A target recognition system and method based on millimeter wave radar
By introducing target speed verification, ID hold, stop processing, afterimage processing, false target elimination and flicker processing modules into the millimeter-wave radar target recognition system, the millimeter-wave radar is easily lost and horizontal target insensitive in urban roads. The problem of undetectable static targets is achieved, and high-precision mobile target detection and tracking is achieved.
Patent Information
- Application Number
- CN202311787483.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-22
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2043-12-22
AI Technical Summary
In urban roads, millimeter-wave radar has shortcomings such as low-speed targets being easily lost, lateral targets are insensitive, static targets cannot be detected, and target flickering, making it difficult to achieve high-precision identification of mobile targets in digital road construction.
A target recognition system based on millimeter wave radar is designed, and the data front-end processing system of the target speed verification module, the target ID holding module, the target stop processing module, the target afterimage processing module, the fake target elimination module and the target flicker processing module are designed. Carry out multiple rounds of data filtering and processing to ensure the accuracy and accuracy of the detection data.
It realizes high-precision locking and tracking of all moving targets in the millimeter-wave radar detection area, maintains consistency of the same target ID, eliminates the push of false radar targets, avoids frequent flash target pushes and abnormal target residues, and ensures the accuracy of detection data. .
Smart Images

Figure CN117970306B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of wireless communications, and in particular, to a target recognition system and method based on millimeter wave radar. Background Art
[0002] With the increasing number of digital road construction projects, the detection of moving targets on urban roads has become a major problem that has long plagued digital road construction. Due to the limitations of camera recognition, radars have been introduced into digital road construction. After a comprehensive comparison of the cost-effectiveness of lidar and millimeter-wave radar, millimeter-wave radar has become the preferred device for massive equipment deployment. However, due to the limitations of millimeter-wave radar technology, when used on urban roads, there are disadvantages such as easy loss of low-speed targets, insensitivity to lateral targets, failure to detect stationary targets, and target flickering. Summary of the invention
[0003] The purpose of the embodiments of the present application is to provide a target recognition system and method based on millimeter wave radar, so as to solve the above-mentioned problems existing in the prior art and to identify moving targets with high accuracy.
[0004] In a first aspect, a target recognition system based on millimeter wave radar is provided, which may include: a target speed verification module and a target ID holding module;
[0005] The target speed verification module is used to respectively obtain a first position, a first speed, a first acceleration, and a second position of a first moving target at a first moment; calculate the first position, the second position, and the time difference between the first moment and the second moment to determine a first initial speed; determine a second initial speed based on the first acceleration, the first speed, and the time difference; determine a second speed at the second moment from the first initial speed and the second initial speed according to a configured speed determination algorithm; wherein the first position at the first moment is obtained by a first radar, the second position at the second moment is obtained by a second radar, and the first moment is less than the second moment;
[0006] The target ID holding module is used to determine the predicted position at the second moment based on the first speed, the first acceleration, the first position and the first heading angle of the first moving target, and match the second position of the second moving target collected by the second radar at the second moment with the predicted position. If the match is successful, the target ID of the first moving target is determined as the target ID of the second moving target.
[0007] In a possible implementation, the system further includes: a target stop processing module;
[0008] The target stop processing module is used to determine that when a moving target stops during the process of the target radar collecting detection data of the moving target, the target distance of the moving target from the target radar is within a configured reference distance, and if the moving target moves again within a first preset time period after the stop time, the moving target is determined to be a stop-and-hold vehicle; so that the target radar collects detection data of the stop-and-hold vehicle after it moves again according to the target ID before the stop-and-hold vehicle stops.
[0009] In a possible implementation, the target stop processing module is also used to, when the moving target stops during the process of the target radar collecting detection data of the moving target, determine that the target is a stop and hold vehicle if the target distance of the moving target from the target radar is not within the reference distance, and if the moving target moves again within a second preset time period after the stop time; so that the target radar collects the detection data of the stop and hold vehicle after it moves again according to the target ID before the moving target stops.
[0010] In a possible implementation, the second preset duration is determined as follows:
[0011] The detection distance and the distance parameter are calculated using a preset duration algorithm to obtain a second preset duration, wherein the detection distance is the farthest distance that the target radar can detect.
[0012] In a possible implementation, the system further includes: a target afterimage processing module;
[0013] The target afterimage processing module is used for determining that the moving target is a stopped vehicle when the target radar collects detection data of the moving target and the target distance of the moving target from the target radar is within a reference distance, if the stopping time of the moving target exceeds a first preset time, so that the target radar discards the target ID of the stopped vehicle;
[0014] Furthermore, when the mobile target stops during the process of the target radar collecting detection data of the mobile target, and it is determined that the target distance of the mobile target from the target radar is not within the reference distance, if the stop time of the mobile target exceeds a second preset time, the mobile target is determined to be a stopped vehicle, so that the target radar discards the target ID of the stopped vehicle.
