Radar data processing method and device based on traffic detection and storage medium

Abstract

The invention discloses a radar data processing method and device based on traffic detection and a storage medium, and belongs to the technical field of road traffic. The method comprises the following steps: performing association updating on original data periodically output by a radar and an existing vehicle track, and generating a first vehicle track set of a current period; traversing the correlation degree between the to-be-determined vehicle track in the first vehicle track set and other vehicle tracks in the first vehicle track set, and adding the vehicle tracks whose correlation degree with the to-be-determined vehicle track is greater than a first preset value into a second vehicle track set; calculating a difference value between the speed of the to-be-judged vehicle track in the historical updating process and the speed of each vehicle track in the second vehicle track set in the historical updating process; and fusing the vehicle track of which the difference value is smaller than the second preset value with the to-be-judged vehicle track. According to the method, the relevance and the speed consistency of the tracks are considered, the situation that different parts of the vehicle are mistakenly recognized as independent tracks is reduced, and the accuracy of vehicle recognition is improved.

Description

Technical Field

[0001] The present application belongs to the field of road traffic technology, and in particular, relates to a radar data processing method, device and storage medium based on traffic detection. Background Art

[0002] Radar technology plays a key role in the field of road traffic and is widely used in traffic monitoring and vehicle navigation to improve road safety and efficiency.

[0003] Currently, radar faces challenges in monitoring the trajectories of large vehicles. Due to the complex structure of large vehicles, radar often misidentifies different parts of them as independent trajectories, resulting in trajectory splitting, which affects the accuracy of traffic management and safety monitoring. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a radar data processing method, device and storage medium based on traffic detection to reduce the situation where different parts of a vehicle are misidentified as independent tracks and improve the accuracy of vehicle identification.

[0005] In a first aspect, the present application provides a radar data processing method based on traffic detection, comprising:

[0006] The original data periodically output by the radar is associated and updated with the existing vehicle trajectory to generate the first vehicle trajectory set of the current period;

[0007] Traversing the correlation degree between the to-be-determined vehicle trajectory in the first vehicle trajectory set and other vehicle trajectories in the first vehicle trajectory set, and adding the vehicle trajectory whose correlation degree with the to-be-determined vehicle trajectory is greater than a first preset value to the second vehicle trajectory set;

[0008] Calculate the difference between the speed of the to-be-determined vehicle trajectory during the historical update process and the speed of each vehicle trajectory in the second vehicle trajectory set during the historical update process;

[0009] The vehicle trajectory whose difference value is less than the second preset value is merged with the vehicle trajectory to be determined.

[0010] According to the radar data processing method based on traffic detection of the present application, by evaluating the degree of correlation between the trajectory of the vehicle to be determined and other trajectories in the first vehicle trajectory set, trajectories with high correlation are screened out to form a second vehicle trajectory set, which is helpful to identify those trajectories that may be the same vehicle from the perspective of spatial correlation, and then by calculating the historical speed difference between the trajectory to be determined and the trajectory in the second trajectory set, and combining the preset threshold to perform trajectory fusion, the spatial correlation and speed consistency of the trajectory are comprehensively considered, and the parts that actually belong to the same vehicle but may be mistakenly identified as multiple independent trajectories are effectively integrated, thereby reducing the situation where different parts of the vehicle are mistakenly identified as independent trajectories, and improving the accuracy of vehicle identification.

[0011] According to one embodiment of the present application, the method further includes:

[0012] Calculating the occlusion range of the to-be-determined vehicle trajectory;

[0013] Each vehicle trajectory in the first vehicle trajectory set is traversed, and the vehicle trajectory in the occlusion range is added to the second vehicle trajectory set.

[0014] In this embodiment, by calculating the occlusion range of the vehicle trajectory to be determined, identifying the occlusion that the trajectory to be determined may cause to other trajectories, and screening the trajectories in the first vehicle trajectory set based on this information, it is possible to more accurately identify those parts that are actually associated with the trajectory to be determined but are not easily directly observed due to line of sight or radar beam occlusion, thereby further enhancing the ability to accurately identify vehicle trajectories.

[0015] According to one embodiment of the present application, the method further includes:

[0016] Acquire a first count value of successful association between the to-be-determined vehicle trajectory and the original data, and a second count value of successful association between each vehicle trajectory in the second vehicle trajectory set and the original data;

[0017] The second vehicle trajectory set is traversed, and vehicle trajectories corresponding to the second count value being greater than the first count value are deleted from the second vehicle trajectory set.

[0018] In this embodiment, by obtaining the first count value of the successful association between the to-be-determined vehicle trajectory and the original data, and the second count value of the successful association between each vehicle trajectory in the second vehicle trajectory set and the original data, since the higher the count value of the successful association, the higher the degree of matching between the trajectory and the original data, those trajectories with a higher degree of association with the original data are effectively excluded, because they are more likely to be independent real vehicle trajectories, rather than misidentifications caused by the complexity of the vehicle structure, thereby further enhancing the ability to accurately identify vehicle trajectories.

[0019] According to an embodiment of the present application, traversing the correlation degree between the to-be-determined vehicle trajectory in the first vehicle trajectory set and other vehicle trajectories in the first vehicle trajectory set includes:

[0020] Establish a wave gate matrix and set weights for different dimensions of the wave gate matrix;

[0021] Calculate, according to the weight, a weighted sum of differences between the to-be-determined vehicle trajectory and other vehicle trajectories in the first vehicle trajectory set in different dimensions of the wave gate matrix;

[0022] The degree of association between the to-be-determined vehicle trajectory and other vehicle trajectories in the first vehicle trajectory set is determined according to the weighted sum.

[0023] In this embodiment, by introducing a wave gate matrix and a weight distribution mechanism, the vehicle trajectory is analyzed in multiple dimensions. By calculating the weighted sum of the differences between the trajectory of the vehicle to be determined and the trajectories of other vehicles in each dimension of the wave gate matrix, the similarity between the trajectories is quantified, thereby improving the accuracy of determining the degree of association between different vehicle trajectories.

[0024] According to one embodiment of the present application, the method further includes:

[0025] The calculating the difference between the speed of the to-be-determined vehicle trajectory during the historical update process and the speed of each vehicle trajectory in the second vehicle trajectory set during the historical update process includes:

[0026] Obtaining a first trajectory speed table corresponding to the to-be-determined vehicle trajectory and a second trajectory speed table corresponding to each vehicle trajectory in the second vehicle trajectory set; wherein the trajectory speed table records the speed of the vehicle trajectory at each association update;

[0027] The difference value is calculated based on the speed difference between the first track speed table and the second track speed table.

[0028] In this embodiment, the speed change of the vehicle at each association update is recorded and tracked through the trajectory speed table, and the speed difference value between the trajectory of the vehicle to be determined and the trajectory in the second vehicle trajectory set is calculated. The dynamic matching degree between the trajectories can be quantified, and it can be more accurately determined whether the trajectory belongs to different parts of the same vehicle or an independent vehicle, thereby reducing misidentification caused by inconsistent speed changes.

[0029] According to an embodiment of the present application, fusing the vehicle trajectory whose difference value is less than the second preset value with the to-be-determined vehicle trajectory includes:

[0030] Adding the vehicle trajectory whose difference value is less than the second preset value and the to-be-determined vehicle trajectory to the split vehicle trajectory set;

[0031] The split vehicle trajectory set is traversed, and the vehicle trajectory located at the front in the vehicle's traveling direction is taken as the head of the split vehicle group for fusion to obtain the vehicle trajectory of the split vehicle group.

