists should be established for obstacle data in different frames, but the obstacle parameters and obstacle data therein need to be adjusted. Establishing an obstacle list based on obstacle numbers, multi-frame obstacle data and obstacle parameters is to provide a reliable data basis for subsequent obstacle tracking, decision-making and planning.

[0078] For each obstacle, try to match it with an obstacle in the previous frame. This can be done by comparing the obstacle's position, speed, size, etc. If a match is successfully found with an obstacle in the previous frame, keep the original number of the obstacle and update its position, speed, etc. If a match cannot be found with any obstacle in the previous frame, assign a new unique number to the current obstacle and add it to the list of obstacle numbers.

[0079] See also Figure 3 , Figure 3 The present invention provides Figure 1 FIG. 1 is a flow chart of an embodiment of step S102 in FIG. 1 . In some embodiments of the present invention, the obstacle parameter includes the number of loops; determining a suspicious obstacle in the obstacle list of the current frame based on the obstacle list includes:

[0080] S301, adding one to the number of loops at the beginning of the loop;

[0081] S302, judging whether the obstacle is stationary relative to the target vehicle and whether the obstacle is within the suspected range of the target vehicle according to the obstacle data;

[0082] S303: When the obstacle is stationary relative to the target vehicle and is within a preset range of the target vehicle, the obstacle is a suspicious obstacle.

[0083] In the above embodiment, when looping through the obstacle list to determine a suspicious obstacle from the obstacle list of the current frame, it is first necessary to increase the corresponding loop count in the obstacle list by one, in order to track the number of loops and ensure that each obstacle is checked, which helps to prevent an infinite loop or ensure that all obstacles are correctly evaluated.

[0084] Only when the obstacle is stationary relative to the target vehicle and within the preset range of the target vehicle, can the obstacle be marked as a suspicious obstacle, which means that it may pose a potential threat to the driving of the target vehicle and requires further attention or processing, and subsequent risk assessment or decision-making is carried out based on this.

[0085] In some embodiments of the present invention, the obstacle data includes the obstacle speed and the obstacle coordinates; determining whether the obstacle is stationary relative to the target vehicle and whether the obstacle is within the suspicious range of the target vehicle according to the obstacle data includes:

[0086] When the absolute value of the difference between the obstacle speed and the speed of the target vehicle is less than a preset speed threshold, the obstacle is stationary relative to the target vehicle;

[0087] When the obstacle coordinates are within a preset range, the obstacle is within the suspicious range of the target vehicle.

[0088] In the above embodiment, as a preferred embodiment, the preset speed threshold is set to 1 m / s. That is to say, when the absolute value of the difference between the obstacle speed and the speed of the target vehicle is less than 1 m / s, the difference between the obstacle speed and the speed of the target vehicle is very small, and it can be considered that the obstacle is stationary relative to the target vehicle.

[0089] Taking the target vehicle at each frame as the origin, a rectangular coordinate system is established. The x-axis direction is longitudinal, positive in the front and negative in the back, the y-axis direction is transverse, positive on the left and negative on the right, and the z-axis direction is vertically oriented, positive on the top and negative on the bottom. The height condition is not used for the time being. The coordinates of the left side of the obstacle are relative to the target vehicle. As a preferred embodiment, the suspicious range is defined as -6 < x < 4 and -3 < y < 3.

[0090] It should be noted that the preset speed threshold and the suspicious range can be set according to actual needs, and the present invention does not further limit this.

[0091] In some embodiments of the present invention, the obstacle parameters include the life cycle; before determining whether the suspicious obstacle in the current frame is a false obstacle based on the obstacle list, it includes:

[0092] When the difference between the loop count corresponding to all obstacles in the current frame and the life cycle exceeds a preset value, clear the obstacle list.

[0093] In the above embodiment, as a preferred embodiment, the preset value is set to 50. When processing the obstacle data of consecutive frames, each obstacle has its life cycle, that is, the time from when it is first detected until it no longer appears in the field of view. If an obstacle is not processed in multiple consecutive frames, this may mean that the obstacle is no longer within the effective range of the current scene, or it has not been correctly tracked due to some reasons (such as sensor failure, data processing delay, etc.). In this case, in order to maintain the accuracy and real-time nature of the obstacle list, it can be considered to clear the entire obstacle list. This can avoid potential problems caused by old data or incorrect data and make room for obstacles newly entering the field of view.