[0015] In a possible implementation, the system further includes: a false target elimination module;
[0016] The false target elimination module is used to mark the initial moment when the target radar first collects any moving target. When the initial moment is delayed by M seconds, if the target radar cannot collect the detection data of the corresponding moving target, the corresponding moving target is determined to be an assumed target, and the assumed target is a moving target that disappears when the initial moment is delayed by M seconds.
[0017] In a possible implementation, the system further includes: a target flicker processing module;
[0018] The target flicker processing module is used to process the target number of mobile targets collected by the target radar at the current moment. If the target number is not greater than the number threshold, the target radar is controlled to collect the corresponding detection data of all mobile targets; if the target number is greater than the number threshold, the target radar is controlled to package the detection data of the stopped vehicle to obtain a stop data identifier.
[0019] In a second aspect, a target recognition method based on millimeter wave radar is provided, which may include:
[0020] The target speed verification module obtains a first position, a first speed, a first acceleration and a second position of a first moving target at a first moment and a second moment respectively;
[0021] Calculate the time difference between the first position, the second position, the first moment and the second moment to determine a first initial velocity;
[0022] determining a second initial velocity based on the first acceleration, the first velocity, and the time difference;
[0023] Determine, according to a configured speed determination algorithm, a second speed at the second moment from the first initial speed and the second initial speed; wherein the first position at the first moment is acquired by the first radar, the second position at the second moment is acquired by the second radar, and the first moment is less than the second moment;
[0024] The method of determining that the first moving target acquired by the first radar and the second moving target acquired by the second radar are the same moving target is:
[0025] The target ID holding module determines the predicted position at the second moment based on the first speed, the first acceleration, the first position and the first heading angle of the first moving target, matches the second position of the second moving target collected by the second radar at the second moment with the predicted position, and if the match is successful, determines that the first moving target and the second moving target are the same moving target, and determines the target ID of the first moving target as the target ID of the second moving target.
[0026] In a third aspect, an electronic device is provided, the electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;
[0027] Memory, used to store computer programs;
[0028] The processor is used to implement any method step described in the second aspect when executing the program stored in the memory.
[0029] In a fourth aspect, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, any method step described in the second aspect is implemented.
[0030] The target recognition system based on millimeter wave radar provided by the present application, according to the characteristics of millimeter wave transmission, is subjected to multiple rounds of data filtering and processing of the detection data output by the millimeter wave radar chip through the data front-end processing system of the target stop processing module, the target afterimage processing module, the target speed verification module, the target ID holding module, the target flicker processing module, and the false target elimination module, to filter out inaccurate, interfering, residual and other problematic data, and finally obtain more accurate detection data. Through the target recognition and detection scheme of the present application, it is possible to achieve high-precision locking and tracking of all mobile targets in the millimeter wave radar detection area, real-time consistency of the same target ID, elimination of radar false target push, no frequent flashing target push, no abnormal target residue, no abnormal speed push, etc. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application 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.
[0032] Figure 1 A system architecture diagram of a target recognition system based on millimeter-wave radar provided in an embodiment of the present application;
[0033] Figure 2 A schematic diagram of a flow chart of a target recognition method based on millimeter wave radar provided in an embodiment of the present application;
[0034] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0035] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0036] With the increasing number of digital road construction projects, the detection of moving targets on urban roads has become a major problem that has long plagued digital road construction. Due to the limitations of camera recognition, radar equipment has been introduced into digital road construction. After a comprehensive comparison of the cost-effectiveness of lidar and millimeter-wave radar, millimeter-wave radar has become the preferred equipment for massive equipment deployment. However, due to the limitations of millimeter-wave radar technology, when used on urban roads, there are disadvantages such as easy loss of low-speed targets, insensitivity to lateral targets, failure to detect stationary targets, and target flickering.
[0037] Therefore, this application proposes a target recognition system based on millimeter-wave radar to solve the above-mentioned problems. The millimeter-wave radar can realize high-precision recognition and detection of mobile targets in complex urban road networks and diverse environments of human and machine mobile targets.
[0038] The preferred embodiments of the present application are described below in conjunction with the drawings in the specification. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application. In addition, the embodiments and features in the embodiments of the present application may be combined with each other if there is no conflict.
[0039] Figure 1 A system architecture diagram of a target recognition system based on millimeter-wave radar provided in this application; Figure 1 As shown, the system may include: a target speed verification module, a target ID holding module, a target stop processing module, a target afterimage processing module, a target elimination module, a target flicker processing module, a left and right turn recognition module and a time high-precision synchronization module;
[0040] It should be noted that the present application also provides a massive equipment debugging system that works in conjunction with the target recognition system. The radars mentioned in the present application are all millimeter wave radars.