[0032] In this embodiment, by comparing the speed difference value and the preset threshold, the vehicle trajectory that matches the speed of the vehicle trajectory to be determined is screened out, so that the trajectories that are similar in speed and may belong to the same vehicle are fused. During the fusion process, according to the direction of vehicle travel, the frontmost trajectory is selected as the head of the split vehicle group, and the trajectory fusion is performed on this basis, taking into account the spatial position and travel direction of the vehicle, so that the fused trajectory can truly reflect the actual motion state of the vehicle, further improving the accuracy of vehicle identification.

[0033] According to one embodiment of the present application, the method further includes:

[0034] When the number of vehicle trajectories in the split vehicle trajectory set is greater than 2, determining the vehicle type of the split vehicle group as a large vehicle;

[0035] When the number of vehicle trajectories in the split vehicle trajectory set is less than or equal to 2, the vehicle type of the head of the split vehicle group is determined as the vehicle type of the split vehicle group.

[0036] In this embodiment, the identification of vehicle types is assisted by judging the number of vehicle trajectories. When the number of vehicle trajectories in the split vehicle trajectory set exceeds two, the vehicle type of the split vehicle group is determined to be a large vehicle. This is because large vehicles are more likely to produce multiple parts that are mistakenly identified as independent trajectories due to their complex structure and large size. If the number of vehicle trajectories in the split vehicle trajectory set is less than or equal to two, the type of the head vehicle is taken as the type of the entire split vehicle group, thereby improving the accuracy of vehicle type identification.

[0037] According to one embodiment of the present application, the method further includes:

[0038] Adding a splitting mark to the vehicle tracks in the splitting vehicle track set except the head of the splitting vehicle group;

[0039] In the process of associating and updating the vehicle trajectories, if the vehicle trajectory with the splitting mark does not belong to the vehicle trajectory of the split vehicle group, the splitting mark is deleted.

[0040] In this embodiment, by adding a split identifier to the vehicle tracks except the head in the split vehicle track set, it helps to clearly distinguish those vehicle parts that may be misidentified as independent tracks due to radar reflection or structural complexity. During the association update process of the vehicle tracks, if it is found that a track with a split identifier is no longer associated with other tracks in the split vehicle group, that is, it no longer meets the track characteristics of a large vehicle, the split identifier of the track will be deleted, which effectively reduces the cases of misidentification and incorrect association and improves the accuracy of vehicle identification.

[0041] In a second aspect, the present application provides a radar data processing device based on traffic detection, comprising:

[0042] An association module, used to associate and update the original data periodically output by the radar with the existing vehicle trajectory to generate a first vehicle trajectory set of the current period;

[0043] A traversal module, configured to traverse the correlation between the to-be-determined vehicle trajectory in the first vehicle trajectory set and other vehicle trajectories in the first vehicle trajectory set, and add a vehicle trajectory whose correlation with the to-be-determined vehicle trajectory is greater than a first preset value to the second vehicle trajectory set;

[0044] A calculation module, used for calculating the difference between the speed of the to-be-determined vehicle trajectory during the historical update process and the speed of each vehicle trajectory in the second vehicle trajectory set during the historical update process;

[0045] The fusion module is used to fuse the vehicle trajectory whose speed difference value is less than the second preset value with the vehicle trajectory to be determined.

[0046] According to the radar data processing device based on traffic detection of the present application, by evaluating the correlation degree between the trajectory of the vehicle to be determined and other trajectories in the first vehicle trajectory set, trajectories with high correlation are screened out to form a second vehicle trajectory set, which is helpful to identify those trajectories that may be the same vehicle from the perspective of spatial correlation, and then by calculating the historical speed difference between the trajectory to be determined and the trajectory in the second trajectory set, and combining the preset threshold to perform trajectory fusion, the spatial correlation and speed consistency of the trajectory are comprehensively considered, and the parts that actually belong to the same vehicle but may be mistakenly identified as multiple independent trajectories are effectively integrated, thereby reducing the situation where different parts of the vehicle are mistakenly identified as independent trajectories, and improving the accuracy of vehicle identification.

[0047] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the radar data processing method based on traffic detection as described in the first aspect above is implemented.

[0048] In a fourth aspect, the present application provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the radar data processing method based on traffic detection as described in the first aspect above is implemented.

[0049] In a fifth aspect, the present application provides a chip, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the radar data processing method based on traffic detection as described in the first aspect above.

[0050] In a sixth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the radar data processing method based on traffic detection as described in the first aspect above.

[0051] The above one or more technical solutions in the embodiments of the present application have at least one of the following technical effects:

[0052] According to the radar data processing method based on traffic detection of the present application, by evaluating the degree of correlation between the trajectory of the vehicle to be determined and other trajectories in the first vehicle trajectory set, trajectories with high correlation are screened out to form a second vehicle trajectory set, which is helpful to identify those trajectories that may be the same vehicle from the perspective of spatial correlation, and then by calculating the historical speed difference between the trajectory to be determined and the trajectory in the second trajectory set, and combining the preset threshold to perform trajectory fusion, the spatial correlation and speed consistency of the trajectory are comprehensively considered, and the parts that actually belong to the same vehicle but may be mistakenly identified as multiple independent trajectories are effectively integrated, thereby reducing the situation where different parts of the vehicle are mistakenly identified as independent trajectories, and improving the accuracy of vehicle identification.

[0053] Furthermore, in some embodiments, by calculating the occlusion range of the vehicle trajectory to be determined, identifying the occlusion that the trajectory to be determined may cause to other trajectories, and combining this information to screen the trajectories in the first vehicle trajectory set, it is possible to more accurately identify those parts that are actually associated with the trajectory to be determined but are not easily directly observed due to line of sight or radar beam occlusion, thereby further enhancing the ability to accurately identify vehicle trajectories.

[0054] Furthermore, in some embodiments, by obtaining a first count value indicating that the to-be-determined vehicle trajectory is successfully associated with the original data, and a second count value indicating that each vehicle trajectory in the second vehicle trajectory set is successfully associated with the original data, since a higher count value indicates that the trajectory has a higher degree of match with the original data, those trajectories with a higher degree of association with the original data are effectively excluded because they are more likely to be independent real vehicle trajectories rather than misidentifications caused by the complexity of the vehicle structure, thereby further enhancing the ability to accurately identify vehicle trajectories.

[0055] Furthermore, in some embodiments, by introducing a wave gate matrix and a weight distribution mechanism, the vehicle trajectory is analyzed in multiple dimensions, and the weighted sum of the differences between the trajectory of the vehicle to be determined and other vehicle trajectories in each dimension of the wave gate matrix is ​​calculated, thereby quantifying the similarity between the trajectories, thereby improving the accuracy of determining the degree of association between different vehicle trajectories.

[0056] Furthermore, in some embodiments, by recording and tracking the speed changes of the vehicle at each association update through a trajectory speed table, and calculating the speed difference value between the trajectory of the vehicle to be determined and the trajectory in the second vehicle trajectory set, the dynamic matching degree between the trajectories can be quantified, and it can be more accurately determined whether the trajectories belong to different parts of the same vehicle, or independent vehicles, thereby reducing misidentification caused by inconsistent speed changes.

[0057] Furthermore, in some embodiments, by comparing the speed difference value and the preset threshold, the vehicle trajectory that matches the speed of the vehicle trajectory to be determined is screened out, so that the trajectories that are similar in speed and may belong to the same vehicle are fused. During the fusion process, the frontmost trajectory is selected as the head of the split vehicle group according to the direction of vehicle travel, and the trajectory fusion is performed on this basis, taking into account the spatial position and direction of travel of the vehicle, so that the fused trajectory can truly reflect the actual motion state of the vehicle, further improving the accuracy of vehicle identification.

