Roadside unit credibility rating method and system based on data skeleton similarity

Abstract

The invention belongs to the technical field of intelligent traffic, and particularly discloses a roadside unit credibility rating method and system based on data skeleton similarity, and the method comprises the following steps: a request vehicle sends request information to a roadside unit, the roadside unit extracts the position (x, y), speed v and direction angle theta features in the information sent by the request vehicle, and the position (x, y), speed v and direction angle theta features are obtained; the method comprises the following steps: constructing a feature matrix, calculating the similarity between the change trend and the direction angle of the spatial position difference and the relative speed difference between a request vehicle and a surrounding target vehicle, judging the visibility of the surrounding target vehicle to the request vehicle through a geometric shielding vector cross product method, and returning the visibility to an on-board unit (OBU) of the request vehicle. And the OBU calculates the credibility of the data of the road side unit. By adopting the technical scheme, the credibility of the data acquired by the road side unit is calculated, the accuracy is higher, and the application of the acquired data in the vehicle end decision is determined.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent transportation technology and relates to a roadside unit credibility rating method and system based on data skeleton similarity. Background Art

[0002] With the rapid growth of motor vehicle numbers and the increasing complexity of urban traffic conditions, ensuring traffic safety and improving traffic efficiency have become two core challenges in current urban development. Against this backdrop, the intelligent transportation industry has entered a period of rapid development, with the emergence of numerous solutions based on both single-vehicle intelligence and connected intelligence.

[0003] However, considering the complexity of urban road conditions, single-vehicle intelligence is difficult to provide a comprehensive solution.

[0004] In contrast, the vehicle-road cooperative system integrates dynamic traffic information from people, vehicles, roads, and the cloud, and uses vehicle-to-everything (V2X) communication technology to achieve information interconnection, allowing traffic information such as vehicle conditions and road conditions to be shared, thereby effectively assisting vehicles and other traffic participants in making safer and more efficient decisions. In this context, vehicles in the vehicle-road cooperative system receive and process data from roadside units through on-board units (OBUs). This data is divided into different levels based on the degree of its impact on vehicle control. However, due to differences in the time and spatial location of data collection between roadside units and vehicles, inconsistencies may occur between the data.

[0005] Existing technologies fail to account for the computing power limitations of onboard terminals and filter data sent back by roadside units. This results in significant latency in vehicle-side processing and response, impacting the real-time nature of the data. Furthermore, they fail to compare the data packets transmitted between the roadside unit and the onboard terminal, failing to account for differences in data collected by the roadside unit and the vehicle itself. This impacts the accuracy of decisions made by the onboard terminal using data collected by the roadside unit. Summary of the Invention

[0006] The purpose of the present invention is to provide a roadside unit credibility rating method and system based on data skeleton similarity, so as to obtain accurate roadside unit data credibility and solve the problem of differences between roadside units and vehicle data.

[0007] To achieve the above objectives, the basic solution of the present invention is: a roadside unit credibility rating method based on data skeleton similarity, comprising the following steps:

[0008] The requesting vehicle sends a request message to the roadside unit, and the roadside unit selects a target vehicle within a radius R around the requesting vehicle according to the location of the requesting vehicle and determines a target time point;

[0009] The roadside unit extracts the position (x, y), speed v, and direction angle θ features from the information sent by the requesting vehicle based on the target time point and constructs a feature matrix.

[0010] Based on the feature matrix, the spatial position difference, relative speed difference change trend and direction angle similarity between the request vehicle and the surrounding target vehicles are calculated respectively;

[0011] Based on the spatial position difference, relative speed difference and direction angle similarity between the requesting vehicle and surrounding target vehicles, the visibility of surrounding target vehicles to the requesting vehicle is determined by the vector cross product method of geometric occlusion.

[0012] The roadside unit transmits the visibility data of surrounding target vehicles to the requesting vehicle back to the on-board unit (OBU) of the requesting vehicle. The OBU matches the target vehicle information in the visibility data with the data of the target vehicle collected by the OBU itself, aligns and compares them, and calculates the credibility of the roadside unit data.

[0013] The working principle and beneficial effects of this basic solution are as follows: This technical solution first proactively sends a request to the roadside unit (RSU) through the vehicle's onboard unit (OBU) to obtain local information about its surrounding environment. Upon receiving the request, the RSU generates and transmits a highly customized data packet based on the vehicle's specific needs and current environmental conditions. Upon receiving this data packet, the OBU extracts information related to the data collected by the vehicle's own sensors, aligns and compares it, and calculates the credibility of the RSU data.

[0014] This not only enhances the vehicle's ability to perceive external information, but multi-sensor fusion technology further increases overall data accuracy, thereby reducing the limitations of a single data source. This improves the reliability and stability of the vehicle-infrastructure cooperative system and enhances the vehicle's decision-making capabilities in complex traffic environments, thereby improving overall driving safety and traffic efficiency. This effectively resolves the discrepancy between roadside unit and vehicle data and promotes the widespread application of vehicle-infrastructure cooperative technology in real-world urban traffic management.

[0015] Furthermore, the requesting vehicle sends a request message to the roadside unit, and the roadside unit selects a target vehicle within a radius R around the requesting vehicle according to the location of the requesting vehicle, and determines the target time point in the following manner:

[0016] Request vehicle A to s The driving information of the vehicle itself and surrounding vehicles is collected at fixed time intervals. A Request vehicle A to send the collected current position information, timestamp and collection frequency information as a high-frequency measurement data set to the roadside unit at any time;

[0017] The roadside unit queries N target vehicles within a radius R around vehicle A based on the location information and calculates the target time point T 目标 :

[0018] T 目标 =T A +T s .