[0041] Specifically, A. Target speed verification module is used to respectively obtain the first position, first speed, first acceleration and second position of the first moving target in the test area at the first moment and the second moment; calculate the first position, the second position and the time difference between the first moment and the second moment to determine the first initial speed. Among them, the first position at the first moment is obtained by the first radar, the second position at the second moment is obtained by the second radar, the first moment is less than the second moment, and the first moment can also be understood as a historical moment relative to the second moment. The collection object of the first radar and the collection object of the second radar are to associate the collection objects of the two radars at different moments to determine that the two radars are collecting the same object.
[0042] And, based on the first acceleration, the first velocity and the time difference, a second initial velocity is determined.
[0043] Further, according to the configured speed determination algorithm, a second speed at the second moment is determined from the first initial speed and the second initial speed.
[0044] For example, the data collected by the first radar at the first moment includes: the first speed V t-1 , the first acceleration A t-1 , the longitude and latitude coordinates of the first position J t-1 (Long1, Lat1);
[0045] The data collected by the second radar at the second moment include: the longitude and latitude coordinates J of the second position t (Long2, Lat2); that is to say, the data collected by the second radar at the second moment is unverified data.
[0046] Then, according to the latitude and longitude calculation formula, the distance moved by the moving target between the first moment and the second moment is Distance(J t-1 , J t ). The first initial speed is determined based on the moving distance and time difference (moving duration).
[0047] It should be noted that the distance between two points is calculated based on Vincenty's formulae approximate algorithm, Distance = ACOS (COS (RADIANS (90-Lat1)) * COS (RADIANS (90-Lat2)) +
[0048] SIN(RADIANS(90-Lat1))*SIN(RADIANS(90-Lat2))*
[0049] COS(RADIANS(Long2-Long1)))*6371000.
[0050] And, according to the first acceleration, the first speed and the time difference, determine the second initial speed V t and the second acceleration is A t .
[0051] Afterwards, a speed determination algorithm is used to determine the optimal speed between the first initial speed and the second initial speed, and the optimal speed is used as the second speed at the second moment.
[0052] The speed determination algorithm is:
[0053]
[0054] Among them, V t is the second initial velocity, T is the time difference, A t-1 is the first acceleration, V t ' is the second speed;
[0055] The process of speed determination algorithm can be understood as:
[0056] When speed deviation Greater than the preset speed deviation (X 阈 ), then determine the first initial velocity The second speed.
[0057] When speed deviation Not greater than the preset speed deviation (X 阈 ), then determine the second initial speed (V t ) is the second speed.
[0058] The preset speed deviation can be set according to actual conditions, and the default value is 50%.
[0059] In this manner, a verified and relatively accurate second velocity at the second moment can be obtained, and further, an accurate second acceleration At can be obtained.
[0060] B. A target ID holding module, used to determine a predicted position at a second moment based on obtaining a first speed, a first acceleration, a first position and a first heading angle of the first moving target.
[0061] Afterwards, matching the second position of the second moving target acquired by the second radar at the second moment with the predicted position;
[0062] If the match is successful, it indicates that the first mobile target and the second mobile target are the same mobile target, and the target ID of the first mobile target is determined as the target ID of the second mobile target.
[0063] In order to more accurately match the relationship between the moving targets, after the predicted position is successfully matched with the second position, the property features of the moving targets collected by each radar can be matched to further determine whether the first moving target and the second moving target are the same moving target. For example, the point cloud features of the height (or width, or the height and width combined to form the vehicle model data, or the line segment point cloud features corresponding to the vehicle) of each moving vehicle collected by each radar are matched to determine that the moving vehicles are the same vehicle when the first height point cloud and the second height point cloud are the same.
[0064] C. Target stop processing module, used for the mobile target to stop during the process of the target radar collecting the detection data of the mobile target, and when it is determined that the target distance of the mobile target from the target radar is within the configured reference distance, if the mobile target moves again within the first preset time length, then the mobile target is determined to be a stop-and-hold vehicle; so that the target radar collects the detection data of the stop-and-hold vehicle after it moves again according to the target ID before the stop-and-hold vehicle stops. For example, vehicle D stops at a distance of 30m from the target radar (reference distance is 50m), vehicle D stops for 60s (the first preset time length is 120s), and then vehicle D drives again. At this time, the target radar collects data on the vehicle according to vehicle D.
[0065] Also, when the target radar collects the detection data of the moving target and the moving target stops, and it is determined that the target distance of the moving target from the target radar is not within the reference distance, if the stopping time of the moving target moves again within the second preset time, the moving target is determined to be a stopped and held vehicle; so that the target radar collects the detection data of the stopped and held vehicle after moving again according to the target ID before the moving target stops. In this case, if a moving target stops at the far end of the target radar, a corresponding holding process will also be performed, but the holding time (the second preset time) will not be too long. Since the detection rate of the far end target is lower than that of the near end, it prevents the target from leaving but not being detected by the radar and generating a large number of afterimages. For example: Vehicle A stops at a distance of 55m (reference distance is 50m) from the target radar, and vehicle D stops for 50s (the second preset time is 60s), and then vehicle D drives again. At this time, the target radar collects data for the vehicle according to vehicle D.