[0058] Furthermore, in some embodiments, the identification of vehicle types is assisted by determining the number of vehicle trajectories. When the number of vehicle trajectories in a split vehicle trajectory set exceeds two, the vehicle type of the split vehicle group is determined to be a large vehicle. This is because large vehicles are more likely to produce multiple parts that are mistakenly identified as independent trajectories due to their complex structure and large size. If the number of vehicle trajectories in the split vehicle trajectory set is less than or equal to two, the type of the head vehicle is taken as the type of the entire split vehicle group, thereby improving the accuracy of vehicle type identification.

[0059] Furthermore, in some embodiments, by adding a split identifier to the vehicle tracks other than the head in the split vehicle track set, it helps to clearly distinguish those vehicle parts that may be misidentified as independent tracks due to radar reflection or structural complexity. During the association update process of the vehicle tracks, if it is found that a track with a split identifier is no longer associated with other tracks in the split vehicle group, that is, it no longer meets the track characteristics of a large vehicle, the split identifier of the track will be deleted, effectively reducing the cases of misidentification and incorrect association, and improving the accuracy of vehicle identification.

[0060] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0062] Figure 1 It is a flow chart of a radar data processing method based on traffic detection provided in an embodiment of the present application;

[0063] Figure 2 is a schematic diagram of the structure of a radar data processing device based on traffic detection provided in an embodiment of the present application;

[0064] Figure 3 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0066] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0067] In the field of modern road traffic, radar technology is increasingly used, playing an important role in traffic monitoring, vehicle navigation, collision prevention and many other aspects. Radar systems can accurately measure the distance, speed and direction of target objects by emitting radio waves and receiving reflected signals. This technology has made a significant contribution to improving road safety and traffic efficiency, especially in dynamic traffic environments, where radar can monitor the driving status of vehicles in real time and provide key information for traffic management and vehicle control systems. For example, in traffic signal control, radar can detect the arrival and departure of vehicles, thereby optimizing the timing of signal lights and reducing waiting time and traffic congestion. In addition, radar is also used for traffic flow monitoring on highways, helping to prevent collisions and rear-end accidents by analyzing vehicle spacing and speed.

[0068] However, despite the many advantages of radar technology in road traffic monitoring, there are still some technical challenges in dealing with trajectory detection of large vehicles. Large vehicles, such as vans and trucks, are usually composed of multiple parts, such as carriages, trailers, etc., due to their large size. In practical applications, the ideal state is to identify the entire vehicle as a single trajectory unit to facilitate effective traffic management and safety monitoring. However, due to the uneven detection quality of current radar manufacturers and the influence of some characteristics of radar detection itself, especially in terms of resolution and signal processing, the different components of these vehicles are often mistakenly identified as multiple independent trajectories, resulting in the trajectory of the vehicle being split in the radar monitoring system, which not only affects the accurate analysis of the behavior of the vehicle, but also may mislead the prediction and management of traffic flow. For example, a vehicle consists of two parts, a trailer and a tractor. When the radar identifies the trailer and the tractor as different trajectories, it may lead to misjudgment of the actual driving path of the vehicle, thereby affecting the optimization of traffic signals and the assessment of vehicle collision risks.

[0069] The present application considers that if the correlation degree and speed consistency of each vehicle trajectory in the vehicle trajectory set can be evaluated, and the parts that may belong to the same vehicle but may be mistakenly identified as multiple independent vehicle trajectories can be fused, it is expected to solve the problem in the prior art that different parts of the vehicle are easily misidentified as independent trajectories.

[0070] The following is a detailed description of the radar data processing method and device based on traffic detection provided by the embodiments of the present application through specific embodiments and their application scenarios in conjunction with the accompanying drawings.

[0071] Among them, the radar data processing method based on traffic detection can be applied to the terminal, and can be specifically executed by hardware or software in the terminal.

[0072] The terminal includes, but is not limited to, a portable communication device such as a mobile phone or tablet computer with a touch-sensitive surface (e.g., a touch screen display and / or a touch pad). It should also be understood that in some embodiments, the terminal may not be a portable communication device, but a desktop computer with a touch-sensitive surface (e.g., a touch screen display and / or a touch pad).

[0073] In the following various embodiments, a terminal including a display and a touch-sensitive surface is described. However, it should be understood that the terminal may include one or more other physical user interface devices such as a physical keyboard, a mouse and a joystick.

[0074] The radar data processing method based on traffic detection provided in the embodiment of the present application may be executed by an electronic device or a functional module or functional entity in the electronic device that can implement the radar data processing method based on traffic detection. The electronic devices mentioned in the embodiment of the present application include but are not limited to mobile phones, tablet computers, computers, cameras, and wearable devices. The radar data processing method based on traffic detection provided in the embodiment of the present application is described below using an electronic device as an example of the execution subject.

[0075] like Figure 1 As shown, the radar data processing method based on traffic detection includes: step 110, step 120, step 130 and step 140.

[0076] Step 110: associate and update the original data periodically output by the radar with the existing vehicle trajectory to generate a first vehicle trajectory set of the current period.

[0077] The raw data output by the radar is the signal reflected by the electromagnetic waves emitted by the radar after encountering the target object. The time difference between the transmitted signal and the received signal, the frequency change and other information are used to calculate the position of the radar and the vehicle, the speed of the vehicle and other data. These data can be used to construct the vehicle trajectory. The raw signal output by the radar is usually periodic, and this periodicity is determined by the repetitive pattern of the electromagnetic waves emitted by the radar. The radar detects the target by emitting electromagnetic waves and receiving echoes. The emission of these electromagnetic waves is repeated at a certain time interval, thus forming a periodic signal pattern. This data is the basis for further processing and analysis by the radar system. This period can be in seconds or milliseconds, depending on the sampling rate and data processing speed of the radar system. The current period refers to the period at the current moment.

[0078] Vehicle trajectory refers to the process of radar tracking a vehicle through a trajectory tracking algorithm, thereby generating a vehicle trajectory. The methods of trajectory tracking can be track initiation, trajectory filtering, trajectory prediction, track management, etc. Track initiation is the process from detecting the target until the target can be continuously tracked; trajectory filtering is to extract the target's motion trajectory from the raw data received by the sensor, and estimate the target's state based on the target motion model and observation model; trajectory prediction is to predict the target's future position and motion state based on the target's historical trajectory data and known motion model; track management is the process of organizing, analyzing and optimizing the target's trajectory, that is, the target's path and state sequence in the target tracking system, to ensure that the tracking system can effectively process and analyze the target data, thereby achieving accurate tracking and prediction of the target. The algorithms involved may include Bayesian estimation, hough transform method, logical method; nearest neighbor method, probabilistic data association algorithm, Kalman filter, extended Kalman filter, interactive multi-model algorithm, etc.

[0079] The embodiments of the present application do not limit the trajectory tracking method and the algorithms involved, but need to ensure that the output trajectory is relatively stable and has strong anti-interference ability.

[0080] In the case of multiple vehicles, the radar system needs to associate the vehicle information detected in continuous time frames to maintain tracking of each vehicle. For example, the radar system monitors the vehicle trajectories of vehicle A, vehicle B, and vehicle C during the monitoring process. Since the monitoring process is continuous, after a period of time, the radar system obtains some raw data. It is unclear which of these raw data corresponds to vehicle A and which corresponds to vehicle B. Therefore, these raw data need to be processed to identify the position, speed and other information of the vehicles in these raw data. Then, the position, speed and other information of the vehicles in these raw data are associated and matched with the vehicle trajectories of vehicles A, vehicle B and vehicle C that have been identified before, so as to determine which data in the raw data corresponds to vehicle A and which data corresponds to vehicle B. In this way, the vehicle trajectory at the current moment can be updated. This process is called the association update process. Similarly, this process is repeated in the next cycle to complete the dynamic update of different vehicle trajectories.

