Multi-millimeter-wave radar vehicle track splicing method and system in tunnel

Abstract

The invention relates to the field of traffic monitoring, in particular to a multi-millimeter-wave radar vehicle track splicing method and system in a tunnel. The method comprises the following steps: carrying out space-time calibration on all millimeter wave radars in a tunnel; respectively acquiring a set of vehicle information of all vehicles at any same moment in the overlapping area of all adjacent millimeter wave radars; similarity of adjacent radar vehicle information; and inputting the similarity into a pre-trained BP network, and determining whether to splice the vehicle tracks or not according to a result. According to the method, the vehicle tracks between the millimeter wave radars are spliced by adopting a BP neural network method, and the problem that the splicing effect of the vehicle tracks in the tunnel is not ideal due to the fact that millimeter wave radar signals have a multipath effect in the tunnel, detection interference is large and monitoring data precision is insufficient is solved.

Description

Technical Field

[0001] The present invention relates to the field of traffic monitoring, and in particular to a method and system for splicing vehicle trajectories using multiple millimeter-wave radars in a tunnel. Background Art

[0002] Traditional highway tunnel control is based on independent equipment monitoring, lacks the means of multi-equipment linkage, and cannot effectively track the entire trajectory of the vehicle in the tunnel. Millimeter-wave radar can effectively track the speed and position of the target to form the target trajectory. However, the detection distance of a single millimeter-wave radar is limited and can only track targets within the radar detection area, and cannot track vehicles throughout the tunnel. At present, most of the trajectory splicing technologies based on millimeter-wave radars are single vehicle trajectories or splicing of multiple vehicle trajectories within a single millimeter-wave radar detection area; at the same time, in the semi-closed environment of the tunnel, the millimeter-wave radar cannot be accurately positioned by conventional means, and the millimeter-wave radar signal has a multipath effect in the tunnel, the detection interference is large, and the monitoring data accuracy is insufficient, resulting in unsatisfactory vehicle trajectory splicing in the tunnel. Summary of the invention

[0003] In view of the above problems, the present invention aims to provide a method and system for splicing vehicle trajectories of multiple millimeter-wave radars in a tunnel. The vehicle trajectories output by multiple millimeter-wave radars are spliced ​​to restore the vehicle's trajectory in the entire tunnel, solving the problem that the vehicle cannot be tracked throughout the tunnel.

[0004] The present invention provides a method for splicing vehicle trajectories of multiple millimeter-wave radars in a tunnel, which is applied to a multiple millimeter-wave radar system in a tunnel, wherein the multiple millimeter-wave radar system in a tunnel comprises multiple millimeter-wave radars sequentially installed on the roadside in the tunnel; the method comprises the following steps:

[0005] S1: Performing time-space calibration on all millimeter-wave radars in the tunnel; the time-space calibration specifically refers to synchronizing the time of all millimeter-wave radars, and the coordinates of the output vehicle information are coordinates in the same coordinate system;

[0006] S2: Obtain the set of vehicle information of all vehicles in the overlapping area of ​​all adjacent millimeter-wave radars at any same time. For any pair of adjacent millimeter-wave radars, they are respectively recorded as radar A and radar B, expressed as:

[0007] ;

[0008] ;

[0009] in, It represents the set of vehicle information of all vehicles in the overlapping area detected by radar A at any same time;

[0010] Represents the set of vehicle information of all vehicles in the overlapping area detected by radar B at any same time;

[0011] Represents the vehicle information of the i-th vehicle in the overlapping area detected by radar A at any same time;

[0012] Represents the vehicle information of the jth vehicle in the overlapping area detected by radar B at any same time;

[0013] S3: Calculation and Similarity features ,in, , ;

[0014] Will A pre-trained BP network is input, where the BP network is a binary classification network. If the result shows that the i-th vehicle of radar A and the j-th vehicle of radar B are the same vehicle, the trajectories of the i-th vehicle of radar A and the j-th vehicle of radar B are spliced, otherwise they are not spliced.