[0019] Obtain the target time point for subsequent use.

[0020] Furthermore, the roadside unit filters and extracts the position (x, y), speed v, and direction angle θ features in the information sent by the requesting vehicle according to the target time point, and constructs the feature matrix as follows:

[0021] The roadside unit selects the timestamp and T from the high-frequency measurement data set received from the requesting vehicle 目标 For the same or closest set of data, extract the position (x, y), velocity v, and direction angle θ features and construct the feature matrix X:

[0022]

[0023] Among them, x N ,y N , v N ,θ N are the horizontal coordinate, vertical coordinate, speed and direction angle characteristics of the Nth vehicle respectively.

[0024] Construct a feature matrix and filter out the required feature data for subsequent visibility screening.

[0025] Furthermore, based on the feature matrix, the steps for calculating the spatial position difference between the request vehicle and the surrounding target vehicles are as follows:

[0026] Perform maximum and minimum normalization on the eigenvalues ​​in the feature matrix X so that each eigenvalue is scaled to the range of 0 to 1:

[0027]

[0028] Among them, X i ' j is the characteristic matrix after maximum and minimum normalization processing; X ij is the jth eigenvalue of the i-th vehicle, X j is the jth eigenvalue of the eigenvalue matrix;

[0029] For the position feature value (x, y), use the Mahalanobis distance to calculate the spatial position difference D between the request vehicle A and the surrounding target vehicles 马氏距离(i,j) :

[0030]

[0031] Among them, P A is the position feature vector of the requesting vehicle A, P 其他车辆 is the position feature vector of the surrounding target vehicles, and S is the covariance matrix of the position feature.

[0032] For the position feature value (x, y), the Mahalanobis distance is used to calculate the spatial position difference between the request vehicle A and the surrounding target vehicles. The operation is simple.

[0033] Furthermore, based on the feature matrix, the change trend Δv of the relative speed difference between the requesting vehicle and the surrounding target vehicles is calculated. 趋势 ,for:

[0034]

[0035] Among them, v A (t) is the speed of the requesting vehicle A at time t, v B (t) is the velocity of the surrounding target vehicles at time t.

[0036] For the rate characteristic value, the changing trend of the relative speed difference between the two vehicles is calculated for subsequent use.

[0037] Furthermore, based on the feature matrix, the similarity of the direction angles of the requesting vehicle and the surrounding target vehicles is calculated as follows:

[0038] Calculate the request vehicle A to point to the target vehicle B i Direction vector

[0039]

[0040] in, The target vehicle B i Location information; x A ,y A To request the location information of vehicle A;

[0041] Calculate the direction vector and the target vehicle B i The angle cos(Δθ) between the driving directions is used to measure the direction consistency by cosine similarity:

[0042]

[0043] in, Indicates the target vehicle B i direction of travel.

[0044] For the direction angle feature value, the similarity of the direction angles of the requesting vehicle A and the surrounding vehicles is calculated for easy use.

[0045] Furthermore, the current visibility of the surrounding target vehicles to the requesting vehicle is determined by the vector cross product method of geometric occlusion. The specific steps are as follows:

[0046] For request vehicle A and target vehicle B i , define a line of sight segment L: from the position of the requesting vehicle A (x A ,y A ) to target vehicle B i Location

[0047] For the obstacle vehicle B j , whose rectangular boundary consists of four vertices (x j1 ,y j1 ),(x j2 ,y j2 ),(x j3 ,y j3 ),

[0048] (x j4 ,y j4 ) constituted as the four boundary segments of the vehicle, used to determine whether the line of sight is blocked;

[0049] Based on the vector cross product method, line segment intersection judgment is performed, specifically:

[0050] Direction vector of line of sight L Defined as:

[0051]

[0052] in, The target vehicle B i Location information; x A ,y A To request the location information of vehicle A;

[0053] Obstacle boundary segment for:

[0054]

[0055] For these two vectors, use the cross product to determine whether they are parallel. If they are not parallel, then further determine whether they intersect.

[0056] Determine the endpoint position of the line segment:

[0057] For non-parallel sight line segment L and obstacle vehicle B j Boundary line segments, determine whether their endpoints are on different sides of the other line segment. Assume that the obstacle vehicle B needs to be determined j The boundary line segment is from point B j1 =(x j1 ,yj1 ) to point B j2 =(x j2 ,y j2 ):

[0058]

[0059] If the signs of D1 and D2 are different, it means that the obstacle vehicle B j The two endpoints of the boundary segment are on different sides of the sight segment, which means that the sight segment and the obstacle vehicle B j The boundary segments may intersect, so mark the target vehicle B i Is invisible.

[0060] The vehicle screening method based on vector cross product visibility determines the current visibility of surrounding vehicles to the requesting vehicle by the vehicle position, reducing the data calculation of the on-board terminal for invisible vehicles and alleviating the computing load of the on-board terminal.

[0061] Furthermore, the steps for determining the future visibility of the surrounding target vehicles to the requesting vehicle are as follows:

[0062] At the future time point t+Δt, the future relative velocity v 相对 (t+Δt) is calculated by the current relative velocity v 相对 (t) and velocity trend Δv 趋势 (t) to make predictions:

[0063] v 相对 (t+Δt)=v 相对 (t)+Δv 趋势 (t)·Δt;

[0064] For a certain car B, suppose its position at time t is (x B ,y B ), velocity is v, direction angle is θ B At the future time t+Δt, the position of vehicle B (x′ B ,y′ B )for:

[0065] (x′ B ,y′ B )=(x B +v 相对 (t+Δt)·cos(θ B )·Δt,y B +v 相对 (t+Δt)·sin(θ B )·

[0066] Δt);

[0067] Based on this prediction, the request vehicle A and the target vehicle B i and obstacle vehicle B j The position at the future time point is then determined using the vector cross product method.