[0081] Specifically, when the radar system receives new raw data (vehicle location, speed, and other information), it can use data association algorithms, such as the nearest neighbor method, probabilistic data association, or multiple hypothesis tracking algorithms, to associate and match the raw data with existing vehicle trajectories. If the algorithm determines that the new raw data is similar enough to a known vehicle trajectory, that is, there is a high probability that they are from the same vehicle, then the association is considered successful. For example, for vehicle A, the new raw data has a low similarity to the trajectory of vehicle A, and the association is considered to have failed.

[0082] It should be noted that for existing vehicle trajectories, due to factors such as occlusion and error, not every update can be successfully associated. The association may fail in this update, but will succeed in the next update. For example, when the number of failed associations for the trajectory of vehicle A reaches a certain number, it can be considered that vehicle A can no longer be tracked, and in the subsequent update process, the trajectory of vehicle A will no longer be associated and updated.

[0083] The operation of associating radar raw data with vehicle trajectory may include data collection, preprocessing, feature extraction, data association, etc. Data collection refers to collecting radar raw data and vehicle trajectory data, data preprocessing may be pre-removing noise and outliers from the collected data, and feature extraction refers to extracting useful features from radar data and vehicle trajectory data respectively. These useful features may be time, position, speed, direction, size of the target, speed, acceleration, steering angle, etc. of the vehicle. Data association is to use an algorithm to associate useful features from the extracted radar data and vehicle trajectory data. The algorithm used here may be a Kalman filter, a particle filter, etc. The embodiment of the present application does not limit the algorithm used.

[0084] In an embodiment of the present application, the raw data periodically output by the radar can be associated with the vehicle trajectory, and the set of vehicle trajectories generated in each cycle can be recorded as the vehicle trajectory set of the cycle. The first vehicle trajectory set is a set of vehicle trajectory data generated by association update in the current processing cycle. The first vehicle trajectory set collects the current trajectories of each vehicle detected by the radar. In the next update cycle, the radar will associate and update the vehicle trajectories in the first vehicle trajectory set according to the new raw data, and generate new trajectories to obtain a new vehicle trajectory set.

[0085] Step 120: traverse the correlation degree between the to-be-determined vehicle trajectory in the first vehicle trajectory set and other vehicle trajectories in the first vehicle trajectory set, and add the vehicle trajectory whose correlation degree with the to-be-determined vehicle trajectory is greater than a first preset value to the second vehicle trajectory set.

[0086] In the embodiment of the present application, the to-be-determined vehicle trajectory may be any one of the vehicle trajectories in the first vehicle trajectory set, and the other vehicle trajectories may be any one or more vehicle trajectories in the first vehicle trajectory set except the to-be-determined vehicle trajectory. By detecting and determining the to-be-determined vehicle trajectory, it is identified whether the to-be-determined vehicle trajectory is an independent trajectory of a vehicle or a trajectory of a part of a vehicle.

[0087] In an embodiment of the present application, the vehicle trajectory with a high degree of correlation with the vehicle trajectory to be determined can be added to the second vehicle trajectory set by traversing the degree of correlation between the vehicle trajectory to be determined and other vehicle trajectories. Specifically, the degree of correlation between the two trajectories can be evaluated by defining a quantitative metric. For example, the degree of correlation between the two trajectories can be evaluated by taking one or more of the parameters such as distance, direction, speed, and time synchronization as a metric; according to the selected metric, a mathematical model or algorithm is established to calculate the correlation between the two trajectories. For example, a distance metric, such as Euclidean distance, or directional consistency, such as angle difference, can be used to evaluate the degree of correlation between different vehicle trajectories. Of course, the degree of correlation between different vehicle trajectories can also be determined in other ways, and the embodiment of the present application is not limited to this.

[0088] For the vehicle trajectory to be determined, the vehicle trajectory to be determined is compared with other vehicle trajectories in the first trajectory set. For each comparison, the degree of association between the vehicle trajectory to be determined and other vehicle trajectories currently traversed can be calculated, and the calculated degree of association can be compared with the first preset value. If the degree of association is greater than or equal to the first preset value, it can be considered that the two vehicle trajectories have a high degree of association. Other vehicle trajectories whose degree of association with the vehicle trajectory to be determined is higher than the first preset value are added to the second vehicle trajectory set. Among them, since the vehicle trajectories in the second vehicle trajectory set have a high degree of association with the vehicle trajectory to be determined, the vehicle trajectories in the second vehicle trajectory set may be vehicle trajectories corresponding to different parts of the same vehicle as the vehicle trajectory to be determined. In this step, the second vehicle trajectory set with a high degree of association with the vehicle trajectory to be determined is screened out, further narrowing the scope of determination.

[0089] Step 130: Calculate the difference between the speed of the to-be-determined vehicle trajectory during the historical update process and the speed of each vehicle trajectory in the second vehicle trajectory set during the historical update process.

[0090] In the embodiment of the present application, in order to further analyze the vehicle trajectories in the second vehicle trajectory set, it is also necessary to compare the speed of the vehicle trajectory to be determined during the historical update process with the speed of each vehicle trajectory in the second vehicle trajectory set during the historical update process.

[0091] The speed in the historical update process is the speed record of the vehicle at a series of time points. Specifically, during the vehicle monitoring process, the radar will periodically update the associated tracks of each vehicle based on the acquired data, for example, the direction, position, speed and other information of the vehicle track can be updated.

[0092] In an embodiment of the present application, during the process of associating and updating the vehicle trajectory, the speed value corresponding to the vehicle trajectory and the time point of the update can be recorded each time the vehicle trajectory is updated. When it is necessary to calculate the speed difference between the vehicle trajectory to be determined and each vehicle trajectory in the second vehicle trajectory set, one or more specific historical time points can be selected first, or all available historical time points can be used to compare the speed difference between the two vehicle trajectories. For example, the speed of the vehicle trajectory to be determined during the historical update process is: time point A corresponds to speed a1, time point B corresponds to speed b1, and time point C corresponds to speed c1. The speed of one of the vehicle trajectories in the second vehicle trajectory set during the historical update process is: time point A corresponds to speed a2, time point B corresponds to speed b2, and time point C corresponds to speed c2. Then the speed at the same time point can be compared, for example, speed a1 is compared with speed a2, speed b1 is compared with speed b2, and speed c1 is compared with speed c2. These comparisons can be to calculate the difference between different speeds, and then the absolute value of the difference between the speeds at different time points is averaged, and the average value is used as the difference value. The smaller the average value, the smaller the difference between the vehicle trajectory to be determined and this vehicle trajectory in the speed dimension. Of course, if the time points of the to-be-determined vehicle trajectory and the vehicle trajectories in the second vehicle trajectory set do not correspond, interpolation or other methods may be used to estimate the speed of the missing time points.

[0093] Step 140: merge the vehicle trajectory whose difference value is less than the second preset value with the trajectory of the vehicle to be determined.

[0094] In the embodiment of the present application, the second preset value may be obtained based on experience and data analysis, and is used to determine whether the speeds of the two vehicle trajectories are close enough. If the difference value is less than the second preset value, the speeds of the two vehicle trajectories during the historical update process are very close, and it can be considered that the two vehicle trajectories may belong to different parts of the same vehicle, and therefore need to be fused.