[0015] Preferably, the specific steps of time-space calibration are:

[0016] Use the satellite timing system to synchronize the time of all millimeter-wave radars in the tunnel;

[0017] A millimeter-wave radar is selected as a reference radar, and the original coordinates output by any other millimeter-wave radar are converted so that the coordinates of the vehicle information output by all millimeter-wave radars are coordinates in the same coordinate system; any other millimeter-wave radar is denoted as radar A, and the conversion method is:

[0018] ;

[0019] in,

[0020] (x_1, y_1) is the original coordinate output by radar A;

[0021] ( , ) is the global coordinate after transformation by radar A;

[0022] R is the rotation matrix; T is the translation matrix;

[0023] ;

[0024] ;

[0025] in, ;

[0026] (lat0, lng0, yaw0) is the radar latitude, longitude and heading angle of the reference radar;

[0027] (latA, lngA, yawA) is the radar latitude, longitude and heading angle of radar A.

[0028] Preferably, the vehicle information is represented as:

[0029]

[0030]

[0031] in, express The horizontal coordinate of the coordinates in the vehicle information;

[0032] express The ordinate of the coordinate in the vehicle information;

[0033] express The lateral speed in the vehicle information;

[0034] express The longitudinal speed in the vehicle information;

[0035] express The vehicle type in the vehicle information;

[0036] express The lane in the vehicle information;

[0037] express The horizontal coordinate of the coordinates in the vehicle information;

[0038] express The ordinate of the coordinate in the vehicle information;

[0039] express The lateral speed in the vehicle information;

[0040] express The longitudinal speed in the vehicle information;

[0041] express The vehicle type in the vehicle information;

[0042] express The lane in the vehicle information.

[0043] Preferably, The calculation method is:

[0044]

[0045] in, express and Similarity features of

[0046] express and The location similarity of

[0047] express and The speed similarity of

[0048] express and Attribute similarity of

[0049] in,

[0050] ;

[0051] ;

[0052] .

[0053] Preferably, the BP network has a total of 4 layers, including an input layer, two hidden layers and an output layer in sequence; the input layer has 3 nodes, which are used to input position similarity, speed similarity and attribute similarity respectively, and the activation function adopts ReLU; the hidden layer uses ReLU as the activation function; the output layer has 1 node, adopts Sigmoid activation function, and the output value is between 0 and 1, indicating the probability of whether it is the same vehicle.

[0054] The present invention also provides a multi-millimeter wave radar system in a tunnel, characterized in that the vehicle track splicing is performed using the method described in any one of claims 1 to 5.

[0055] Beneficial effects:

[0056] 1. Strong adaptability. The present invention uses millimeter wave radar to splice vehicle tracks, which is not affected by harsh environmental conditions such as smoke, fog and brightness changes in the tunnel, and can work stably under various environmental conditions.

[0057] 2. Easy deployment. The present invention converts multiple millimeter-wave radars into time and space, and can splice vehicle tracks without precise longitude and latitude coordinates, eliminating the need for high-precision maps, total stations, and other steps to obtain precise longitude and latitude of millimeter-wave radars in tunnels, greatly facilitating the deployment of millimeter-wave radars in tunnels.

[0058] 3. Improve the track splicing effect. The present invention uses the BP neural network method to splice the vehicle tracks between millimeter wave radars, solving the problem that the millimeter wave radar signal has multipath effect in the tunnel, large detection interference, insufficient monitoring data accuracy, and leads to unsatisfactory track splicing effect in the tunnel.

[0059] 4. The system is upgradeable. As time goes by and the sample size grows, the vehicle splicing effect can be further improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a system framework diagram of the present invention; DETAILED DESCRIPTION

[0061] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] like Figure 1 As shown, a method for splicing vehicle trajectories of multiple millimeter-wave radars in a tunnel is applied to a multiple millimeter-wave radar system in a tunnel, wherein the multiple millimeter-wave radar system in a tunnel comprises multiple millimeter-wave radars sequentially installed on the roadside in the tunnel; the method comprises the following steps:

[0063] S1: Performing time-space calibration on all millimeter-wave radars in the tunnel; the time-space calibration specifically refers to synchronizing the time of all millimeter-wave radars, and the coordinates of the output vehicle information are coordinates in the same coordinate system;

[0064] In this implementation, the specific steps of time-space calibration are:

[0065] Use the satellite timing system to synchronize the time of all millimeter-wave radars in the tunnel;

[0066] A millimeter-wave radar is selected as a reference radar, and the original coordinates output by any other millimeter-wave radar are converted so that the coordinates of the vehicle information output by all millimeter-wave radars are coordinates in the same coordinate system; any other millimeter-wave radar is denoted as radar A, and the conversion method is:

[0067] ;

[0068] in,

[0069] (x_1, y_1) is the original coordinate of the output of radar A;

[0070] ( , ) is the global coordinate after transformation by radar A;

[0071] R is the rotation matrix; T is the translation matrix;

[0072] ;

[0073] ;

[0074] in, ;

[0075] (lat0, lng0, yaw0) is the radar latitude, longitude and heading angle of the reference radar;

[0076] (latA, lngA, yawA) is the radar latitude and longitude heading angle of radar A;

[0077] In a tunnel environment, the millimeter-wave radar cannot obtain accurate longitude and latitude coordinates, and the vehicle trajectory will jump when crossing the millimeter-wave radar device. In order to achieve smooth trajectory splicing, the present invention needs to convert the millimeter-wave radar coordinate system in the tunnel to a unified global coordinate system. According to the rotation matrix and translation matrix of each radar, the coordinates in the vehicle position information output by each radar are converted to a global coordinate system with the reference radar as the origin of the coordinate system.

[0078] S2: Obtain the set of vehicle information of all vehicles in the overlapping area of ​​all adjacent millimeter-wave radars at any same time. For any pair of adjacent millimeter-wave radars, they are respectively recorded as radar A and radar B, expressed as:

[0079] ;

[0080] ;

[0081] in, Represents the set of vehicle information of all vehicles in the overlapping area detected by radar A at any same time;

[0082] Represents the set of vehicle information of all vehicles in the overlapping area detected by radar B at any same time;

[0083] Represents the vehicle information of the i-th vehicle in the overlapping area detected by radar A at any same time;

[0084] represents the vehicle information of the jth vehicle in the overlapping area detected by radar B at any same time;

[0085] Among them, the vehicle information is expressed as:

[0086]

[0087]

[0088] in, express The horizontal coordinate of the coordinates in the vehicle information;

[0089] express The ordinate of the coordinate in the vehicle information;

[0090] express The lateral speed in the vehicle information;

[0091] express The longitudinal speed in the vehicle information;

[0092] express The vehicle type in the vehicle information;

[0093] express The lane in the vehicle information;

[0094] express The horizontal coordinate of the coordinates in the vehicle information;

[0095] express The ordinate of the coordinate in the vehicle information;

[0096] express The lateral speed in the vehicle information;

[0097] express The longitudinal speed in the vehicle information;

[0098] express The vehicle type in the vehicle information;

[0099] express The lane in the vehicle information;

[0100] S3: Calculation and Similarity features ,in, , ;

[0101] Will Input the pre-trained BP network, which is a binary classification network. If the result shows that the i-th vehicle of radar A and the j-th vehicle of radar B are the same vehicle, then the trajectories of the i-th vehicle of radar A and the j-th vehicle of radar B are spliced, otherwise they are not spliced;

[0102] In this embodiment The calculation method is:

[0103]

[0104] in, express and Similarity features of

[0105] express and The location similarity of

[0106] express and The speed similarity of

[0107] express and Attribute similarity of

[0108] in,

[0109] ;

[0110] ;

[0111] ;

[0112] The BP network has a total of 4 layers, including an input layer, two hidden layers and an output layer in sequence; the input layer has 3 nodes, which are used to input position similarity, speed similarity and attribute similarity respectively, and the activation function adopts ReLU; the hidden layer uses ReLU as the activation function; the output layer has 1 node, using Sigmoid activation function, and the output value is between 0 and 1, indicating the probability of whether it is the same vehicle.

[0113] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for splicing vehicle trajectories of multiple millimeter-wave radars in a tunnel, applied to a multiple millimeter-wave radar system in a tunnel, wherein the multiple millimeter-wave radar system in a tunnel comprises multiple millimeter-wave radars sequentially installed on the roadside in the tunnel; characterized in that: The method comprises the following steps: S1: Performing time-space calibration on all millimeter-wave radars in the tunnel; the time-space calibration specifically refers to synchronizing the time of all millimeter-wave radars, and the coordinates of the output vehicle information are coordinates in the same coordinate system; S2: Obtain the set of vehicle information of all vehicles in the overlapping area of ​​all adjacent millimeter-wave radars at any same time. For any pair of adjacent millimeter-wave radars, they are respectively recorded as radar A and radar B, expressed as: ； ； in, It represents the set of vehicle information of all vehicles in the overlapping area detected by radar A at any same time; Represents the set of vehicle information of all vehicles in the overlapping area detected by radar B at any same time; Represents the vehicle information of the i-th vehicle in the overlapping area detected by radar A at any same time; represents the vehicle information of the jth vehicle in the overlapping area detected by radar B at any same time; S3: Calculation and Similarity features ,in, , ; Will A pre-trained BP network is input, where the BP network is a binary classification network. If the result shows that the i-th vehicle of radar A and the j-th vehicle of radar B are the same vehicle, the trajectories of the i-th vehicle of radar A and the j-th vehicle of radar B are spliced, otherwise they are not spliced.

2. The method for splicing multi-millimeter-wave radar vehicle trajectories in a tunnel according to claim 1, characterized in that: The specific steps of space-time calibration are: Use the satellite timing system to synchronize the time of all millimeter-wave radars in the tunnel; A millimeter-wave radar is selected as a reference radar, and the original coordinates output by any other millimeter-wave radar are converted so that the coordinates of the vehicle information output by all millimeter-wave radars are coordinates in the same coordinate system; any other millimeter-wave radar is denoted as radar A, and the conversion method is: ； in, (x_1, y_1) is the original coordinate output by radar A; ( , ) is the global coordinate after transformation by radar A; R is the rotation matrix; T is the translation matrix; ； ； in, ; (lat0, lng0, yaw0) is the radar latitude, longitude and heading angle of the reference radar; (latA, lngA, yawA) is the radar latitude, longitude and heading angle of radar A.

3. The method for splicing multi-millimeter-wave radar vehicle trajectories in a tunnel according to claim 1, characterized in that: Vehicle information is represented as: in, express The horizontal coordinate of the coordinates in the vehicle information; express The ordinate of the coordinate in the vehicle information; express The lateral speed in the vehicle information; express The longitudinal speed in the vehicle information; express The vehicle type in the vehicle information; express The lane in the vehicle information; express The horizontal coordinate of the coordinates in the vehicle information; express The ordinate of the coordinate in the vehicle information; express The lateral speed in the vehicle information; express The longitudinal speed in the vehicle information; express The vehicle type in the vehicle information; express The lane in the vehicle information.

4. The method for splicing multi-millimeter-wave radar vehicle trajectories in a tunnel according to claim 3, characterized in that: The calculation method is: in, express and Similarity features of express and The location similarity of express and The speed similarity of express and Attribute similarity of in, ； ； 。 5. The method for splicing multi-millimeter-wave radar vehicle trajectories in a tunnel according to claim 4, characterized in that: The BP network has a total of 4 layers, including an input layer, two hidden layers and an output layer in sequence; the input layer has 3 nodes, which are used to input position similarity, speed similarity and attribute similarity respectively, and the activation function adopts ReLU; the hidden layer uses ReLU as the activation function; the output layer has 1 node, using Sigmoid activation function, and the output value is between 0 and 1, indicating the probability of whether it is the same vehicle.

6. A multi-millimeter wave radar system in a tunnel, characterized in that: Vehicle track splicing is performed using the method described in any one of claims 1 to 5.

Images