Key area vehicle positioning monitoring and early warning method
Through the combination of pulse transmission base station and Beidou positioning, combined with improved long-term and short-term memory neural network to predict vehicle trajectory, the problems of low positioning accuracy and insufficient future behavior prediction capabilities in traditional technology are solved, and efficient monitoring of vehicles in key areas and Early warning.
Patent Information
- Application Number
- CN202411235020.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-04
- Publication Date
- 2024-11-26
AI Technical Summary
Traditional vehicle positioning technology has low positioning accuracy in complex environments, insufficient real-time performance, and lacks the ability to predict the future behavior of the target vehicle, making it difficult to effectively monitor and early warning of vehicle behavior in key areas.
The method of combining pulse transmission base station and Beidou positioning is adopted to determine the vehicle position through the echo reflection principle, and the improved long-term and short-term memory neural network is used to predict the vehicle trajectory, and combined with the space-time attention mechanism, the preset trajectory is matched for early warning.
It improves the positioning accuracy and real-timeness of vehicles in key areas, can quickly identify and warn vehicles that do not meet the expected behavior, improves safety prevention and emergency response capabilities, and reduces safety risks.
Smart Images

Figure CN119024388A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a vehicle positioning monitoring and early warning method in a key area. Background Art
[0002] Key areas generally refer to specific areas that are important for safety, management or other specific purposes, including but not limited to safety-critical areas, traffic control areas, environmental protection areas and isolation areas.
[0003] Traditional systems may rely on relatively simple positioning technologies, such as GPS positioning. These technologies may be affected by multipath effects in complex environments (such as high-rise buildings in cities), resulting in reduced positioning accuracy. The real-time performance of traditional systems may be limited by data update frequency and transmission speed, resulting in slow monitoring response. In addition, traditional systems generally lack the ability to predict the future behavior of target vehicles. Summary of the invention
[0004] The purpose of the present invention is to solve the problems existing in the background technology and to propose a method for positioning, monitoring and warning of vehicles in key areas.
[0005] The technical solution of the present invention is a method for positioning, monitoring and early warning of vehicles in key areas, comprising:
[0006] Setting a pulse transmitting base station and setting a pulse receiving terminal on the target vehicle, and determining the geographical location of the target vehicle relative to the pulse transmitting base station based on the echo reflection principle between the pulse transmitting base station and the pulse receiving terminal;
[0007] Dividing key areas and obtaining a sequence of geographical locations of the boundaries of the key areas relative to the pulse transmitting base station;
[0008] Determine whether the geographical position of the target vehicle relative to the pulse transmitting base station is within the geographical position sequence of the boundary of the key area relative to the pulse transmitting base station. If not, it indicates that the target vehicle is outside the key area.
[0009] If yes, the target vehicle in the key area is positioned in real time by using the Beidou positioning principle to obtain the real-time position of the target vehicle;
[0010] Arranging the real-time position of the target vehicle in a time series to obtain trajectory data of the target vehicle, transmitting the trajectory data of the target vehicle to a trained trajectory prediction model, and outputting the trajectory data of the target vehicle at the next preset time through the trained trajectory prediction model;
[0011] The trajectory data of the target vehicle at the next preset time is matched with the preset standard trajectory data to obtain the matching degree between the two. If the matching degree between the two is lower than the preset matching degree threshold, an early warning is issued through the alarm device.
[0012] Preferably, determining the geographical location of the target vehicle relative to the pulse transmitting base station based on the principle of echo reflection by the pulse transmitting base station and the pulse receiving end includes:
[0013] Calculating the straight-line distance between the pulse transmitting base station and the pulse receiving end;
[0014] The geographical position of the target vehicle relative to the pulse transmitting base station is calculated based on the straight-line distance between the pulse transmitting base station and the pulse receiving end.
[0015] Preferably, the straight-line distance calculation formula is as follows:
[0016]
[0017] Among them, L total represents the straight-line distance between the pulse transmitting base station and the pulse receiving end, ΔT represents the time delay difference between the transmitted pulse and the received pulse, V o Indicates the propagation speed of the pulse signal in the air, f o Indicates the vibration amplitude of the transmitting end in the pulse transmitting base station.
[0018] Preferably, the calculation formula for the geographical location of the target vehicle relative to the pulse transmitting base station is as follows:
[0019]
[0020] Among them, (x d ,y d ) represents the geographical location of the target vehicle relative to the pulse transmitting base station, that is, the horizontal axis value and the vertical axis value, and α represents the tilt angle of the transmitting end in the pulse transmitting base station.