[0066] When the target radar detects the parking direction at the far end, a parking retention algorithm will be implemented, and it will also be processed based on the queuing strategy. The specific retention time is inversely proportional to the distance between the target and the radar, and is not a fixed value.
[0067] Using a preset duration determination algorithm to determine a preset duration, wherein the preset duration includes a first preset duration and a second preset duration;
[0068] The preset duration determination algorithm is:
[0069] Td=S>S 阈 ? (L / C):X
[0070] Among them, Td is the preset time, L is the detection distance, C is the distance parameter, S 阈 is the reference distance, and S is the target distance. The detection distance is the maximum distance that the target radar can detect. The distance parameter is generally set according to the actual situation and is generally set to 5;
[0071] The process of determining the preset duration algorithm can be understood as follows:
[0072] When the target distance is greater than the reference distance, (L / C) is determined as the preset time length, ie, the second preset time length. In other words, the preset time length algorithm is Td=L / C.
[0073] When the target distance is not greater than the reference distance, X is set to be a preset time length, that is, a first preset time length, and the first preset time length is generally set to 120 seconds by default.
[0074] This method can be understood as a target stop processing mechanism. Since millimeter-wave radar works based on the principle of Doppler effect, it inherently filters stationary targets, so millimeter-wave radar does not have the ability to detect static targets. However, when we use millimeter-wave radar to detect moving targets, moving targets often stop. The conventional processing is to put the judgment logic on the upper platform, and the sensor only provides simple data based on the principle. However, this method involves a large number of devices and will seriously affect the response efficiency of the system. This method proposes to place the judgment of target stop at the front end of the millimeter-wave radar for processing to improve the intelligence of the terminal device.
[0075] D. Target afterimage processing module, used for when the target radar collects the detection data of the moving target and the moving target stops, and it is determined that the target distance of the moving target from the target radar is not within the reference distance, if the stopping time of the moving target exceeds the second preset time, the moving target is determined to be a stopped vehicle, so that the target radar discards the target ID of the stopped vehicle. For example, vehicle A stops at a distance of 55m from the target radar (the reference distance is 50m), and vehicle A stops for 65 (the first preset time is 60s). At this time, when the stopping time of vehicle A reaches 60s, the vehicle A identification of the vehicle and all previous detection data of vehicle A are discarded. Afterwards, if the vehicle drives again, at this time, the target radar collects data for the vehicle according to the new vehicle identification.
[0076] And, when the target radar collects the detection data of the mobile target and the mobile target stops, and it is determined that the target distance of the mobile target from the target radar is within the reference distance, if the stop time of the mobile target exceeds the first preset time, the mobile target is determined to be a stopped vehicle, so that the target radar discards the target ID of the stopped vehicle. For example, vehicle D stops at a distance of 30m (reference distance is 50m) from the target radar, and vehicle D stops for 130s (the first preset time is 120s). At this time, when the stop time of vehicle D reaches 120s, the vehicle D identification of the vehicle and all previous detection data of vehicle D are discarded. Afterwards, if the vehicle drives again, at this time, the target radar collects data for the vehicle according to the new vehicle identification.
[0077] In this method, when a moving object stops (or other situations) and leaves at a certain moment, but the radar does not detect the moving target, the actual moving target has left, but the radar will continue to report a target information, which is manifested as a stuck target in the massive equipment debugging system, referred to as "afterimage". The above processing method is designed to compensate for the target sticking problem that may be caused by the "target stop processing mechanism".
[0078] The target afterimage processing module works together with the target stop processing module to obtain the optimal parking time according to the distance between the moving high target and the radar, which can improve the accuracy of the radar detection data.
[0079] E. False target elimination module, used to mark the initial moment when the target radar first collects any mobile target. When the initial moment is delayed by M seconds, if the target radar cannot collect the detection data of the corresponding mobile target, the corresponding mobile target is determined to be a hypothetical target, and the hypothetical target is a mobile target that disappears when the initial moment is delayed by M seconds. If the target radar can collect the detection data of the corresponding mobile target, it indicates that the mobile target is a mobile target for which detection data can be collected. This process can be understood as a judgment mechanism for the first time of the mobile target. For example: the target radar first detects the mobile target at 15:48:30 on December 19, 2023. After a delay of Ms (2s), that is, at 15:48:32, it can still detect the mobile target, indicating that the mobile target is a mobile target for which detection data can be collected. When the mobile target is not detected at 15:48:32, the mobile target is a false target (it does not exist itself), and there is no need to report the detection data of the mobile target detected at 15:48:30.
[0080] Due to interference from the millimeter-wave radar chip itself or possible external environment, it may occasionally detect a false target, which will disappear after a short period of reporting. It should be noted that when the same moving target stops and stays in the future, there is no need to perform a judgment mechanism at the first time.
[0081] F. Target flicker processing module, used to process the target number of moving targets collected by the target radar at the current moment. If the target number is not greater than the number threshold, the target radar is controlled to collect the corresponding detection data of all moving targets; if the target number is greater than the number threshold, the target radar is controlled to package the detection data of the corresponding stopped vehicles in the target number to obtain the stop data identifier.
[0082] That is to say, first determine the number of moving targets currently collected by the target radar. When the number of targets is small (a threshold value can be set internally), all the detection data will be reported; when the number of targets exceeds the set threshold, the vehicle data will be stopped and streamlined, and the original speed, acceleration, longitude and latitude, heading angle and other information data will be streamlined into a retention mark, but not discarded. Through the above judgment and simplification, the length of data transmission can be reduced and information loss can be guaranteed.
[0083] When the millimeter-wave radar reports data, the detection data contained in some data frames within the same second is incomplete, causing the test platform to display the moving target as a "flickering" phenomenon that appears and disappears from time to time. The problem of incomplete data frames is caused by the incomplete collection of data reported by the internal firmware of the millimeter-wave radar. In order to reduce the load of data processing, the business data of some frames of data in multiple frames within 1 second are trimmed. This method can solve the "flickering" phenomenon of the moving target.
[0084] The target flicker processing module can ensure the integrity of target data while reducing the data pressure in multi-target situations.
[0085] G. Left and right turn recognition module: used in specific areas. The millimeter-wave radar is preset with a specific left and right turn judgment algorithm to improve the detection rate and accuracy of the millimeter-wave radar for the left and right turn actions of moving targets.
[0086] Among them, specific areas refer to community entrances and exits, traffic intersections, number of lanes, lane attributes at intersections (straight lanes, left turn lanes, right turn lanes, etc.), U-turn signs, etc.
[0087] Millimeter-wave radars are insensitive to stationary and low-speed targets, so their ability to identify left and right turns is very low. This method relies on vehicle position information and combines the actual conditions of the real road to make a comprehensive judgment on the left and right turns of moving targets. Therefore, a high-precision map is collected for the global road conditions detected by the millimeter-wave radar, and a series of road information such as community entrances and exits, traffic intersections, number of lanes, lane attributes at intersections (straight lanes, left turn lanes, right turn lanes, etc.), and U-turn signs are entered into the millimeter-wave radar so that the millimeter-wave radar can determine the left and right turns of moving targets in a specific area.
[0088] H. High-precision time synchronization module is used to deploy more than two PTP timing servers in the network of the target recognition system and ensure time synchronization at the ns level. The clock system of each millimeter-wave radar is set to the PTP timing scheme, and the path delay of the timing server is periodically determined internally to select the optimal timing path. This ensures that its own time is synchronized with the central timing time at the ns level, thereby ensuring that the time synchronization between each millimeter-wave radar and between the millimeter-wave radar and the massive equipment commissioning system is at the ns level.
[0089] The target recognition system provided by this application is based on the characteristics of millimeter wave transmission. Through the data front-end processing system of the target stop processing module, the target afterimage processing module, the target speed verification module, the target ID retention module, the target flicker processing module, and the false target elimination module, the detection data output by the millimeter wave radar chip is subjected to multiple rounds of data filtering and processing to filter out inaccurate, interfering, residual and other problematic data, and finally obtain more accurate detection data. Through the target recognition and detection scheme of this application, it is possible to achieve high-precision locking and tracking of all mobile targets in the millimeter wave radar detection area, real-time consistency of the same target ID, elimination of radar false target push, no frequent flashing target push, no abnormal target residue, no abnormal speed push, etc.
[0090] In some embodiments, since the current road testing method uses a small-scale calibration of a single device or a single intersection device, the debugging system of radar equipment in a large range is not practical enough and inefficient, and cannot accurately identify overlaps or blind spots across radar monitoring areas. This application builds a three-dimensional high-precision map scene as the scene base for a massive equipment debugging system. It does not rely on point cloud data or the ability to identify stationary targets, and is applicable to both millimeter-wave radar and lidar calibration.
[0091] The massive equipment debugging system can specifically include: high-precision map scene base map module, roadside equipment high-precision positioning module, high-precision lane acquisition module, target detection data acquisition module, high-precision time synchronization module, millimeter-wave radar coverage range detection module, high-precision ID fusion module, and real-time perception presentation module.
[0092] The debugging method applied to the massive equipment debugging system may specifically include:
[0093] Step S1: Build a three-dimensional high-precision map scene through drone oblique photography or access to urban CIM data as the scene base for the massive equipment commissioning system.
[0094] Step S2: The device point location data reported by the high-precision positioning module of the roadside equipment is loaded into the scene base to achieve a one-to-one matching between the radar and the detected road, which serves as the basis for later determining the coverage range of each millimeter-wave radar.
[0095] Step S3: The centimeter-level lane lines are loaded into the scene base through the high-precision lane acquisition module as the precise line of the road to accurately determine the relative position of the moving target.
[0096] Step S4: The target detection data acquisition module reports the millimeter wave radar detection data (including accurate timestamp, target ID, speed, longitude and latitude, etc.) to the scene for real-time rendering.
[0097] Step S5: Introduce a high-precision network time synchronization system (PTP) to control the time synchronization error of the scene base and all targets attached to the base to the microsecond level, ensuring the time and space synchronization of the targets. High-precision time synchronization is the premise and guarantee of ID fusion work.
[0098] Step S6: Through the millimeter-wave radar coverage detection module, the effective coverage range of each radar is accurately identified, and the overlapping area range between multiple millimeter-wave radars is reasonably set, providing a technical prerequisite for ID fusion in the overlapping area between multiple radars, thereby turning the overlapping area into a fusion area.
[0099] Step S7: The high-precision ID fusion module uses overlapping area selection, high-precision time synchronization, and real-time collection of mobile target detection data to fuse multiple IDs of the same target on different radars to obtain a global unique ID for the target.
[0100] Step S8: The real-time perception presentation module uses the detection data (fusion ID, speed, longitude and latitude, etc.) obtained after processing by the high-precision ID fusion module to realize real-time presentation of the moving target in the scene base.
[0101] Step S9: Present the detection data of the moving target of a single millimeter-wave radar in the scene base (single-point presentation), and use the big data algorithm, combined with the lane lines and video surveillance as the true value precise size, to reversely verify the accuracy of the millimeter-wave radar's detection data.
[0102] Step S10: Use the final verified detection data to present (fused presentation) in the scene base, and use the big data algorithm, combined with lane lines and video surveillance as the true value precise size, to reversely verify the accuracy of the millimeter wave radar detection data.
[0103] Step S11: using the reverse verification data presented at a single point and presented in a fusion manner as correction parameters of the millimeter-wave radar, and optimizing the internal firmware or configuration parameters of the millimeter-wave radar to achieve high-precision detection of the entire massive equipment debugging system.
[0104] The massive equipment commissioning system of the present application, by establishing a high-precision three-dimensional simulated urban area, maps the massive equipment in the area into a three-dimensional virtual scene in a high-precision positioning manner, and then presents the massive radar-sensed targets in real time in the scene. By introducing roadside monitoring video to verify the accuracy of radar observation targets, the efficiency of actual scene and detection data comparison is improved. Focus on abnormal data (such as speed and trajectory mismatch, lane abnormality, target residue, quantity abnormality, etc.). Through dot positioning, ranging and other mapping functions, the equipment commissioning work such as detection distance, coverage, and detection direction of the equipment can be quickly realized. Through free and flexible perspective adjustment, a macro to detail multi-perspective observation method is provided, which can realize accurate observation of a single intersection or a single section and global perception of perspective. Through the detection data of millimeter-wave radar, a twin world is established for the massive mobile targets on the actual road in the remote digital platform, realizing the one-to-one correspondence between the actual world and the digital world, and intuitively realizing the visualization of detection data, which can greatly improve the commissioning speed of massive millimeter-wave radar systems.
[0105] Simply put, the target recognition system is the front-end system, and the massive equipment debugging system is the back-end debugging system. The two systems work together to systematically optimize the city's massive millimeter-wave radar detection data, improve the data quality of end users, and provide technical support for data quality improvement in digital road construction.
[0106] The system working process of the target recognition system and the massive equipment debugging system is as follows:
[0107] The scene base of the millimeter-wave radar mass equipment debugging system is built, in which single millimeter-wave radar is tuned, IDs of multiple millimeter-wave radars are fused, and the global scene is visualized.
[0108] High-precision map data is introduced into the millimeter-wave radar, and a series of road information such as community entrances and exits, traffic intersections, number of lanes, lane attributes at intersections (straight lanes, left turn lanes, right turn lanes, etc.), and U-turn signs are entered into the millimeter-wave radar.
[0109] In specific areas, the millimeter-wave radar is pre-installed with a specific left or right turn judgment algorithm to improve the detection rate and accuracy of the millimeter-wave radar for the left or right turn movements of moving targets.
[0110] The verification process of the millimeter-wave radar detecting the speed of the moving target is a supplement to the conventional speed extraction of the front-end equipment, which can obtain the speed at the corresponding moment more accurately.
[0111] Stop-and-hold and afterimage elimination work together to obtain the optimal stop-and-hold value through an algorithm based on the distance between the target and the radar, effectively improving the radar's accuracy in judging target stop events.
[0112] The target recognition system works in conjunction with the massive equipment debugging system to effectively improve the availability of roadside millimeter-wave radar on complex urban roads.
[0113] This application can be used in scenarios such as massive equipment tuning in digital road construction, performance tuning of roadside millimeter-wave radars, research and development and production of high-precision millimeter-wave radars, holographic intersections, intelligent transportation, cloud control platforms, and vehicle-road collaborative unmanned bus routes. It can improve the detection accuracy of millimeter-wave radars, increase the application scope of millimeter-wave radars, and solve the problem of low accuracy in target detection in urban and complex traffic scenarios.
[0114] Figure 2 A flow chart of a target recognition method based on millimeter wave radar provided in an embodiment of the present application is as follows: Figure 2 As shown, the method may include:
[0115] Step S210: The target speed verification module obtains the first position, the first speed, the first acceleration at the first moment and the second position at the second moment of the first moving target respectively.
[0116] Step S220: Calculate the time difference between the first position, the second position, the first moment and the second moment to determine a first initial speed.
[0117] Step S230: determining a second initial speed based on the first acceleration, the first speed and the time difference.
[0118] Step S240: Determine a second speed at a second moment from the first initial speed and the second initial speed according to the configured speed determination algorithm.
[0119] The first position at the first moment is obtained by the first radar, the second position at the second moment is obtained by the second radar, and the first moment is smaller than the second moment.
[0120] The method of determining that the first moving target acquired by the first radar and the second moving target acquired by the second radar are the same moving target is:
[0121] The target ID holding module determines the predicted position at the second moment based on the first speed, the first acceleration, the first position and the first heading angle of the first moving target, matches the second position of the second moving target collected by the second radar at the second moment with the predicted position, and if the match is successful, determines that the first moving target and the second moving target are the same moving target, and determines the target ID of the first moving target as the target ID of the second moving target.
[0122] The present application also provides an electronic device, such as Figure 3 As shown, it includes a processor 310 , a communication interface 320 , a memory 330 and a communication bus 340 , wherein the processor 310 , the communication interface 320 , and the memory 330 communicate with each other via the communication bus 340 .
[0123] Memory 330, for storing computer programs;
[0124] The processor 310 is used to execute the program stored in the memory 330 to implement the following steps:
[0125] The target speed verification module obtains a first position, a first speed, a first acceleration at a first moment, and a second position at a second moment of the first moving target respectively.
[0126] The first position, the second position, and the time difference between the first moment and the second moment are calculated to determine a first initial speed.
[0127] A second initial velocity is determined based on the first acceleration, the first velocity, and the time difference.
[0128] According to the configured speed determination algorithm, the second speed at the second moment is determined from the first initial speed and the second initial speed; wherein the first position at the first moment is obtained by the first radar, the second position at the second moment is obtained by the second radar, and the first moment is less than the second moment.
[0129] The method of determining that the first moving target acquired by the first radar and the second moving target acquired by the second radar are the same moving target is:
[0130] The target ID holding module determines the predicted position at the second moment based on the first speed, the first acceleration, the first position and the first heading angle of the first moving target, matches the second position of the second moving target collected by the second radar at the second moment with the predicted position, and if the match is successful, determines that the first moving target and the second moving target are the same moving target, and determines the target ID of the first moving target as the target ID of the second moving target.
[0131] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.
[0132] The communication interface is used for communication between the above electronic device and other devices.
[0133] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.
[0134] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
[0135] Since the implementation methods and beneficial effects of the various components of the electronic device in the above embodiments to solve the problems can be achieved by referring to the various steps in the embodiments, the specific working process and beneficial effects of the electronic device provided in the embodiments of the present application are not repeated here.
[0136] In another embodiment provided in the present application, a computer-readable storage medium is also provided, in which instructions are stored. When the computer-readable storage medium is executed on a computer, the computer executes a target recognition method based on millimeter-wave radar as described in any of the above embodiments.
[0137] In another embodiment provided in the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute a target recognition method based on millimeter-wave radar as described in any one of the above embodiments.
[0138] Those skilled in the art will appreciate that the embodiments in the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt a complete hardware embodiment, a complete software embodiment, or a form of an embodiment combining software and hardware. Moreover, the present application may adopt a form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0139] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0140] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0141] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0142] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they are aware of the basic creative concepts. Therefore, the present application embodiments are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application embodiments.
[0143] Obviously, those skilled in the art can make various changes and modifications to the embodiments in the present application without departing from the spirit and scope of the embodiments in the present application. Thus, if these modifications and variations of the embodiments in the present application are within the scope of the embodiments in the present application and their equivalents, the embodiments in the present application are also intended to include these modifications and variations.
Claims
1. A target recognition system based on millimeter wave radar, characterized in that: The system comprises: a target speed verification module and a target ID holding module; The target speed verification module is used to respectively obtain a first position, a first speed, a first acceleration, and a second position of a first moving target at a first moment; calculate the first position, the second position, and the time difference between the first moment and the second moment to determine a first initial speed; determine a second initial speed based on the first acceleration, the first speed, and the time difference; determine a second speed at the second moment from the first initial speed and the second initial speed according to a configured speed determination algorithm; wherein the first position at the first moment is obtained by a first radar, the second position at the second moment is obtained by a second radar, and the first moment is less than the second moment; The target ID holding module is used to determine the predicted position at the second moment according to the first speed, the first acceleration, the first position and the first heading angle of the first moving target, match the second position of the second moving target collected by the second radar at the second moment with the predicted position, and if the match is successful, determine the target ID of the first moving target as the target ID of the second moving target; The system further comprises: a target stop processing module; The target stop processing module is used for determining that the moving target is a stop-and-hold vehicle if the moving target stops during the process of the target radar collecting the detection data of the moving target and the target distance of the moving target from the target radar is within a configured reference distance, and if the stop duration of the moving target moves again within a first preset duration, so that the target radar collects the detection data of the stop-and-hold vehicle after it moves again according to the target ID before the stop-and-hold vehicle stops; The target stop processing module is also used for, when the moving target stops during the process of the target radar collecting the detection data of the moving target, and it is determined that the target distance of the moving target from the target radar is not within the reference distance, if the moving target moves again within a second preset time period after the stop time, then determining that the moving target is a stop and hold vehicle; so that the target radar collects the detection data of the stop and hold vehicle after it moves again according to the target ID of the moving target before it stops.
2. The system according to claim 1, characterized in that The second preset duration is determined as follows: The detection distance and the distance parameter are calculated using a preset duration algorithm to obtain a second preset duration, wherein the detection distance is the farthest distance that the target radar can detect.
3. The system according to claim 1, characterized in that The system further comprises: a target afterimage processing module; The target afterimage processing module is used for, when the moving target stops during the process of the target radar collecting detection data of the moving target, determining that the moving target is a stopped vehicle if the stop duration of the moving target exceeds a first preset duration, so that the target radar discards the target ID of the stopped vehicle; Furthermore, when the mobile target stops during the process of the target radar collecting detection data of the mobile target, and it is determined that the target distance of the mobile target from the target radar is not within the reference distance, if the stop time of the mobile target exceeds a second preset time, the mobile target is determined to be a stopped vehicle, so that the target radar discards the target ID of the stopped vehicle.
4. The system according to claim 1, characterized in that The system also includes: a false target elimination module; The false target elimination module is used to mark the initial moment when the target radar first collects any moving target. When the initial moment is delayed by M seconds, if the target radar cannot collect the detection data of the corresponding moving target, the corresponding moving target is determined to be an assumed target, and the assumed target is a moving target that disappears when the initial moment is delayed by M seconds.
5. The system according to claim 3, characterized in that The system also includes: a target flicker processing module; The target flicker processing module is used to process the target number of mobile targets collected by the target radar at the current moment. If the target number is not greater than the number threshold, the target radar is controlled to collect the corresponding detection data of all mobile targets; if the target number is greater than the number threshold, the target radar is controlled to package the detection data of the stopped vehicle to obtain a stop data identifier.
6. A target recognition method based on millimeter wave radar, characterized in that: Applied to a target recognition system including a target speed verification module and a target ID holding module, the method comprises: The target speed verification module obtains a first position, a first speed, a first acceleration and a second position of a first moving target at a first moment and a second moment respectively; Calculate the time difference between the first position, the second position, the first moment and the second moment to determine a first initial velocity; determining a second initial velocity based on the first acceleration, the first velocity, and the time difference; Determine, according to a configured speed determination algorithm, a second speed at the second moment from the first initial speed and the second initial speed; wherein the first position at the first moment is acquired by the first radar, the second position at the second moment is acquired by the second radar, and the first moment is less than the second moment; The method of determining that the first moving target acquired by the first radar and the second moving target acquired by the second radar are the same moving target is: The target ID holding module determines the predicted position at the second moment according to the first speed, the first acceleration, the first position and the first heading angle of the first moving target, matches the second position of the second moving target acquired by the second radar at the second moment with the predicted position, and if the match is successful, determines that the first moving target and the second moving target are the same moving target, and determines the target ID of the first moving target as the target ID of the second moving target; When the moving target stops during the process of the target radar collecting the detection data of the moving target, and it is determined that the target distance of the moving target from the target radar is within the configured reference distance, if the moving target moves again within the first preset time length after the stop time, the moving target is determined to be a stop-and-hold vehicle; so that the target radar collects the detection data of the stop-and-hold vehicle after it moves again according to the target ID before the stop-and-hold vehicle stops; When the moving target stops during the process of the target radar collecting detection data of the moving target, and it is determined that the target distance of the moving target from the target radar is not within the reference distance, if the moving target moves again within a second preset time period after being stopped, the moving target is determined to be a stop and hold vehicle; so that the target radar collects the detection data of the stop and hold vehicle after it moves again according to the target ID before the moving target stops.
7. An electronic device, characterized in that: The electronic device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; The processor is used to implement the method steps described in claim 6 when executing the program stored in the memory.
8. 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 method steps described in claim 6 are implemented.