[0095] Specifically, the data points of the vehicle trajectory to be determined and the vehicle trajectory whose difference value is less than the second preset value can be merged into one vehicle trajectory, or an identifier indicating the same vehicle trajectory can be set for the vehicle trajectory to be determined and the vehicle trajectory whose difference value is less than the second preset value. Of course, the fusion can also be performed in any other way, which is not limited in the embodiments of the present application.

[0096] According to the radar data processing method based on traffic detection of the present application, by evaluating the degree of correlation between the trajectory of the vehicle to be determined and other trajectories in the first vehicle trajectory set, trajectories with high correlation are screened out to form a second vehicle trajectory set, which is helpful to identify those trajectories that may be the same vehicle from the perspective of spatial correlation, and then by calculating the historical speed difference between the trajectory to be determined and the trajectory in the second trajectory set, and combining the preset threshold to perform trajectory fusion, the spatial correlation and speed consistency of the trajectory are comprehensively considered, and the parts that actually belong to the same vehicle but may be mistakenly identified as multiple independent trajectories are effectively integrated, thereby reducing the situation where different parts of the vehicle are mistakenly identified as independent trajectories, and improving the accuracy of vehicle identification.

[0097] In some embodiments, the method further comprises:

[0098] Calculate the occlusion range of the vehicle trajectory to be determined;

[0099] Traverse each vehicle trajectory in the first vehicle trajectory set, and add the vehicle trajectory in the occlusion range to the second vehicle trajectory set.

[0100] The occlusion ability of a vehicle usually refers to its blocking effect on line of sight, radar or other detection means in a specific environment. Since the vehicle corresponding to the vehicle trajectory to be determined has a certain volume space, there will be an occlusion range. The occlusion range is relatively close to the vehicle corresponding to the vehicle trajectory to be determined in spatial position. The vehicle trajectory within the occlusion range is also likely to be a different part of the same vehicle as the vehicle trajectory to be determined. Therefore, the vehicle trajectory within the occlusion range can also be added to the second vehicle trajectory set.

[0101] In this embodiment, the radar device installation height h, the vehicle height h car , the vehicle's longitudinal axis coordinate information y0 determines the occlusion range of the vehicle at a specific position. The longitudinal axis of the vehicle can be the length direction of the vehicle body, and the transverse axis of the vehicle can be the width direction of the vehicle body, or the longitudinal axis of the vehicle can also be the length direction of the road, and the transverse axis of the vehicle can be the width direction of the road. The occlusion range can be divided into the starting transverse coordinate C xs , end the horizontal axis C xe , starting ordinate C ys , end vertical coordinate C ye Therefore, the calculation method of each value in the occlusion area (Cxs, Cxe, Cys, Cye) is as follows:

[0102] Starting horizontal coordinate C xs : The boundary of the left side of the current vehicle. If the distance to the lane boundary is less than the preset distance, such as less than 1m, 0.8m, 0.5m, etc., the left boundary of the lane is used.

[0103] End horizontal axis Cxe : The boundary of the current right side of the vehicle. If the distance to the lane boundary is less than the preset distance, such as less than 1m, 0.8m, 0.5m, etc., the right boundary of the lane is used.

[0104] The starting ordinate Cys: can be the ordinate y coordinate of the head position of the current vehicle.

[0105] End vertical coordinate: C ye =(h car *y0) / (hh car ).

[0106] In this embodiment, by calculating the occlusion range of the vehicle trajectory to be determined, identifying the occlusion that the trajectory to be determined may cause to other trajectories, and screening the trajectories in the first vehicle trajectory set based on this information, it is possible to more accurately identify those parts that are actually associated with the trajectory to be determined but are not easily directly observed due to line of sight or radar beam occlusion, thereby further enhancing the ability to accurately identify vehicle trajectories.

[0107] In some embodiments, the method further comprises:

[0108] Obtaining the vehicle type and radar scattering area identified by the radar for the vehicle trajectory to be determined;

[0109] When the vehicle type and the radar scattering area meet the preset relationship, the vehicle type is determined as the type of vehicle corresponding to the vehicle trajectory to be determined;

[0110] When the vehicle type and the target radar scattering area do not conform to a preset relationship, the vehicle type with the highest frequency in the historical vehicle types of the vehicle track to be determined is determined as the type of vehicle corresponding to the vehicle track to be determined.

[0111] Radar Cross Section (RCS) is a physical quantity that measures an object's ability to reflect radar waves. The size of RCS varies with factors such as the viewing angle, radar frequency, and the shape and material of the target. Different types of vehicles have different geometric shapes and material compositions, so their theoretical RCS values ​​are also different. Generally speaking, large vehicles have larger geometric cross-sections, so large vehicles have higher RCS.

[0112] In this embodiment, a preset relationship between vehicle type and radar scattering area can be established in advance. This preset relationship can be set as follows: for example, the RCS of non-motor vehicles is ≤7dbsm, the RCS of small cars is between 7dbsm and 15dbsm, and the dbsm of large vehicles is ≥15dbsm. Of course, the RCS preset range corresponding to different vehicle types can be adjusted. For vehicles with larger geometric cross-sectional areas, the corresponding RCS value should be higher. This application does not limit the specific range.

[0113] When the vehicle type and the radar scattering area conform to a preset relationship, that is, the vehicle type and the radar scattering area correspond to each other, the vehicle type can be recorded in the corresponding vehicle track.

[0114] When the vehicle type and the target radar scattering area do not conform to the preset relationship, the historical identification type information of the vehicle trajectory to be determined can be retrieved for comparison, and the vehicle type with the highest frequency of historical type identification is selected as the vehicle type corresponding to the vehicle trajectory to be determined.

[0115] In this embodiment, the vehicle type of the vehicle corresponding to the to-be-determined vehicle trajectory is determined in combination with the radar scattering area, thereby further improving the accuracy of vehicle type determination.

[0116] In some embodiments, the method further comprises:

[0117] Obtaining a first count value of successful association between the to-be-determined vehicle trajectory and the original data, and a second count value of successful association between each vehicle trajectory in the second vehicle trajectory set and the original data;

[0118] The second vehicle trajectory set is traversed, and the vehicle trajectories corresponding to the second count value being greater than the first count value are deleted from the second vehicle trajectory set.

[0119] The vehicle trajectory is associated with the original data of the radar and updated. If the association is successful, it means that the vehicle trajectory and the original data of the radar can match, the vehicle trajectory conforms to the data obtained by the radar, and the radar can successfully detect the vehicle status and process the vehicle trajectory. If the vehicle trajectory and the original data of the radar fail to be associated, it means that the radar cannot successfully detect the vehicle and cannot track the vehicle trajectory. In each cycle, the original data output by the radar will be associated with the existing vehicle trajectory. For each vehicle trajectory, if the association is successful, the successful association count of the vehicle trajectory can be increased. For example, each time the association is successful, the successful association count value can be increased by one. If the association fails, the failed association count of the vehicle trajectory can be increased by one.

[0120] The larger the successful association count value is, the higher the stability of the vehicle track over a period of time is, the more consistent it is with the original data of the radar, and the higher the consistency is with the original data. Therefore, when the second count value corresponding to a vehicle track in the second vehicle track set is greater than the first count value of the vehicle track to be determined, it can be considered that the vehicle track is likely to be an independent vehicle track, so the vehicle track can be deleted from the second vehicle track set.

[0121] For example, in the second vehicle trajectory set, the higher the number of successful associations of vehicle A's trajectory, the more stable the trajectory monitoring process of vehicle A is during the previous association update process, because more new raw data can be associated with vehicle A during multiple updates. If the number of successful associations of the trajectory of the to-be-determined vehicle B is less than that of vehicle A, it means that during the entire monitoring period, the stability of the monitored vehicle A is higher than that of the to-be-determined vehicle B, so vehicle A is most likely an independent vehicle, so the trajectory of vehicle A can be deleted from the second vehicle trajectory set.

[0122] In this embodiment, by obtaining the first count value of the successful association between the to-be-determined vehicle trajectory and the original data, and the second count value of the successful association between each vehicle trajectory in the second vehicle trajectory set and the original data, since the higher the count value of the successful association, the higher the degree of matching between the trajectory and the original data, those trajectories with a higher degree of association with the original data are effectively excluded, because they are more likely to be independent real vehicle trajectories, rather than misidentifications caused by the complexity of the vehicle structure, thereby further enhancing the ability to accurately identify vehicle trajectories.

[0123] In some embodiments, traversing the correlation between the to-be-determined vehicle trajectory in the first vehicle trajectory set and other vehicle trajectories in the first vehicle trajectory set includes:

[0124] Establish a wave gate matrix and set weights for different dimensions of the wave gate matrix;

[0125] Calculate the weighted sum of the differences between the to-be-determined vehicle trajectory and other vehicle trajectories in the first vehicle trajectory set in different dimensions of the gate matrix according to the weights;

[0126] The degree of association between the to-be-determined vehicle trajectory and other vehicle trajectories in the first vehicle trajectory set is determined according to the weighted sum.

[0127] The Gate Matrix is ​​a concept used in Multi-Target Tracking (MTT) and data association problems, especially in radar or sensor systems, to quantify and evaluate the degree of association between different tracks or observations. The Gate Matrix is ​​a multi-dimensional data structure, each dimension represents a specific physical quantity or feature, such as time, spatial position, velocity, acceleration, etc. Each dimension can be assigned a weight to represent the importance of the feature in data association, and the weight can be determined based on prior knowledge, experience, or through optimization algorithms.

[0128] In this embodiment, taking different dimensions including x-direction spacing, y-direction spacing, x-direction speed difference sx, and y-direction speed difference sy as examples, corresponding weights can be set for these dimensions of the wave gate matrix, for example, the weight of the x-direction spacing is set to 50, the weight of the y-direction spacing is set to 20, the weight of the x-direction speed difference sx is set to 0, and the weight of the y-direction speed difference sy is set to 30.

[0129] Then, according to the set dimensions and weights, the weighted sums of the differences in different dimensions of the wave gate matrix between the vehicle trajectory to be determined and other vehicle trajectories in the first vehicle trajectory set are calculated respectively. For example, for the difference in each dimension, the corresponding weight is multiplied, and then the weighted sum is summed up. The weighted sum quantifies the comprehensive differences between the vehicle trajectory to be determined and other vehicle trajectories in multiple dimensions. The larger the weighted sum of other vehicle trajectories and the vehicle trajectory to be determined, the greater the difference with the vehicle trajectory to be determined, and the smaller the degree of correlation. Therefore, the vehicle trajectory whose weighted sum is greater than the preset value can be added to the second vehicle trajectory set.

[0130] In this embodiment, by introducing a wave gate matrix and a weight distribution mechanism, the vehicle trajectory is analyzed in multiple dimensions. By calculating the weighted sum of the differences between the trajectory of the vehicle to be determined and the trajectories of other vehicles in each dimension of the wave gate matrix, the similarity between the trajectories is quantified, thereby improving the accuracy of determining the degree of association between different vehicle trajectories.

[0131] In some embodiments, the method further comprises:

[0132] Calculating the difference between the speed of the to-be-determined vehicle trajectory during the historical update process and the speed of each vehicle trajectory in the second vehicle trajectory set during the historical update process, including:

[0133] Obtaining a first trajectory speed table corresponding to the to-be-determined vehicle trajectory and a second trajectory speed table corresponding to each vehicle trajectory in the second vehicle trajectory set; wherein the trajectory speed table records the speed of the vehicle trajectory at each association update;

[0134] The difference value is calculated based on the speed difference between the first track speed table and the second track speed table.

[0135] In this embodiment, during the process of associating and updating the vehicle trajectory, the speed value corresponding to the vehicle trajectory and the time point of the update can be recorded each time the vehicle trajectory is updated, and stored in the trajectory speed table corresponding to the vehicle trajectory. For example, the first trajectory speed table corresponding to the vehicle trajectory to be determined stores the speed a1 corresponding to time point A, the speed b1 corresponding to time point B, and the speed c1 corresponding to time point C. The first trajectory speed table corresponding to one of the vehicle trajectories in the second vehicle trajectory set stores the speed a2 corresponding to time point A, the speed b2 corresponding to time point B, and the speed c2 corresponding to time point C. Then, the speeds at the same time point can be compared, for example, speed a1 is compared with speed a2, speed b1 is compared with speed b2, and speed c1 is compared with speed c2. These comparisons can be to calculate the difference between different speeds, and then average the absolute values ​​of the differences in speed at different time points, and use the average value as the difference value. The smaller the average value, the smaller the difference between the vehicle trajectory to be determined and this vehicle trajectory in the speed dimension.

[0136] In some embodiments, the difference value may also be calculated in other ways, for example, by calculating the mean square error of the speeds in the first trajectory speed table and the second trajectory speed table, thereby effectively reflecting the difference in speeds of the two vehicle trajectories during the historical operation process. The calculation formula is as follows:

[0137]

[0138] Where MSR represents mean square error; S 1i represents the i-th speed value in the first trajectory speed table corresponding to the vehicle trajectory 1 to be determined, S 2i Represents the speed value of the i-th item in the second trajectory speed table corresponding to vehicle trajectory 2 in the second vehicle trajectory set.

[0139] In some embodiments, the vehicle trajectory with MSR less than the preset value can be fused with the trajectory to be determined. In one example, based on engineering experience, the vehicle trajectory with MSR<1.2 can be fused with the trajectory to be determined. Of course, the preset value can also be other values, such as 1.0, 1.3, etc., which is not limited in the embodiments of the present application.

[0140] In this embodiment, the speed change of the vehicle at each association update is recorded and tracked through the trajectory speed table, and the speed difference value between the trajectory of the vehicle to be determined and the trajectory in the second vehicle trajectory set is calculated. The dynamic matching degree between the trajectories can be quantified, and it can be more accurately determined whether the trajectory belongs to different parts of the same vehicle or an independent vehicle, thereby reducing misidentification caused by inconsistent speed changes.

[0141] In some embodiments, fusing the vehicle trajectory whose difference value is less than the second preset value with the vehicle trajectory to be determined includes:

[0142] Adding the vehicle trajectory whose difference value is less than the second preset value and the vehicle trajectory to be determined to the split vehicle trajectory set;

[0143] The set of split vehicle trajectories is traversed, and the vehicle trajectory located at the front in the direction of vehicle travel is taken as the head of the split vehicle group for fusion to obtain the vehicle trajectory of the split vehicle group.

[0144] In this embodiment, for the vehicle trajectory in the second vehicle trajectory set, if the difference value is less than the second preset value, it can be considered that the vehicle trajectory and the vehicle trajectory to be determined may belong to different parts of the same vehicle, and therefore need to be fused.

[0145] Specifically, the vehicle trajectories with difference values ​​less than the second preset value and the vehicle trajectories to be determined can be added to a new set, called a split vehicle trajectory set. For the vehicle trajectories in the split vehicle trajectory set, the direction of travel of the vehicle trajectory can be determined by the order of points on the trajectory or the direction of the velocity vector, for example, the vehicle trajectory at the front is found according to the y-coordinate information of the trajectory. Each vehicle trajectory in the split vehicle trajectory set is traversed to determine the relative position of the vehicle trajectory in the direction of vehicle travel, and the vehicle trajectory at the front of the vehicle travel direction is selected as the head of the split vehicle group. The head trajectory is fused with other vehicle trajectories in the split vehicle trajectory set. The fusion can be a simple trajectory merging, or a more complex algorithm, such as a Kalman filter, is used to estimate the optimal trajectory representation.