[0021] Preferably, real-time positioning of the target vehicle in the key area is performed by using the Beidou positioning principle to obtain the real-time position of the target vehicle, including:
[0022] Calculating the distance between the Beidou satellite and the target vehicle;
[0023] The real-time position of the target vehicle is calculated based on the distance between the Beidou satellite and the target vehicle.
[0024] Preferably, the distance calculation formula between the Beidou satellite and the target vehicle is as follows:
[0025]
[0026] Where S represents the distance between the Beidou satellite and the target vehicle. represents the vertical distance from the Beidou satellite to the ground plane, ε o represents pseudorange error, β represents auxiliary calculation parameters, ΔT b represents the time error, and ΔT b =ΔT r -ΔT e , ΔT r Represents the receiver clock difference, ΔT e Represents the satellite clock error.
[0027] Preferably, the real-time position calculation formula of the target vehicle is as follows:
[0028]
[0029] Among them, S1 and S2 represent the real-time position of the target vehicle, (x b ,y b ) represents the real-time position of the target vehicle, (x1, y1) represents the coordinate information of the first BeiDou satellite, (x2, y2) represents the coordinate information of the second BeiDou satellite, and ε 1 and ε 2 represents the pseudorange error between the first and second BeiDou satellites, ΔT 1 b and ΔT 2 b Indicates the time error between the first and second BeiDou satellites.
[0030] Preferably, the trajectory data expression of the target vehicle is as follows:
[0031]
[0032] Among them, (X, Y) represents the trajectory data of the target vehicle, Represents the real-time position of the target vehicle at time t.
[0033] Preferably, the trajectory prediction model adopts an improved long short-term memory neural network, which is improved by introducing a spatiotemporal attention mechanism, wherein the spatiotemporal attention mechanism includes a temporal attention layer and a spatial attention layer. The trajectory prediction model takes the trajectory data of the target vehicle as input, and obtains a hidden state through the long short-term memory neural network. Get the time attention weight of the target vehicle The hidden state and the temporal attention weight Combined to get the tensor unit value associated with the target vehicle Based on the associated tensor unit value of the target vehicle Calculating spatial attention weights The temporal attention weight and the spatial attention weights Combine to get the spatiotemporal attention weight V t .
[0034] Preferably, the hidden state The calculation formula is as follows:
[0035]
[0036] in, represents the hidden state of the jth layer of the long short-term memory neural network;
[0037] The temporal attention weight The calculation formula is as follows:
[0038]
[0039] Among them, softmax represents the softmax activation function, tanh represents the hyperbolic tangent activation function, W α represents the learnable weights in the temporal attention layer;
[0040] The target vehicle's associated tensor unit value The calculation formula is as follows:
[0041]
[0042] Where T represents the transpose of the matrix;
[0043] The spatial attention weights The calculation formula is as follows:
[0044]
[0045] Among them, W β represents the learnable weights in the spatial attention layer;
[0046] The spatiotemporal attention weight V t The calculation formula is as follows:
[0047]
[0048] Compared with the prior art, the above technical solution of the present invention has the following beneficial technical effects:
[0049] 1. The present invention uses a combination of a pulse transmitting base station and a receiving end to determine the geographical location of the target vehicle relative to the base station using the echo reflection principle, and can accurately determine whether the vehicle is in or outside the set key area. For target vehicles entering the key area, the Beidou positioning principle is used for real-time positioning to obtain the real-time position of the target vehicle, and the historical trajectory data is transmitted to the trained trajectory prediction model to predict the trajectory of the target vehicle at future times.