[0094] In some embodiments of the present invention, the obstacle parameter includes the accumulated number of times that the obstacle position does not change; judging whether the suspicious obstacle in the current frame is a false obstacle based on the obstacle list includes:

[0095] If the obstacle coordinates of the suspected obstacle in the current frame are the same as those in the previous frame, the accumulated number is increased by one, and the number of cycles is assigned to the life cycle;

[0096] If the accumulated number of suspicious obstacles in the current frame is greater than the preset accumulated number threshold, the suspicious obstacle is a false obstacle.

[0097] In the above embodiment, by comparing whether the obstacle coordinates corresponding to the obstacle ID in the current frame are the same as the obstacle coordinates corresponding to the obstacle ID in the previous frame, it is determined whether the obstacle position has not changed, and the accumulated number of times the obstacle position has not changed is increased by one. The obstacle is within the effective range of the current scene, and the number of cycles needs to be assigned to the life cycle.

[0098] As a preferred embodiment, the preset cumulative number threshold is 10, and the preset cumulative number threshold is used to determine how many consecutive frames an obstacle remains stationary or unchanged in position before it can be determined as a false obstacle. The preset cumulative number threshold in the present invention can be set according to actual needs, and the present invention does not impose further restrictions on this.

[0099] In some embodiments of the present invention, judging whether a suspicious obstacle in the current frame is a false obstacle based on the obstacle list further includes:

[0100] If the obstacle coordinates of the suspected obstacle in the current frame are different from those in the previous frame, the accumulated times are set to zero;

[0101] If the accumulated number of suspicious obstacles in the current frame is not greater than the preset accumulated number threshold, the suspicious obstacle is a normal obstacle.

[0102] In the above embodiment, for each suspicious obstacle in the current frame, it is first necessary to check whether its coordinates are different from the corresponding obstacle coordinates in the previous frame. If the coordinates have changed, it means that the obstacle is moving, so it should not be judged as a false obstacle. If the coordinates of the suspicious obstacle in the current frame are different from those in the previous frame, it means that the obstacle is moving, and the accumulated number should be set to zero, because it may be a normal, moving obstacle.

[0103] If the coordinates of the suspected obstacle in the current frame are the same as those in the previous frame or the changes are within an acceptable range, keep increasing the number of accumulations. Then, check whether the number of accumulations exceeds the preset number of accumulations threshold. If not, then the obstacle should be considered a normal obstacle.

[0104] If the accumulated number exceeds the threshold, the obstacle is determined to be a false obstacle and can be removed from the list or processed in other ways.

[0105] In practical applications, obstacle matching may involve more complex logic, especially when there are a large number of obstacles and they may occlude or be close to each other. In addition, for obstacle matching between consecutive frames, it may be necessary to consider using more advanced algorithms, such as the Hungarian algorithm (used to solve the assignment problem) or feature-based matching methods.

[0106] See also Figure 4 , Figure 4 A schematic diagram of a scene of an embodiment of an obstacle and a target vehicle provided by the present invention. In a specific embodiment of the present invention, the target vehicle is traveling at a speed of 25 m / s. The target vehicle is currently traveling in the middle lane. At this time, the millimeter-wave radar detects three obstacles around it.

[0107] At the first frame:

[0108] Obstacle 1 is a false target, and the obstacle coordinate information (x / y) remains unchanged. The information in the first frame is: id=1, x=2m, y=1.5m, v=24.8m / s.

[0109] Obstacle 2 is a real target, and the obstacle coordinate information (x / y) changes. The information in the first frame is: id=2, x=-10m, y=4m, V=24.5m / s.

[0110] Obstacle 3 is a real target, and the obstacle coordinate information (x / y) changes. The information in the first frame is: id=3, x=-4m, y=1.5m, V=25.5m / s.