[0146] The fused vehicle trajectory will contain updated position, velocity and other relevant state information to reflect the actual motion state of the vehicle. For the vehicle trajectory fused with the head trajectory, it can be deleted from the split vehicle trajectory set or marked as fused to avoid repeated processing.

[0147] In this embodiment, by comparing the speed difference value and the preset threshold, the vehicle trajectory that matches the speed of the vehicle trajectory to be determined is screened out, so that the trajectories that are similar in speed and may belong to the same vehicle are fused. During the fusion process, according to the direction of vehicle travel, the frontmost trajectory is selected as the head of the split vehicle group, and the trajectory fusion is performed on this basis, taking into account the spatial position and travel direction of the vehicle, so that the fused trajectory can truly reflect the actual motion state of the vehicle, further improving the accuracy of vehicle identification.

[0148] In some embodiments, when the number of vehicle trajectories in the split vehicle trajectory set is greater than 2, the vehicle type of the split vehicle group is determined to be a large vehicle;

[0149] When the number of vehicle trajectories in the split vehicle trajectory set is less than or equal to 2, the vehicle type of the head of the split vehicle group is determined as the vehicle type of the split vehicle group.

[0150] In this embodiment, the identification of vehicle types is assisted by judging the number of vehicle trajectories. When the number of vehicle trajectories in the split vehicle trajectory set exceeds two, the vehicle type of the split vehicle group is determined to be a large vehicle. This is because large vehicles are more likely to produce multiple parts that are mistakenly identified as independent trajectories due to their complex structure and large size. If the number of vehicle trajectories in the split vehicle trajectory set is less than or equal to two, the type of the head vehicle is taken as the type of the entire split vehicle group, thereby improving the accuracy of vehicle type identification.

[0151] In some embodiments, the method further comprises:

[0152] Add splitting marks to the vehicle trajectories in the splitting vehicle trajectory set except the head of the splitting vehicle group;

[0153] In the process of associating and updating the vehicle trajectories, if the vehicle trajectory with the splitting flag does not belong to the vehicle trajectory of the split vehicle group, the splitting flag is deleted.

[0154] In this embodiment, for other vehicle tracks in the split vehicle track set except the head track, a split mark can be added. The split mark is used to mark that these vehicle tracks are parts split from the same vehicle due to radar detection or other reasons. In the process of association update of the vehicle track, the state of the vehicle track, including position, speed, etc., can be updated according to the new original data. In the update process, it can be checked whether the vehicle track with the split mark is still associated with the head track of the split vehicle group. For example, the degree of association between the vehicle track with the split mark and the head track of the split vehicle group and the difference value of the speed in the historical update process can be calculated. If the track with the split mark is found to no longer be associated with the head track of the split vehicle group during the update process, that is, they no longer meet the association condition, then it can be considered that these vehicle tracks no longer belong to the same vehicle. For the vehicle track that no longer belongs to the split vehicle group, the split mark is deleted to indicate that these vehicle tracks are independent and not part of the split vehicle group.

[0155] In this embodiment, by adding a split identifier to the vehicle tracks except the head in the split vehicle track set, it helps to clearly distinguish those vehicle parts that may be misidentified as independent tracks due to radar reflection or structural complexity. During the association update process of the vehicle tracks, if it is found that a track with a split identifier is no longer associated with other tracks in the split vehicle group, that is, it no longer meets the track characteristics of a large vehicle, the split identifier of the track will be deleted, which effectively reduces the cases of misidentification and incorrect association and improves the accuracy of vehicle identification.

[0156] The radar data processing method based on traffic detection provided in the embodiment of the present application can be executed by a radar data processing device based on traffic detection. In the embodiment of the present application, the radar data processing device based on traffic detection performs the radar data processing method based on traffic detection as an example to illustrate the radar data processing device based on traffic detection provided in the embodiment of the present application.

[0157] The embodiment of the present application also provides a radar data processing device based on traffic detection.

[0158] like Figure 2 As shown, the radar data processing device based on traffic detection includes:

[0159] An association module 210 is used to associate and update the original data periodically output by the radar with the existing vehicle trajectory to generate a first vehicle trajectory set of the current period;

[0160] A traversal module 220, configured to traverse the correlation between the to-be-determined vehicle trajectory in the first vehicle trajectory set and other vehicle trajectories in the first vehicle trajectory set, and add a vehicle trajectory whose correlation with the to-be-determined vehicle trajectory is greater than a first preset value to the second vehicle trajectory set;

[0161] A calculation module 230, used to calculate the difference between the speed of the to-be-determined vehicle trajectory during the historical update process and the speed of each vehicle trajectory in the second vehicle trajectory set during the historical update process;

[0162] The fusion module 240 is used to fuse the vehicle trajectory whose difference value is less than the second preset value with the vehicle trajectory to be determined.

[0163] According to the radar data processing device based on traffic detection of the present application, by evaluating the correlation degree between the trajectory of the vehicle to be determined and other trajectories in the first vehicle trajectory set, trajectories with high correlation are screened out to form a second vehicle trajectory set, which is helpful to identify those trajectories that may be the same vehicle from the perspective of spatial correlation, and then by calculating the historical speed difference between the trajectory to be determined and the trajectory in the second trajectory set, and combining the preset threshold to perform trajectory fusion, the spatial correlation and speed consistency of the trajectory are comprehensively considered, and the parts that actually belong to the same vehicle but may be mistakenly identified as multiple independent trajectories are effectively integrated, thereby reducing the situation where different parts of the vehicle are mistakenly identified as independent trajectories, and improving the accuracy of vehicle identification.

[0164] In some embodiments, the traversal module 220 is further configured to:

[0165] Calculate the occlusion range of the vehicle trajectory to be determined;

[0166] Traverse each vehicle trajectory in the first vehicle trajectory set, and add the vehicle trajectory in the occlusion range to the second vehicle trajectory set.

[0167] In some embodiments, the traversal module 220 is further configured to:

[0168] Obtaining a first count value of successful association between the to-be-determined vehicle trajectory and the original data, and a second count value of successful association between each vehicle trajectory in the second vehicle trajectory set and the original data;

[0169] The second vehicle trajectory set is traversed, and the vehicle trajectories corresponding to the second count value being greater than the first count value are deleted from the second vehicle trajectory set.

[0170] In some embodiments, the traversal module 220 is further configured to:

[0171] Establish a wave gate matrix and set weights for different dimensions of the wave gate matrix;

[0172] Calculate the weighted sum of the differences between the to-be-determined vehicle trajectory and other vehicle trajectories in the first vehicle trajectory set in different dimensions of the gate matrix according to the weights;

[0173] The degree of association between the to-be-determined vehicle trajectory and other vehicle trajectories in the first vehicle trajectory set is determined according to the weighted sum.

[0174] In some embodiments, the computing module 230 is further configured to:

[0175] Calculating the difference between the speed of the to-be-determined vehicle trajectory during the historical update process and the speed of each vehicle trajectory in the second vehicle trajectory set during the historical update process, including:

[0176] Obtaining a first trajectory speed table corresponding to the to-be-determined vehicle trajectory and a second trajectory speed table corresponding to each vehicle trajectory in the second vehicle trajectory set; wherein the trajectory speed table records the speed of the vehicle trajectory at each association update;

[0177] The difference value is calculated based on the speed difference between the first track speed table and the second track speed table.

[0178] In some embodiments, the fusion module 240 is further configured to:

[0179] Adding the vehicle trajectory whose difference value is less than the second preset value and the vehicle trajectory to be determined to the split vehicle trajectory set;

[0180] The set of split vehicle trajectories is traversed, and the vehicle trajectory located at the front in the direction of vehicle travel is taken as the head of the split vehicle group for fusion to obtain the vehicle trajectory of the split vehicle group.

[0181] In some embodiments, the fusion module 240 is further configured to:

[0182] When the number of vehicle trajectories in the split vehicle trajectory set is greater than 2, the vehicle type of the split vehicle group is determined to be a large vehicle;

[0183] When the number of vehicle trajectories in the split vehicle trajectory set is less than or equal to 2, the vehicle type of the head of the split vehicle group is determined as the vehicle type of the split vehicle group.

[0184] In some embodiments, the fusion module 240 is further configured to:

[0185] Add splitting marks to the vehicle trajectories in the splitting vehicle trajectory set except the head of the splitting vehicle group;

[0186] In the process of associating and updating the vehicle trajectories, if the vehicle trajectory with the splitting flag does not belong to the vehicle trajectory of the split vehicle group, the splitting flag is deleted.

[0187] The radar data processing device based on traffic detection in the embodiment of the present application can be an electronic device, or a component in an electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal, or other devices other than a terminal. Exemplarily, the electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, a vehicle-mounted electronic device, a mobile Internet device (Mobile Internet Device, MID), an augmented reality (augmented reality, AR) / virtual reality (virtual reality, VR) device, a robot, a wearable device, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a netbook or a personal digital assistant (personal digital assistant, PDA), etc. It can also be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television, TV), a teller machine or a self-service machine, etc., and the embodiment of the present application is not specifically limited.

[0188] The radar data processing device based on traffic detection in the embodiment of the present application may be a device having an operating system. The operating system may be a Microsoft (Windows) operating system, an Android (Android) operating system, an IOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0189] In some embodiments, Figure 3As shown, the embodiment of the present application also provides an electronic device 300, including a processor 301, a memory 302, and a computer program stored in the memory 302 and executable on the processor 301. When the program is executed by the processor 301, each process of the above-mentioned radar data processing method embodiment based on traffic detection is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0190] It should be noted that the electronic devices in the embodiments of the present application include the above-mentioned mobile electronic devices and non-mobile electronic devices.

[0191] The embodiment of the present application also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the above-mentioned radar data processing method embodiment based on traffic detection are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0192] The processor is the processor in the electronic device in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0193] An embodiment of the present application also provides a computer program product, including a computer program, which implements the above-mentioned radar data processing method based on traffic detection when executed by a processor.

[0194] The processor is the processor in the electronic device in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0195] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned radar data processing method embodiment based on traffic detection, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0196] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0197] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0198] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0199] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

[0200] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0201] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present application, and that the scope of the present application is defined by the claims and their equivalents.

Claims

1. A radar data processing method based on traffic detection, characterized in that: include: The original data periodically output by the radar is associated and updated with the existing vehicle trajectory to generate the first vehicle trajectory set of the current period; Traversing the correlation degree between the to-be-determined vehicle trajectory in the first vehicle trajectory set and other vehicle trajectories in the first vehicle trajectory set, and adding the vehicle trajectory whose correlation degree with the to-be-determined vehicle trajectory is greater than a first preset value to the second vehicle trajectory set; Calculate the difference between the speed of the to-be-determined vehicle trajectory during the historical update process and the speed of each vehicle trajectory in the second vehicle trajectory set during the historical update process; The vehicle trajectory whose difference value is less than the second preset value is merged with the vehicle trajectory to be determined.

2. The method according to claim 1, characterized in that The method further comprises: Calculating the occlusion range of the to-be-determined vehicle trajectory; Each vehicle trajectory in the first vehicle trajectory set is traversed, and the vehicle trajectory in the occlusion range is added to the second vehicle trajectory set.

3. The method according to claim 1, characterized in that The method further comprises: Acquire a first count value of successful association between the to-be-determined vehicle trajectory and the original data, and a second count value of successful association between each vehicle trajectory in the second vehicle trajectory set and the original data; The second vehicle trajectory set is traversed, and vehicle trajectories corresponding to the second count value being greater than the first count value are deleted from the second vehicle trajectory set.

4. The method according to claim 1, characterized in that: The traversing the correlation degree between the to-be-determined vehicle trajectory in the first vehicle trajectory set and other vehicle trajectories in the first vehicle trajectory set includes: Establish a wave gate matrix and set weights for different dimensions of the wave gate matrix; Calculate, according to the weight, a weighted sum of differences between the to-be-determined vehicle trajectory and other vehicle trajectories in the first vehicle trajectory set in different dimensions of the wave gate matrix; The degree of association between the to-be-determined vehicle trajectory and other vehicle trajectories in the first vehicle trajectory set is determined according to the weighted sum.

5. The method according to claim 1, characterized in that The calculating the difference between the speed of the to-be-determined vehicle trajectory during the historical update process and the speed of each vehicle trajectory in the second vehicle trajectory set during the historical update process includes: Obtaining a first trajectory speed table corresponding to the to-be-determined vehicle trajectory and a second trajectory speed table corresponding to each vehicle trajectory in the second vehicle trajectory set; wherein the trajectory speed table records the speed of the vehicle trajectory at each association update; The difference value is calculated based on the speed difference between the first track speed table and the second track speed table.

6. The method according to claim 1, characterized in that The step of fusing the vehicle trajectory whose difference value is less than the second preset value with the to-be-determined vehicle trajectory includes: Adding the vehicle trajectory whose difference value is less than the second preset value and the to-be-determined vehicle trajectory to the split vehicle trajectory set; The split vehicle trajectory set is traversed, and the vehicle trajectory located at the front in the vehicle's traveling direction is taken as the head of the split vehicle group for fusion to obtain the vehicle trajectory of the split vehicle group.

7. The method according to claim 6, characterized in that The method further comprises: When the number of vehicle trajectories in the split vehicle trajectory set is greater than 2, determining the vehicle type of the split vehicle group as a large vehicle; When the number of vehicle trajectories in the split vehicle trajectory set is less than or equal to 2, the vehicle type of the head of the split vehicle group is determined as the vehicle type of the split vehicle group.

8. The method according to claim 6, characterized in that The method further comprises: Adding a splitting mark to the vehicle tracks in the splitting vehicle track set except the head of the splitting vehicle group; In the process of associating and updating the vehicle trajectories, if the vehicle trajectory with the splitting mark does not belong to the vehicle trajectory of the split vehicle group, the splitting mark is deleted.

9. A radar data processing device based on traffic detection, characterized in that: include: An association module, used to associate and update the original data periodically output by the radar with the existing vehicle trajectory to generate a first vehicle trajectory set of the current period; A traversal module, configured to traverse the correlation between the to-be-determined vehicle trajectory in the first vehicle trajectory set and other vehicle trajectories in the first vehicle trajectory set, and add a vehicle trajectory whose correlation with the to-be-determined vehicle trajectory is greater than a first preset value to the second vehicle trajectory set; A calculation module, used for calculating the difference between the speed of the to-be-determined vehicle trajectory during the historical update process and the speed of each vehicle trajectory in the second vehicle trajectory set during the historical update process; A fusion module is used to fuse the vehicle trajectory whose difference value is less than the second preset value with the vehicle trajectory to be determined.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images