[0050] 2. The present invention analyzes the matching degree of the predicted trajectory of the target vehicle with the preset standard trajectory. If the matching degree is lower than the preset threshold, an early warning can be triggered, thereby effectively improving the ability to identify and respond to potential threats or abnormal behaviors, and can quickly identify and warn vehicles that do not conform to expected behaviors in key areas, which helps to improve safety precautions and emergency response capabilities, and reduce potential safety risks and the possibility of incidents. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] Figure 1 A schematic diagram of the overall method in an embodiment of the present invention ;
[0052] Figure 2 The relationship between the pulse transmitting base station, the vehicle and the Beidou satellite in one embodiment of the present invention is shown in FIG. intention . DETAILED DESCRIPTION
[0053] Embodiment 1, as Figure 1 As shown, a key area vehicle positioning monitoring and early warning method proposed by the present invention comprises:
[0054] S1. Set up a pulse transmitting base station and a pulse receiving terminal on the target vehicle, and determine the geographical location of the target vehicle relative to the pulse transmitting base station based on the principle of echo reflection between the pulse transmitting base station and the pulse receiving terminal;
[0055] S2. Divide the key areas and obtain a sequence of geographical locations of the boundaries of the key areas relative to the pulse transmitting base stations;
[0056] S3, determining whether the geographical location of the target vehicle relative to the pulse transmitting base station is within the geographical location sequence of the boundary of the key area relative to the pulse transmitting base station. If not, it indicates that the target vehicle is outside the key area;
[0057] S4. If yes, the target vehicle in the key area is positioned in real time by using the Beidou positioning principle to obtain the real-time position of the target vehicle;
[0058] S5, arranging the real-time position of the target vehicle in time series to obtain the trajectory data of the target vehicle, transmitting the trajectory data of the target vehicle to the trained trajectory prediction model, and outputting the trajectory data of the target vehicle at the next preset time through the trained trajectory prediction model;
[0059] S6. Match the trajectory data of the target vehicle at the next preset time with the preset standard trajectory data to obtain a matching degree between the two. If the matching degree between the two is lower than a preset matching degree threshold, an alarm device is used to issue an early warning.
[0060] In the present invention, the pulse transmitting base station and the pulse receiving end: this refers to the pulse radar in radar technology, which transmits an electromagnetic pulse and receives its echo. The time delay and intensity of the echo can be used to calculate the distance and direction of the target object (such as a vehicle) relative to the radar; the Beidou positioning principle can accurately determine the position of the target vehicle in the geographic space; key areas: refers to specific areas divided in the geographic space, which are security-critical areas and restricted areas; alarm equipment includes but is not limited to sound alarms, light flash alarms, vibration alarms, electronic alarm systems and automatic telephone call systems.
[0061] Embodiment 2, a method for positioning, monitoring and warning of vehicles in key areas proposed by the present invention, compared with embodiment 1, this embodiment further includes: determining the geographical location of the target vehicle relative to the pulse transmitting base station based on the principle of echo reflection by the pulse transmitting base station and the pulse receiving end, including:
[0062] A1. Calculate the straight-line distance between the pulse transmitting base station and the pulse receiving end;
[0063] A2. Calculate the geographic location of the target vehicle relative to the pulse transmitting base station based on the straight-line distance between the pulse transmitting base station and the pulse receiving end.
[0064] In this embodiment, the pulse transmitting base station is a device for transmitting radar signals, which generates and transmits electromagnetic pulses; the pulse receiving end is used to receive the radar signal reflected by the target object; the echo reflection principle means that when the radar signal (pulse) collides with the target object, part of the signal will be reflected back by the target object to form an echo. By analyzing the time delay and intensity of these echoes, the distance, speed and direction of the target object relative to the radar can be determined.
[0065] In an optional embodiment, the straight-line distance calculation formula is as follows:
[0066]
[0067] Among them, L total represents the straight-line distance between the pulse transmitting base station and the pulse receiving end, ΔT represents the time delay difference between the transmitted pulse and the received pulse, Vo Indicates the propagation speed of the pulse signal in the air, f o Indicates the vibration amplitude of the transmitting end in the pulse transmitting base station.
[0068] In an optional embodiment, the geographical location of the target vehicle relative to the pulse transmitting base station is calculated as follows:
[0069]
[0070] Among them, (x d ,y d ) represents the geographical location of the target vehicle relative to the pulse transmitting base station, that is, the horizontal axis value and the vertical axis value, and α represents the tilt angle of the transmitting end in the pulse transmitting base station.
[0071] In an optional embodiment, real-time positioning of a target vehicle in a key area is performed using the Beidou positioning principle to obtain the real-time position of the target vehicle, including:
[0072] B1. Calculate the distance between the Beidou satellite and the target vehicle;
[0073] B2. Calculate the real-time position of the target vehicle based on the distance between the Beidou satellite and the target vehicle.
[0074] In an optional embodiment, the distance calculation formula between the Beidou satellite and the target vehicle is as follows:
[0075]
[0076] Where S represents the distance between the Beidou satellite and the target vehicle. represents the vertical distance from the Beidou satellite to the ground plane, ε o represents pseudorange error, β represents auxiliary calculation parameters, ΔT b represents the time error, and ΔT b =ΔT r -ΔT e , ΔT r Represents the receiver clock difference, ΔT e Represents the satellite clock error.