[0068] The vehicle screening method based on vector cross product visibility determines the current and future visibility of surrounding vehicles to the requesting vehicle through the vehicle position, reducing the data calculation of invisible vehicles by the on-board terminal and alleviating the computing load of the on-board terminal.

[0069] Further, comprehensive judgment of future visibility:

[0070] Whether the target vehicle B is visible to the requesting vehicle A at a future time depends on whether the following conditions are met:

[0071] Current visibility: target vehicle B is currently visible to requesting vehicle A;

[0072] The future occlusion situation is visible: at the future time point, the sight segments of the target vehicle B and the requesting vehicle A are not blocked;

[0073] The directions remain consistent or the difference is small: cos(Δθ)≈1, that is, the direction consistency between the target vehicle B and the request vehicle A is high;

[0074] Speed ​​trend shows close to: v 相对 (t+Δt)<0, the speed trend of target vehicle B shows that it is gradually approaching request vehicle A;

[0075] When all conditions are met, the target vehicle B is considered visible to the requesting vehicle A at the current moment and in the future, which means it passes the visibility screening;

[0076] Traverse the feature matrix to filter and return the data:

[0077] Traverse all vehicles in the feature matrix and filter their visibility;

[0078] For vehicles that pass the visibility screening, the roadside unit will send multiple sets of measurement data of the vehicle at different times to the requesting vehicle A for continuous calculation of credibility.

[0079] The original vehicle driving data collected by the roadside unit is filtered based on current and future visibility, and the vehicle nodes with a strong correlation with the requesting vehicle are selected and transmitted back. This can effectively reduce the computing load of the on-board terminal in dynamic traffic scenarios with heavy traffic volume, while ensuring the accuracy of the credibility calculation.

[0080] Furthermore, the method for the on-board unit OBU to calculate the credibility of the roadside unit data is:

[0081] Data reception and matching: After receiving the data returned from the roadside unit, the requesting vehicle A identifies the corresponding vehicle from the vehicles it has detected based on the vehicle information;

[0082] Data validity check: After identifying the corresponding vehicle, check whether the characteristic values ​​in the returned data are valid data, specifically:

[0083] If a key feature is invalid data, the data of the target vehicle is discarded;

[0084] When non-critical features have null values, or are only partially measured at the moment, data filling can be performed;

[0085] Based on the graph structure formed by the roadside unit for the requesting vehicle A and its surrounding vehicles that have passed the screening, and the graph structure generated by the requesting vehicle A and the same surrounding vehicles, the similarity scores of edges and nodes are calculated through matrix iteration to evaluate their overall similarity. Specifically,

[0086] Based on the measurement and screening of the roadside unit, an undirected complete graph G is generated for the requesting vehicle A and its surrounding vehicles that have passed the screening A , based on the measurement of the requesting vehicle A and the surrounding vehicles that are the same as those filtered by the roadside unit, the graph G is generated in the same way B ;

[0087] Calculate the node similarity score S through matrix iteration N and edge similarity score S E , and finally calculate the overall similarity score S G :

[0088] S G =S N +S E

[0089] Compare S G With τ G , τ G is a pre-set threshold, when S G >τ G When , it is considered to have reached a low credibility level and enters the next step of calculating the credibility of the speed measurement value;

[0090] Calculate the mean square error (MSE) of the speed measurements of the roadside unit and the requesting vehicle A for the same surrounding vehicles v :

[0091]

[0092] Among them, v R,t is the speed measurement of the target vehicle by the roadside unit at time t, v A,tis the speed measurement of the target vehicle by requesting vehicle A at time t, and N is the total number of surrounding vehicles;

[0093] Comparing MSE v With τ v , τ v is a pre-set threshold, when MSE v <τ v When , it is considered to have reached the medium reliability level, and the next step is to calculate the reliability of the direction angle measurement value;

[0094] Calculate the mean square error (MSE) of the direction angle measurements of the roadside unit and the requesting vehicle A for the same surrounding vehicles;

[0095] Consider the periodicity of the direction angle:

[0096] Δθ t =min(|θ R,t -θ A,t |,360°-|θ R,t -θ A,t |)

[0097] Among them, θ R,t is the direction angle measured by the roadside unit to the target vehicle at time t, θ A,t is the direction angle measured by the requesting vehicle A to the target vehicle at time t, Δθ t is the difference in the direction angle measurement of the target vehicle between the roadside unit and the requesting vehicle A at time t, ranging from [0°, 180°];

[0098] Direction angle mean square error (MSE) θ for:

[0099]

[0100] Comparing MSE θ With τ θ , τ θ is a pre-set threshold, when MSE θ <τ θ , it is considered to have reached a high confidence level.

[0101] For the data skeletons constructed based on the data sent back by the roadside unit and the detection data of the requesting vehicle itself, the graph similarity is first calculated to determine the similarity from the overall perspective, and then the differences in the remaining eigenvalues ​​are calculated to gradually classify them. This can effectively evaluate and classify the credibility of roadside units with different collected data differences, thereby determining the application of their collected data in vehicle-side decision-making.