[0111] Construct a list of suspicious obstacles, store obstacles 1, 2, and 3 in the list, and assign the information to the list synchronously. At the same time, initialize the age, duration, and err_count parameters of the three obstacles to 0, as shown in Table 1 below:

[0112] Table 1: Initial obstacle list in the first frame

[0113] First Frame Obstacle 1 Obstacle 2 Obstacle 3 id 1 2 3 × 2 -10 -4 y 1.5 4 1.5 v 24.9 24.9 25.6 age 0 0 0 duration 0 0 0 err_count 0 0 0

[0114] When entering the loop, the duration of all obstacles is increased by one. At this time, the obstacle list is shown in Table 2 below:

[0115] Table 2: Obstacle list at the beginning of the first frame loop

[0116] First Frame Obstacle 1 Obstacle 2 Obstacle 3 id 1 2 3 x 2 -10 -4 y 15 4 1.5 v 24.9 24.9 25.6 age 0 0 0 duration 1 1 1 err_count 0 0 0

[0117] The speeds of the three obstacles satisfy |obstacle speed - target vehicle speed| < 1m / s, and they are all relatively stationary with respect to the target vehicle. Obstacles 1 and 3 are within the suspicious range. Obstacle 2 is outside the suspicious range.

[0118] Obstacles 1 and 3 proceed to the next step.

[0119] Obstacle 2 is not a suspicious target and is marked as a normal obstacle.

[0120] If the duration-age of all data in the current frame is not greater than 50, the obstacle list will not be cleared and the judgment will continue.

[0121] In the second frame:

[0122] Obtain millimeter-wave radar obstacle data, and determine that the three obstacles with IDs 1 / 2 / 3 are in the list of suspected obstacles based on the obstacle IDs, then proceed to the next step.

[0123] The duration of all obstacles entering the loop is increased by 1. The obstacle list at this time is shown in Table 3 below:

[0124] Table 3 Obstacle list at the beginning of the second frame loop

[0125] Second frame Obstacle 1 Obstacle 2 Obstacle 3 id 1 2 3 x 2 -11 -4.2 y 1.5 4 1.5 v 24.9 24.9 25.6 age 0 0 0 duration 2 2 2 err_count 0 0 0

[0126] The speeds of the three obstacles satisfy |obstacle speed - target vehicle speed| < 1m / s, and they are all relatively stationary with respect to the target vehicle. Obstacles 1 and 3 are within the suspicious range. Obstacle 2 is outside the suspicious range.

[0127] Obstacles 1 and 3 proceed to the next step.

[0128] Obstacle 2 is not a suspicious target and is marked as a normal obstacle.

[0129] If the duration-age of all data in the current frame is not greater than 50, the obstacle list will not be cleared and the judgment will continue.

[0130] The x and y values ​​of obstacle 1 are the same as in the previous frame. err_count is incremented by 1, as shown in Table 4 below, and the next step is continued.

[0131] Table 4 List of suspicious obstacles in the second frame

[0132] Second frame Obstacle 1 Obstacle 2 Obstacle 3 id 1 2 3 x 2 -11 -4.2 y 1.5 4 1.5 v 24.9 24.9 25.6 age 0 0 0 duration 2 2 2 err_count 1 0 0

[0133] The x and y values ​​of obstacles 2 and 3 are different from those in the previous frame, indicating that the obstacles are normal obstacles.

[0134] At this time, if obstacle 1err_count is equal to 1 and not greater than 10, it means that the obstacle is a normal obstacle.

[0135] By analogy, suppose that at the 12th frame, the err_count of obstacle 1 is 11, which is greater than 10. At this time, obstacle 1 can be judged as a false obstacle and can be deleted.

[0136] In order to better implement the method for processing false obstacles in the embodiment of the present invention, based on the method for processing false obstacles, correspondingly, please refer to Figure 5 , Figure 5 This is a schematic diagram of the structure of an embodiment of a device for processing a false obstacle provided by the present invention. The embodiment of the present invention provides a device for processing a false obstacle 500, including:

[0137] A table building module 501 is used to build a multi-frame obstacle list according to the multi-frame obstacle data obtained by continuously detecting the obstacle data around the target vehicle;

[0138] A suspicious determination module 502, configured to determine a suspicious obstacle from a current frame obstacle list based on the obstacle list;

[0139] A false determination module 503, used to determine whether a suspicious obstacle in the current frame is a false obstacle based on the obstacle list;

[0140] The deletion module 504 is used to delete the obstacle data corresponding to the false obstacle from the obstacle list.

[0141] It should be noted here that: the device 500 provided in the above embodiment can implement the technical solutions described in the above method embodiments. The specific implementation principles of the above modules or units can refer to the corresponding contents in the above method embodiments, which will not be repeated here.

[0142] See also Figure 6 , Figure 6 A schematic diagram of the structure of a false obstacle processing device provided by an embodiment of the present invention. Based on the above-mentioned false obstacle processing method, the present invention also provides a false obstacle processing device, which can be a computing device such as a mobile terminal, a desktop computer, a notebook, a PDA, and a server. The false obstacle processing device 600 includes a processor 601, a memory 602, and a display 603. Figure 6 Only some components of the false obstacle processing device are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0143] In some embodiments, the memory 602 may be an internal storage unit of the processing device 600 of the false obstacle, such as a hard disk or memory of the processing device 600 of the false obstacle. In other embodiments, the memory 602 may also be an external storage device of the processing device 600 of the false obstacle, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the processing device 600 of the false obstacle. Further, the memory 602 may also include both an internal storage unit and an external storage device of the processing device 600 of the false obstacle. The memory 602 is used to store application software and various data of the processing device 600 for installing the false obstacle, such as the program code of the processing device 600 for installing the false obstacle. The memory 602 may also be used to temporarily store data that has been output or is to be output. In one embodiment, the memory 602 stores a processing program 604 for a false obstacle, which may be executed by the processor 601, thereby realizing the processing method for a false obstacle in each embodiment of the present application.

[0144] In some embodiments, the processor 601 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 602, such as executing a method for processing false obstacles.

[0145] In some embodiments, the display 603 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 603 is used to display information on the false obstacle processing device 600 and to display a visual user interface. The components 601-603 of the false obstacle processing device 600 communicate with each other through a system bus.

[0146] In one embodiment, when the processor 601 executes the false obstacle processing program 604 in the memory 602 , the steps in the false obstacle processing method described above are implemented.

[0147] This embodiment further provides a computer-readable storage medium, on which a false obstacle processing program is stored. When the false obstacle processing program is executed by a processor, the following steps are implemented:

[0148] Establishing a multi-frame obstacle list according to multi-frame obstacle data obtained by continuously detecting obstacle data around the target vehicle;

[0149] Determine the suspicious obstacles in the obstacle list of the current frame based on the obstacle list;

[0150] Based on the obstacle list, determine whether the suspicious obstacle in the current frame is a false obstacle;

[0151] Delete the obstacle data corresponding to the false obstacle from the obstacle list.

[0152] In summary, the present embodiment provides a method for processing false obstacles. First, a multi-frame obstacle list is established according to multi-frame obstacle data obtained by continuously detecting obstacle data around a target vehicle by a millimeter-wave radar. The multi-frame obstacle list contains obstacle data of different frames. Then, the obstacle data in the obstacle lists of different frames are used to analyze the obstacles to determine suspicious obstacles. The suspicious obstacles are further analyzed to find out false obstacles. The obstacle data corresponding to the detected false obstacles are removed, and the normal obstacle data is retained, thereby confirming the true detection result of the vehicle's surrounding environment, so as to accurately reflect the driving conditions around the target vehicle and reduce safety hazards during driving.

[0153] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0154] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for processing false obstacles, used for detecting obstacles around a target vehicle based on a millimeter-wave radar, characterized in that: include: Establishing a multi-frame obstacle list according to multi-frame obstacle data obtained by continuously detecting obstacle data around the target vehicle, wherein the obstacle list includes obstacle numbers, multi-frame obstacle data and obstacle parameters, wherein the obstacle parameters include the number of cycles, the life cycle and the accumulated number of times the obstacle position does not change, and the obstacle data includes obstacle coordinates; Determining a suspicious obstacle in the obstacle list of the current frame based on the obstacle list includes: adding one to the number of loops at the beginning of a loop; judging whether the obstacle is stationary relative to the target vehicle and the obstacle is within a suspicious range of the target vehicle according to the obstacle data; when the obstacle is stationary relative to the target vehicle and within a preset range of the target vehicle, the obstacle is the suspicious obstacle; Determining whether the suspected obstacle in the current frame is a false obstacle based on the obstacle list includes: if the obstacle coordinates of the suspected obstacle in the current frame are the same as those in the previous frame, adding one to the accumulated number and assigning the cycle number to the life cycle; if the accumulated number of the suspected obstacle in the current frame is greater than a preset accumulated number threshold, the suspected obstacle is a false obstacle; The obstacle data corresponding to the false obstacle is deleted from the obstacle list.

2. The method for processing false obstacles according to claim 1, characterized in that: The step of establishing a multi-frame obstacle list according to the multi-frame obstacle data obtained by continuously detecting the obstacle data around the target vehicle includes: Setting initialization obstacle parameters for the multiple frames of obstacle data; Numbering multiple frames of obstacle data obtained through continuous detection to obtain obstacle numbers; A multi-frame obstacle list is established according to the obstacle number, the multi-frame obstacle data and the obstacle parameters.

3. The method for processing false obstacles according to claim 1, characterized in that: The obstacle data includes obstacle speed; and judging whether the obstacle is stationary relative to the target vehicle and the obstacle is within the suspicious range of the target vehicle according to the obstacle data includes: When the absolute value of the difference between the speed of the obstacle and the speed of the target vehicle is less than a preset speed threshold, the obstacle is stationary relative to the target vehicle; When the obstacle coordinates are within a preset range, the obstacle is within the suspicious range of the target vehicle.

4. The method for processing false obstacles according to claim 1, characterized in that: Before judging whether the suspected obstacle in the current frame is a false obstacle based on the obstacle list, the method includes: When the difference between the number of cycles corresponding to all obstacles in the current frame and the life cycle exceeds a preset value, the obstacle list is cleared.

5. The method for processing false obstacles according to claim 1, characterized in that: The determining whether the suspected obstacle in the current frame is a false obstacle based on the obstacle list further includes: If the obstacle coordinates of the suspected obstacle in the current frame are different from those in the previous frame, the accumulated times are set to zero; If the accumulated number of times of the suspicious obstacle in the current frame is not greater than the preset accumulated number threshold, the suspicious obstacle is a normal obstacle.

6. A device for processing false obstacles, characterized in that: include: A table building module, used to build a multi-frame obstacle list according to multi-frame obstacle data obtained by continuously detecting obstacle data around the target vehicle, wherein the obstacle list includes obstacle numbers, multi-frame obstacle data and obstacle parameters, wherein the obstacle parameters include the number of cycles, the life cycle and the accumulated number of times the obstacle position does not change, and the obstacle data includes obstacle coordinates; A suspicious determination module, for determining a suspicious obstacle in the obstacle list of the current frame based on the obstacle list, comprising: adding one to the number of loops at the beginning of a loop; judging whether the obstacle is stationary relative to the target vehicle and the obstacle is within a suspicious range of the target vehicle according to obstacle data; when the obstacle is stationary relative to the target vehicle and within a preset range of the target vehicle, the obstacle is the suspicious obstacle; a false judgment module, used for judging whether the suspected obstacle in the current frame is a false obstacle based on the obstacle list, including: if the obstacle coordinates of the suspected obstacle in the current frame are the same as those in the previous frame, then increasing the accumulated number by one, and assigning the cycle number to the life cycle; if the accumulated number of the suspected obstacle in the current frame is greater than a preset accumulated number threshold, then the suspected obstacle is a false obstacle; The deleting module is used to delete the obstacle data corresponding to the false obstacle from the obstacle list.

7. A false obstacle processing device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the method for processing false obstacles as described in any one of claims 1 to 5 above.

8. A computer-readable storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps of the method for processing false obstacles described in any one of claims 1 to 5 above.

Images