[0077] In an optional embodiment, the real-time position calculation formula of the target vehicle is as follows:
[0078]
[0079] Among them, S1 and S2 represent the real-time position of the target vehicle, (x b ,y b) represents the real-time position of the target vehicle, (x1, y1) represents the coordinate information of the first BeiDou satellite, (x2, y2) represents the coordinate information of the second BeiDou satellite, and ε 1 and ε 2 represents the pseudorange error between the first and second BeiDou satellites, ΔT 1 b and ΔT 2 b Indicates the time error between the first and second BeiDou satellites.
[0080] In an optional embodiment, the trajectory data of the target vehicle is expressed as follows:
[0081]
[0082] Among them, (X, Y) represents the trajectory data of the target vehicle, Represents the real-time position of the target vehicle at time t.
[0083] In an optional embodiment, the trajectory prediction model uses an improved long short-term memory neural network, which is improved by introducing a spatiotemporal attention mechanism. The spatiotemporal attention mechanism includes a temporal attention layer and a spatial attention layer. The trajectory prediction model takes the trajectory data of the target vehicle as input and obtains the hidden state through the long short-term memory neural network. Get the temporal attention weight of the target vehicle Hide the state and temporal attention weights Combined to get the tensor unit value associated with the target vehicle Based on the associated tensor unit value of the target vehicle Calculating spatial attention weights The temporal attention weight and spatial attention weights Combine to get the spatiotemporal attention weight V t .
[0084] It should be noted that the long short-term memory network (LSTM) is a recurrent neural network (RNN) specially designed to process and predict time series data, such as trajectory data. It can capture long-term dependencies, avoid the gradient vanishing problem in traditional RNNs, and control the input and output of information through gating units; the temporal attention layer is used to dynamically learn the importance weight of each time step in the time series, which allows the model to focus on the most important time points or time periods in the prediction process instead of treating all time steps uniformly; the spatial attention layer is used to learn the importance weights of different features or inputs in space. For trajectory prediction models, it refers to the contribution of different locations or regions in predicting the target trajectory; the hidden state in the LSTM network is a memory unit used to store and transmit the information learned by the model in the input sequence for use in prediction; the tensor unit refers to the data structure in the neural network, similar to a multidimensional array, which is used to store and transmit data and information in the neural network.
[0085] In an alternative embodiment, the hidden state The calculation formula is as follows:
[0086]
[0087] in, represents the hidden state of the jth layer of the long short-term memory neural network;
[0088] Temporal attention weight The calculation formula is as follows:
[0089]
[0090] Among them, softmax represents the softmax activation function, tanh represents the hyperbolic tangent activation function, W α represents the learnable weights in the temporal attention layer;
[0091] The associated tensor cell value of the target vehicle The calculation formula is as follows:
[0092]
[0093] Where T represents the transpose of the matrix;
[0094] Spatial attention weights The calculation formula is as follows:
[0095]
[0096] Among them, W β represents the learnable weights in the spatial attention layer;
[0097] Spatial-temporal attention weight Vt The calculation formula is as follows:
[0098]
[0099] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto, and various changes can be made within the knowledge scope of technicians in the relevant technical field without departing from the purpose of the present invention.
Claims
1. A method for positioning, monitoring and early warning of vehicles in key areas, It is characterized in that include: Setting a pulse transmitting base station and setting a pulse receiving terminal on the target vehicle, and determining the geographical location of the target vehicle relative to the pulse transmitting base station based on the echo reflection principle between the pulse transmitting base station and the pulse receiving terminal; Dividing key areas and obtaining a sequence of geographical locations of the boundaries of the key areas relative to the pulse transmitting base station; Determine whether the geographical position of the target vehicle relative to the pulse transmitting base station is within the geographical position sequence of the boundary of the key area relative to the pulse transmitting base station. If not, it indicates that the target vehicle is outside the key area. If yes, the target vehicle in the key area is positioned in real time by using the Beidou positioning principle to obtain the real-time position of the target vehicle; Arranging the real-time position of the target vehicle in a time series to obtain trajectory data of the target vehicle, transmitting the trajectory data of the target vehicle to a trained trajectory prediction model, and outputting the trajectory data of the target vehicle at the next preset time through the trained trajectory prediction model; The trajectory data of the target vehicle at the next preset time is matched with the preset standard trajectory data to obtain the matching degree between the two. If the matching degree between the two is lower than the preset matching degree threshold, an early warning is issued through the alarm device.