[0102] Furthermore, according to the pre-set τ 位置 , τ v , τ θ, and the similarity S to the whole graph G and MSE v and MSE θ The calculation results are used to assign different credibility levels C 等级 :

[0103]

[0104] Based on the credibility level, it is decided whether the requesting vehicle A adopts the data provided by the roadside unit for subsequent decision making.

[0105] The credibility grading calculation method based on the similarity of the overall graph and the difference of each item is constructed by constructing a data skeleton based on the data collected by the roadside unit and the vehicle terminal. It realizes the similarity analysis and calculation from the overall level to the specific feature value level to evaluate the credibility level of the roadside unit data.

[0106] The present invention also provides a roadside unit credibility rating system based on the method of the present invention, comprising a roadside unit and an onboard unit (OBU), wherein the onboard unit (OBU) is used to collect the current requesting vehicle's own location information, timestamp, and collection frequency information, and send the information to the roadside unit;

[0107] The roadside unit is used to obtain visibility data of surrounding target vehicles to the requesting vehicle and transmit it back to the on-board unit (OBU) of the requesting vehicle. The on-board unit (OBU) calculates the credibility of the roadside unit data.

[0108] The system processes the raw vehicle driving data collected by the roadside unit (RSU), adds the requesting vehicle's timestamp and location, and returns the vehicle data, filtered by visibility, to the requesting vehicle. The vehicle terminal then constructs a data skeleton based on the returned vehicle feature data and compares it with its own detected data to calculate similarity and evaluate the RSU's credibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0109] Figure 1 It is a flow chart of the roadside unit credibility rating method based on data skeleton similarity of the present invention. DETAILED DESCRIPTION

[0110] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0111] In the description of the present invention, it should be understood that the terms "longitudinal", "transverse", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention.

[0112] In the description of the present invention, unless otherwise specified and limited, it should be noted that the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a mechanical connection or an electrical connection, or it can be the internal communication between two components. It can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to the specific circumstances.

[0113] The present invention discloses a roadside unit credibility rating method based on data skeleton similarity, which realizes data verification and credibility evaluation between vehicles and roadside units through an interactive data request and response mechanism. Figure 1 As shown in FIG, the roadside unit credibility rating method based on data skeleton similarity includes the following steps:

[0114] The requesting vehicle sends a request message to the roadside unit (the vehicle's onboard unit (OBU) actively sends a request to the roadside unit). The roadside unit selects a target vehicle within a radius R around the requesting vehicle based on the requesting vehicle's location and determines a target time point.

[0115] The roadside unit extracts the position (x, y), speed v, and direction angle θ features from the information sent by the requesting vehicle based on the target time point and constructs a feature matrix.

[0116] Based on the feature matrix, the spatial position difference, relative speed difference change trend and direction angle similarity between the request vehicle and the surrounding target vehicles are calculated respectively;

[0117] Based on the spatial position difference between the requesting vehicle and surrounding target vehicles, the changing trend of the relative speed difference, and the similarity of the azimuth angle, the visibility of the surrounding target vehicles to the requesting vehicle is judged through the vector cross product method of geometric occlusion. From the multi-vehicle data collected by the roadside unit, the target vehicles that meet the detection range of the requesting vehicle are screened out, and the current and future visibility of the target vehicles to the requesting vehicle in dynamic traffic scenarios is taken into account, so that the on-board terminal of the requesting vehicle can calculate the temporally continuous roadside unit credibility score.

[0118] The RSU transmits visibility data of surrounding target vehicles to the requesting vehicle's OBU. The OBU then matches the target vehicle information in the visibility data with the OBU's own collected data about the target vehicle, aligning and comparing them to calculate the reliability of the RSU data. This enhances the vehicle's ability to perceive external information, and multi-sensor fusion technology further enhances overall data accuracy, reducing the limitations of a single data source.

[0119] In a preferred embodiment of the present invention, a requesting vehicle sends a request message to a roadside unit, and the roadside unit selects a target vehicle within a radius R around the requesting vehicle based on the location of the requesting vehicle, and determines the target time point in the following manner:

[0120] Request vehicle A to s The driving information of the vehicle itself and surrounding vehicles is collected at fixed time intervals. A Request vehicle A to send the collected information such as its current location, timestamp, and collection frequency as a high-frequency measurement data set to the roadside unit at any time;

[0121] The roadside unit queries N target vehicles within a radius R around vehicle A based on the location information and calculates the target time point T 目标 :

[0122] T 目标 =T A +T s .

[0123] In a preferred embodiment of the present invention, the roadside unit extracts the position (x, y), speed v, and direction angle θ features from the information sent by the requesting vehicle according to the target time point, and constructs the feature matrix in the following manner:

[0124] The roadside unit selects the timestamp and T from the high-frequency measurement data set received from the requesting vehicle 目标 For the same or closest set of data, extract the position (x, y), velocity v, and direction angle θ features and construct the feature matrix X:

[0125]

[0126] Among them, x N ,y N , v N ,θ N are the horizontal coordinate, vertical coordinate, speed and direction angle characteristics of the Nth vehicle respectively.

[0127] In a preferred embodiment of the present invention, the step of calculating the spatial position difference between the requesting vehicle and the surrounding target vehicles based on the feature matrix is ​​as follows:

[0128] Perform maximum and minimum normalization on the eigenvalues ​​in the feature matrix X so that each eigenvalue is scaled to the range of 0 to 1:

[0129]

[0130] Among them, X i ' j is the characteristic matrix after maximum and minimum normalization processing; X ij is the jth eigenvalue of the i-th vehicle, X j is the jth eigenvalue of the eigenvalue matrix; in the characteristic matrix, the jth eigenvalue of the first row X 1j , and the j-th eigenvalue X in the second row 2j , respectively represent the jth eigenvalue of the first car and the jth eigenvalue of the second car. So from the perspective of matrix rows and columns, X j Refers to the jth column, which is the jth eigenvalue of all vehicles.

[0131] For the position feature value (x, y), use the Mahalanobis distance to calculate the spatial position difference D between the request vehicle A and the surrounding target vehicles 马氏距离(i,j) :

[0132]

[0133] Among them, P A is the position feature vector of the requesting vehicle A, P 其他车辆 is the position feature vector of the surrounding target vehicles, and S is the covariance matrix of the position feature.

[0134] In a preferred embodiment of the present invention, based on the characteristic matrix, the change trend Δv of the relative speed difference between the requesting vehicle and the surrounding target vehicles is calculated. 趋势 ,for:

[0135]

[0136] Among them, v A (t) is the speed of the requesting vehicle A at time t, v B (t) is the velocity of the surrounding target vehicles at time t.

[0137] In a preferred embodiment of the present invention, the similarity of the direction angles of the requesting vehicle and the surrounding target vehicles is calculated based on the feature matrix, and the steps are as follows:

[0138] Calculate the request vehicle A to point to the target vehicle B i Direction vector

[0139]

[0140] in, The target vehicle B i Location information; x A ,y A To request the location information of vehicle A;

[0141] Calculate the direction vector and the target vehicle B i The angle cos(Δθ) between the driving directions is used to measure the direction consistency by cosine similarity:

[0142]

[0143] in, Indicates the target vehicle B i direction of travel.

[0144] In a preferred embodiment of the present invention, the current visibility of the surrounding target vehicles to the requesting vehicle is determined by the vector cross product method of geometric occlusion. The specific steps are as follows:

[0145] For request vehicle A and target vehicle B i , define a line of sight segment L: from the position of the requesting vehicle A (x A ,y A ) to target vehicle B i Location

[0146] For the obstacle vehicle B j , whose rectangular boundary consists of four vertices (x j1 ,y j1 ),(x j2 ,y j2 ),(x j3 ,y j3 ),

[0147] (x j4 ,y j4 ) constituted as the four boundary segments of the vehicle, used to determine whether the line of sight is blocked;

[0148] Based on the vector cross product method, line segment intersection judgment is performed, specifically:

[0149] Direction vector of line of sight L Defined as:

[0150]

[0151] in, The target vehicle B i Location information; x A ,y A To request the location information of vehicle A;

[0152] Obstacle boundary segment (For example, vehicle B j A certain boundary of ) is:

[0153]

[0154] For these two vectors, use the cross product to determine whether they are parallel. If they are not parallel, then further determine whether they intersect.

[0155] Determine the endpoint position of the line segment:

[0156] For non-parallel sight line segment L and obstacle vehicle B j Boundary line segments, determine whether their endpoints are on different sides of the other line segment. Assume that the obstacle vehicle B needs to be determined j The boundary line segment is from point B j1 =(x j1 ,y j1 ) to point B j2 =(x j2 ,y j2 ):

[0157]

[0158] If the signs of D1 and D2 are different, it means that the obstacle vehicle B j The two endpoints of the boundary segment are on different sides of the sight segment, which means that the sight segment and the obstacle vehicle B j The boundary segments may intersect, so mark the target vehicle B i Is invisible.

[0159] In a preferred embodiment of the present invention, the steps of determining the future visibility of the surrounding target vehicles to the requesting vehicle are as follows:

[0160] At the future time point t+Δt, the future relative velocity v 相对 (t+Δt) is calculated by the current relative velocity v 相对 (t) and velocity trend Δv 趋势 (t) to make predictions:

[0161] v 相对 (t+Δt)=v 相对 (t)+Δv 趋势 (t)·Δt;

[0162] Future position prediction: For a certain vehicle B, suppose its position at time t is (x B ,y B ), velocity is v, direction angle is θ B At the future time t+Δt, the position of vehicle B (x′ B ,y′ B )for:

[0163] (x′ B ,y′ B )=(x B +v 相对 (t+Δt)·cos(θ B )·Δt,y B +v 相对 (t+Δt)·sin(θ B )·

[0164] Δt);

[0165] Based on this prediction, the request vehicle A and the target vehicle B i and obstacle vehicle B j Position at a future time point (B j For an obstacle vehicle, it is necessary to determine whether it blocks the path from vehicle A to target vehicle B. i Line of sight segment, thus judging B i In the future for the visibility of A), and then use the vector cross product method to determine the visibility.

[0166] In a preferred embodiment of the present invention, the future visibility is comprehensively judged as follows:

[0167] Whether the target vehicle B is visible to the requesting vehicle A at a future time depends on whether the following conditions are met:

[0168] Current visibility: target vehicle B is currently visible to requesting vehicle A;

[0169] The future occlusion situation is visible: at the future time point, the sight segments of the target vehicle B and the requesting vehicle A are not blocked;

[0170] The directions remain consistent or the difference is small: cos(Δθ)≈1, that is, the direction consistency between the target vehicle B and the request vehicle A is high;

[0171] Speed ​​trend shows close to: v 相对 (t+Δt)<0, the speed trend of target vehicle B shows that it is gradually approaching request vehicle A;

[0172] When all conditions are met, the target vehicle B is considered visible to the requesting vehicle A at the current moment and in the future, which means it passes the visibility screening;

[0173] Traverse the feature matrix to filter and return the data:

[0174] Traverse all vehicles in the feature matrix and filter their visibility;

[0175] For vehicles that pass the visibility screening, the roadside unit will send multiple sets of measurement data of the vehicle at different times to the requesting vehicle A for continuous calculation of credibility.

[0176] In a preferred embodiment of the present invention, the method for the on-board unit OBU to calculate the credibility of the roadside unit data is:

[0177] Data reception and matching: After receiving the data returned from the roadside unit, the requesting vehicle A identifies the corresponding vehicle from the vehicles it has detected based on the vehicle information (vehicle ID, etc.);

[0178] Data validity check: After identifying the corresponding vehicle, check whether the characteristic values ​​in the returned data are valid data, specifically:

[0179] If a key feature (such as position or speed) is invalid data, it will significantly affect the accuracy of the credibility assessment and the data of the target vehicle will be discarded. When a non-key feature (such as direction angle) is null, or only partially measured at the moment, data filling can be performed: Filling with data from the last measurement: If a feature is missing at the current moment, you can try to fill it with the same feature value field from the last measurement. If a continuous feature is missing, the null value can be filled by linear interpolation or estimation based on data from multiple previous and subsequent time points.

[0180] Low confidence calculation based on graph structure similarity:

[0181] Based on the graph structure formed by the roadside unit for the requesting vehicle A and its surrounding vehicles that have passed the screening, and the graph structure generated by the requesting vehicle A and the same surrounding vehicles, the similarity scores of edges and nodes are calculated through matrix iteration to evaluate their overall similarity. Specifically,

[0182] Based on the measurement and screening of the roadside unit, an undirected complete graph G is generated for the requesting vehicle A and its surrounding vehicles that have passed the screening A , based on the measurement of the requesting vehicle A and the surrounding vehicles that are the same as those filtered by the roadside unit, the graph G is generated in the same way B ;

[0183] Calculate the node similarity score S through matrix iteration N and edge similarity score S E , and finally calculate the overall similarity score S G :

[0184] S G =S N +S E

[0185] Compare S G With τ G , τ G is a pre-set threshold, when S G >τ GWhen , it is considered to have reached a low credibility level and enters the next step of calculating the credibility of the speed measurement value;

[0186] Calculation of medium credibility based on rate characteristic value:

[0187] Calculate the mean square error (MSE) of the speed measurements of the roadside unit and the requesting vehicle A for the same surrounding vehicles v :

[0188]

[0189] Among them, v R,t is the speed measurement of the target vehicle by the roadside unit at time t, v A,t is the speed measurement of the target vehicle by requesting vehicle A at time t, and N is the total number of surrounding vehicles;

[0190] Comparing MSE v With τ v , τ v is a pre-set threshold, when MSE v <τ v When , it is considered to have reached the medium reliability level, and the next step is to calculate the reliability of the direction angle measurement value;

[0191] High confidence calculation based on azimuth eigenvalues:

[0192] Calculate the mean square error (MSE) of the direction angle measurements of the roadside unit and the requesting vehicle A for the same surrounding vehicles;

[0193] Consider the periodicity of the direction angle:

[0194] Δθ t =min(|θ R,t -θ A,t |,360°-|θ R,t -θ A,t |)

[0195] Among them, θ R,t is the direction angle measured by the roadside unit to the target vehicle at time t, θ A,t is the direction angle measured by the requesting vehicle A to the target vehicle at time t, Δθ t is the difference in the direction angle measurement of the target vehicle between the roadside unit and the requesting vehicle A at time t, ranging from [0°, 180°];

[0196] Direction angle mean square error (MSE) θ for:

[0197]

[0198] Comparing MSE θ With τ θ , τθ is a pre-set threshold, when MSE θ <τ θ , it is considered to have reached a high confidence level.

[0199] More preferably, according to the pre-set τ 位置 , τ v , τ θ , and the similarity S to the whole graph G and MSE v and MSE θ The calculation results are used to assign different credibility levels C 等级 :

[0200]

[0201] Based on the credibility level, the system determines whether requesting vehicle A will use the data provided by the roadside unit for subsequent decision-making. Through comprehensive analysis, differential analysis and calculations are performed step by step, from the overall data to the specific feature value level. This system further assesses the credibility of the roadside unit data in a graded manner, providing a reliable reference for intelligent vehicle decision-making.

[0202] The present invention also provides a roadside unit credibility rating system based on the method of the present invention, comprising a roadside unit and an onboard unit (OBU). The onboard unit (OBU) is used to collect current location information, timestamp and collection frequency information, and send it to the roadside unit.

[0203] The roadside unit is used to obtain visibility data of surrounding target vehicles to the requesting vehicle and transmit it back to the on-board unit (OBU) of the requesting vehicle. The on-board unit (OBU) calculates the credibility of the roadside unit data.

[0204] The present invention processes the raw vehicle driving data collected by the roadside unit, combines it with the requesting vehicle's timestamp and location, and returns the vehicle data, filtered by visibility, to the requesting vehicle. The vehicle terminal then constructs a data skeleton based on the returned vehicle feature value data, compares it with its own detected data, and calculates similarity to evaluate the roadside unit's credibility.

[0205] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0206] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

Claims

1. A roadside unit credibility rating method based on data skeleton similarity, characterized in that: The steps include: The requesting vehicle sends a request message to the roadside unit, and the roadside unit selects a target vehicle within a radius R around the requesting vehicle according to the position of the requesting vehicle, and determines a target time point; The roadside unit filters and extracts the position (x, y), speed v, and direction angle θ features in the information sent by the requesting vehicle according to the target time point, and constructs a feature matrix; Based on the feature matrix, the spatial position difference between the request vehicle and the surrounding target vehicles, the change trend of the relative speed difference and the similarity of the direction angle are calculated respectively; According to the spatial position difference between the requesting vehicle and the surrounding target vehicles, the changing trend of the relative speed difference and the similarity of the direction angle, the visibility of the surrounding target vehicles to the requesting vehicle is determined by the vector cross product method of geometric occlusion; The roadside unit transmits the visibility data of the surrounding target vehicles to the requesting vehicle to the on-board unit OBU of the requesting vehicle. The on-board unit OBU matches the target vehicle information in the visibility data with the data of the target vehicle collected by the on-board unit OBU itself, aligns and compares them, and calculates the credibility of the roadside unit data.

2. The roadside unit credibility rating method based on data skeleton similarity according to claim 1, characterized in that: The requesting vehicle sends a request message to the roadside unit, and the roadside unit selects a target vehicle within a radius R around the requesting vehicle according to the position of the requesting vehicle, and determines the target time point by: Request vehicle A to s The driving information of the vehicle itself and surrounding vehicles is collected at fixed time intervals. A Request vehicle A to send the collected current position information, timestamp and collection frequency information to the roadside unit as a high-frequency measurement data set at any time; The roadside unit queries the N target vehicles within a radius R around the request vehicle A based on the location information and calculates the target time point T 目标 : T 目标 =T A +T s 。 3. The roadside unit credibility rating method based on data skeleton similarity according to claim 1, characterized in that: The roadside unit extracts the position (x, y), speed v, and direction angle θ features in the information sent by the requesting vehicle according to the target time point, and constructs the feature matrix as follows: The roadside unit selects the timestamp and T from the high-frequency measurement data set received from the requesting vehicle. 目标 The same or closest set of data, extract the position (x, y), speed v, direction angle θ features, and construct the feature matrix X: Among them, x N ,y N , v N ,θ N They are respectively the horizontal coordinate, vertical coordinate, speed and direction angle characteristics of the Nth vehicle.

4. The roadside unit credibility rating method based on data skeleton similarity according to claim 1, characterized in that: Based on the feature matrix, the steps to calculate the spatial position difference between the request vehicle and the surrounding target vehicles are: Perform maximum and minimum normalization on the eigenvalues ​​in the feature matrix X so that each eigenvalue is scaled to the range of 0 to 1: Among them, X i ' j is the feature matrix after maximum and minimum normalization; X ij is the jth eigenvalue of the i-th vehicle, X j is the jth eigenvalue of the eigenvalue matrix; For the position feature value (x, y), the Mahalanobis distance is used to calculate the spatial position difference D between the request vehicle A and the surrounding target vehicles 马氏距离(i,j) : Among them, P A is the position feature vector of the requesting vehicle A, P 其他车辆 is the position feature vector of the surrounding target vehicles, and S is the covariance matrix of the position feature.

5. The roadside unit credibility rating method based on data skeleton similarity according to claim 1, characterized in that: Based on the feature matrix, calculate the change trend Δv of the relative speed difference between the requesting vehicle and the surrounding target vehicles 趋势 ,for: Among them, v A (t) is the speed of the requesting vehicle A at time t, v B (t) is the velocity of the surrounding target vehicles at time t.

6. The method for roadside unit credibility rating based on data skeleton similarity according to claim 1, characterized in that: Based on the feature matrix, the similarity of the direction angles of the request vehicle and the surrounding target vehicles is calculated in the following steps: Calculate the request vehicle A to point to the target vehicle B i Direction vector in, The target vehicle B i Location information; x A ,y A To request the location information of vehicle A; Calculate the direction vector and the target vehicle B i The angle cos(Δθ) between the driving directions is used to measure the direction consistency by cosine similarity: in, Indicates the target vehicle B i direction of travel.

7. The method for roadside unit credibility rating based on data skeleton similarity according to claim 1, characterized in that: The current visibility of the surrounding target vehicles to the requesting vehicle is determined by the vector cross product method of geometric occlusion. The specific steps are as follows: For request vehicle A and target vehicle B i , define a line of sight L: from the position of the requesting vehicle A (x A ,y A ) to the target vehicle B i Location For obstacle vehicle B j , whose rectangular boundary consists of four vertices (x j1 ,y j1 ),(x j2 ,y j2 ),(x j3 ,y j3 ), (x j4 ,y j4 ) as four boundary line segments of the vehicle, used to determine whether the line of sight is blocked; Based on the vector cross product method, the line segment intersection judgment is performed, specifically: The direction vector of the line of sight L Defined as: in, The target vehicle B i Location information; x A ,y A To request the location information of vehicle A; Obstacle boundary segment for: For these two vectors, use the cross product to determine whether they are parallel. If they are not parallel, further determine whether they intersect. Determine the endpoint position of the line segment: For non-parallel sight line segment L and obstacle vehicle B j Boundary line segments, determine whether their endpoints are on different sides of the other line segment. Assume that the obstacle vehicle B needs to be determined j The boundary line segment is from point B j1 =(x j1 ,y j1 ) to point B j2 =(x j2 ,y j2 ): If the signs of D1 and D2 are different, it means that the obstacle vehicle B j The two endpoints of the boundary segment are located on different sides of the line of sight segment, which means that the line of sight segment and the obstacle vehicle B j The boundary segments may intersect, so the target vehicle B is marked i Is invisible.

8. The method for roadside unit credibility rating based on data skeleton similarity according to claim 7, characterized in that: The steps to determine the future visibility of surrounding target vehicles to the requesting vehicle are as follows: At the future time point t+Δt, the future relative speed v 相对 (t+Δt) through the current relative velocity v 相对 (t) and velocity trend Δv 趋势 (t) to make predictions: v 相对 (t+Δt)=v 相对 (t)+Δv 趋势 (t)·Δt; For a vehicle B, suppose its position at time t is (x B ,y B ), velocity is v, direction angle is θ B , at the future time t+Δt, the position of vehicle B (x′ B ,y′ B )for: (x′ B ,y′ B )=(x B +v 相对 (t+Δt)·cos(θ B )·Δt,y B +v 相对 (t+Δt)·sin(θ B )·Δt); Based on this prediction, the request vehicle A and the target vehicle B i and obstacle vehicle B j The position at a future time point is then determined using the vector cross product method.

9. The method for roadside unit credibility rating based on data skeleton similarity according to claim 8, characterized in that: Comprehensive judgment of future visibility: Whether the target vehicle B is visible to the requesting vehicle A at a future time depends on whether the following conditions are met: Current visibility: target vehicle B is currently visible to requesting vehicle A; The future occlusion situation is visible: the sight segments of the target vehicle B and the requesting vehicle A are not blocked at the future time point; The directions are consistent or the difference is small: cos(Δθ)≈1, that is, the direction consistency between the target vehicle B and the request vehicle A is high; Speed ​​trend shows close to: v 相对 (t+Δt)<0, the speed trend of target vehicle B shows that it is gradually approaching request vehicle A; When all conditions are met, the target vehicle B is considered to be visible to the requesting vehicle A at the current moment and in the future, which means it passes the visibility screening; Traverse the feature matrix to filter and return the data: Traverse all vehicles in the feature matrix and filter their visibility; For vehicles that pass the visibility screening, the roadside unit will send multiple sets of measurement data of the vehicle at different times to the requesting vehicle A for continuous calculation of credibility.

10. The method for roadside unit credibility rating based on data skeleton similarity according to claim 1, characterized in that: The method by which the on-board unit OBU calculates the credibility of the roadside unit data is: Data reception and matching: After receiving the data returned from the roadside unit, the requesting vehicle A identifies the corresponding vehicle from the vehicles it has detected based on the vehicle information; Data validity check: After identifying the corresponding vehicle, check whether the feature value in the returned data is valid data, specifically: If a key feature is invalid data, choose to discard the data of the target vehicle; When non-critical features are empty, or only partially measured at the moment, data filling can be performed; Based on the graph structure formed by the roadside unit for the request vehicle A and its surrounding vehicles that have passed the screening, and the graph structure generated by the request vehicle A and the same surrounding vehicles, the similarity scores of the edges and nodes are calculated by matrix iteration to evaluate their overall similarity, specifically: Based on the measurement and screening of the roadside unit, an undirected complete graph G is generated for the requesting vehicle A and its surrounding vehicles that have passed the screening A , based on the measurement of the requesting vehicle A and the surrounding vehicles that are the same as those filtered by the roadside unit, the graph G is generated in the same way B ; Calculate the node similarity score S through matrix iteration N and edge similarity score S E , and finally calculate the overall similarity score S G : S G =S N +S E Compare S G With τ G , τ G is a preset threshold. When S G >τ G When , it is considered to have reached a low credibility level, and the next step is to calculate the credibility of the speed measurement value; Calculate the mean square error (MSE) of the speed measurements of the roadside unit and the requesting vehicle A for the same surrounding vehicles v : Among them, v R,t is the speed measurement of the target vehicle by the roadside unit at time t, v A,t is the measurement value of the target vehicle speed by requesting vehicle A at time t, and N is the total number of surrounding vehicles; Comparing MSE v With τ v , τ v is a pre-set threshold, when MSE v <τ v When , it is considered to have reached the medium credibility level, and the next step is to calculate the credibility of the azimuth measurement value; Calculate the mean square error (MSE) of the direction angle measurements of the roadside unit and the requesting vehicle A for the same surrounding vehicles; Consider the periodicity of the direction angle: Dth t =min(|θ R,t -θ A,t |,360°-|θ R,t -θ A,t |) Among them, θ R,t is the direction angle measured by the roadside unit to the target vehicle at time t, θ A,t is the direction angle measured by the requesting vehicle A to the target vehicle at time t, Δθ t is the difference in the direction angle measurement of the target vehicle by the roadside unit and the requesting vehicle A at time t, ranging from [0°, 180°]; Direction angle mean square error (MSE) θ for: Comparing MSE θ With τ θ , τ θ is a pre-set threshold, when MSE θ <τ θ , it is considered to have reached a high confidence level.

11. The method for roadside unit credibility rating based on data skeleton similarity according to claim 10, characterized in that: According to the preset τ 位置 , τ v , τ θ , and the similarity S to the whole graph G and MSE v and MSE θ The calculation results are used to assign different credibility levels C 等级 : Based on the credibility level, it is decided whether the requesting vehicle A adopts the data provided by the roadside unit for subsequent decision making.

12. A roadside unit credibility rating system based on the method according to any one of claims 1 to 11, characterized in that: It includes a roadside unit and an onboard unit OBU, wherein the onboard unit OBU is used to collect the current request vehicle's own location information, timestamp and collection frequency information, and send it to the roadside unit; The roadside unit is used to obtain visibility data of surrounding target vehicles to the requesting vehicle, and transmit it back to the on-board unit OBU of the requesting vehicle. The on-board unit OBU calculates the credibility of the roadside unit data.

Images