2. A method for monitoring and warning vehicle positioning in key areas according to claim 1, It is characterized in that Determining the geographical location of the target vehicle relative to the pulse transmitting base station based on the pulse transmitting base station and the pulse receiving end through the echo reflection principle includes: Calculating the straight-line distance between the pulse transmitting base station and the pulse receiving end; The geographical position of the target vehicle relative to the pulse transmitting base station is calculated based on the straight-line distance between the pulse transmitting base station and the pulse receiving end.
3. A method for monitoring and warning vehicle positioning in key areas according to claim 2, It is characterized in that The straight-line distance calculation formula is as follows: Among them, L total represents the straight-line distance between the pulse transmitting base station and the pulse receiving end, ΔT represents the time delay difference between the transmitted pulse and the received pulse, V o Indicates the propagation speed of the pulse signal in the air, f o Indicates the vibration amplitude of the transmitting end in the pulse transmitting base station.
4. A method for monitoring and warning vehicle positioning in key areas according to claim 3, It is characterized in that The calculation formula for the geographical location of the target vehicle relative to the pulse transmitting base station is as follows: Among them, (x d ,y d ) represents the geographical location of the target vehicle relative to the pulse transmitting base station, that is, the horizontal axis value and the vertical axis value, and α represents the tilt angle of the transmitting end in the pulse transmitting base station.
5. A method for monitoring and warning vehicle positioning in key areas according to claim 1, It is characterized in that The target vehicle in the key area is positioned in real time by using the Beidou positioning principle to obtain the real-time position of the target vehicle, including: Calculating the distance between the Beidou satellite and the target vehicle; The real-time position of the target vehicle is calculated based on the distance between the Beidou satellite and the target vehicle.
6. A method for monitoring and warning vehicle positioning in key areas according to claim 5, It is characterized in that The distance calculation formula between the Beidou satellite and the target vehicle is as follows: Where S represents the distance between the Beidou satellite and the target vehicle. represents the vertical distance from the Beidou satellite to the ground plane, ε o represents pseudorange error, β represents auxiliary calculation parameters, ΔT b represents the time error, and ΔT b =ΔT r -ΔT e , ΔT r Represents the receiver clock difference, ΔT e Represents the satellite clock error.
7. A method for monitoring and warning vehicle positioning in key areas according to claim 6, It is characterized in that The real-time position calculation formula of the target vehicle is as follows: Among them, S 1 and S 2 represents the real-time position of the target vehicle, (x b ,y b ) represents the real-time position of the target vehicle, (x 1 ,y 1 ) represents the coordinate information of the first Beidou satellite, (x 2 ,y 2 ) represents the coordinate information of the second BeiDou satellite, indicating that ε 1 and ε 2 represents the pseudorange error between the first and second BeiDou satellites, ΔT 1 b and ΔT 2 b Indicates the time error between the first and second BeiDou satellites.
8. A method for monitoring and warning vehicle positioning in key areas according to claim 1, It is characterized in that The trajectory data expression of the target vehicle is as follows: Among them, (X, Y) represents the trajectory data of the target vehicle, Represents the real-time position of the target vehicle at time t.
9. A method for monitoring and warning vehicle positioning in key areas according to claim 1, It is characterized in that The trajectory prediction model adopts an improved long short-term memory neural network, which is improved by introducing a spatiotemporal attention mechanism. The spatiotemporal attention mechanism includes a temporal attention layer and a spatial attention layer. The trajectory prediction model takes the trajectory data of the target vehicle as input and obtains a hidden state through the long short-term memory neural network. Get the time attention weight of the target vehicle The hidden state and the temporal attention weight Combined to get the tensor unit value associated with the target vehicle Based on the associated tensor unit value of the target vehicle Calculating spatial attention weights The temporal attention weight and the spatial attention weight B t v Combine to get the spatiotemporal attention weight V t .
10. A method for monitoring and warning vehicle positioning in key areas according to claim 9, It is characterized in that The hidden state The calculation formula is as follows: in, represents the hidden state of the jth layer of the long short-term memory neural network; The temporal attention weight The calculation formula is as follows: Among them, softmax represents the softmax activation function, tanh represents the hyperbolic tangent activation function, W α represents the learnable weights in the temporal attention layer; The target vehicle's associated tensor unit value The calculation formula is as follows: Where T represents the transpose of the matrix; The spatial attention weights The calculation formula is as follows: Among them, W β represents the learnable weights in the spatial attention layer; The spatiotemporal attention weight V t The calculation formula is as follows:
