Vehicle behavior identification method, device and equipment

Abstract

The invention relates to the technical field of traffic safety monitoring, and particularly discloses a vehicle behavior recognition method, device and equipment. According to the invention, the distance data of the vehicle can be reliably acquired in real time by distributing the communication time slot and using the fixed base station to measure the distance, and the data processing flow from the distance data to the behavior identification is simplified by adopting the behavior of the vehicle based on deviation identification. The method can adapt to different traffic road sections and environments, and has good adaptability and flexibility.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic safety monitoring, and particularly relates to a method, device and equipment for vehicle behavior recognition. Background Art

[0002] With the increasingly complex traffic conditions accompanied by the increasing number of current vehicles, traditional vehicle behavior recognition methods have inaccurate recognition. Especially in high-traffic sections, how to communicate with vehicles in real time and reliably and obtain accurate distance data is the key to ensuring the effectiveness of the vehicle behavior recognition system. Summary of the Invention

[0003] In order to overcome the above problems existing in the existing vehicle behavior recognition, the present invention provides a method, device and equipment for vehicle behavior recognition.

[0004] In order to achieve the above invention purpose, the present invention provides the following technical solutions:

[0005] In a first aspect, the present invention provides a method for vehicle behavior recognition, the method comprising:

[0006] Construct a lane coordinate system of the target section;

[0007] Based on the target vehicle entering the target section, allocate communication time slots to the target vehicle according to the traffic flow of the target section, so that the target vehicle performs ranging with a fixed base station and returns distance data;

[0008] Calculate the coordinates of the target vehicle according to the distance data, and perform trajectory prediction to obtain the trajectory of the target vehicle;

[0009] Judge the deviation of the trajectory of the target vehicle in the lane coordinate system, and identify the behavior of the target vehicle based on the deviation; the behavior of the target vehicle includes overtaking behavior and lane-changing behavior.

[0010] According to a specific implementation manner, in the above recognition method, at least three fixed base stations are set in the target section, and the layout spacing is set according to the minimum coverage distance of the fixed base station, and they are arranged at intervals on both sides of the target section.

[0011] According to a specific implementation manner, in the above recognition method, the width of the target section is obtained through the position of the fixed base station, and then combined with the lane width of the target section, a lane coordinate system of the target section is constructed.

[0012] According to a specific implementation manner, in the above recognition method, calculating the coordinates of the target vehicle according to the distance data includes:

[0013] Adopt a geometric positioning method to obtain the estimated coordinates of the target vehicle;

[0014] The weighted least squares method and the Taylor-based weighted least squares method are used to eliminate the errors of the estimated coordinates, and the coordinates of the target vehicle are obtained.

[0015] According to a specific implementation manner, in the above recognition method, determining the deviation of the trajectory of the target vehicle in the lane coordinate system includes:

[0016] Based on a preset trajectory, obtain the lateral distance between each trajectory point in the trajectory of the target vehicle and each trajectory point in the preset trajectory;

[0017] Judge the deviation based on the lateral distance.

[0018] According to a specific implementation manner, in the above recognition method, identifying the behavior of the target vehicle based on the deviation includes:

[0019] Based on the fact that within a first time window, there are consecutive lateral distances in the lateral distances that are greater than a preset threshold and reach a preset ratio, identify the behavior of the target vehicle as a lane change behavior;

[0020] Based on the fact that the behavior of the target vehicle is a lane change behavior, and by the lateral distance within a second time window returning to be less than the preset threshold, identify the behavior of the target vehicle as an overtaking behavior.

[0021] In a second aspect, the present invention provides a vehicle behavior recognition device, and the device includes:

[0022] A monitor, configured to allocate communication time slots to the target vehicle according to the traffic flow of the target road section;

[0023] A fixed base station, configured to complete ranging with the target vehicle;

[0024] An in-vehicle tag, configured to receive the communication time slot, perform ranging with the fixed base station, and return distance data to the data processing center;

[0025] A data processing center, configured to receive the distance data, and use a vehicle behavior recognition method as described in any one of the above to obtain the behavior of the target vehicle;

[0026] Wherein, the monitor, the fixed base station and the data processing center are communicatively connected.

[0027] According to a specific implementation manner, in the above recognition device, the device further includes:

[0028] A remote control center, communicatively connected to the data processing center, configured to display the coordinates and trajectory of the target vehicle, and make predictions according to the trajectory of the target vehicle.

[0029] In a third aspect, the present invention further provides an electronic device, which includes a memory and a processor, where

[0030] the memory is used to store a computer program;

[0031] the processor is used to call and execute the computer program so that the device executes a vehicle behavior recognition method as described in any one of the above.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0033] By allocating communication time slots and using a fixed base station for ranging, the present invention can obtain the distance data of vehicles in real time and reliably, and adopts the behavior of vehicles based on deviation recognition, simplifying the data processing flow from distance data to behavior recognition. The present invention can adapt to different traffic sections and environments, and has good adaptability and flexibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a schematic structural diagram of a vehicle behavior recognition device provided by an embodiment of the present invention;

[0035] Figure 2 is a schematic flowchart of a vehicle behavior recognition method provided by an embodiment of the present invention;

[0036] Figure 3 is a schematic diagram of the principle of the geometric positioning method provided by an embodiment of the present invention;

[0037] Figure 4 is a schematic diagram of the principle of solving positioning by the least squares method provided by an embodiment of the present invention;

[0038] Figure 5 is a schematic diagram of an overtaking trajectory provided by an embodiment of the present invention;

[0039] Figure 6 is a schematic flowchart of a trajectory classification algorithm provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The present invention will be further described in detail below in conjunction with test examples and specific embodiments. However, this should not be construed as limiting the scope of the above-mentioned subject matter of the present invention to the following embodiments. All technologies implemented based on the content of the present invention belong to the scope of the present invention.

[0041] First of all, it should be noted that a vehicle behavior recognition method provided by the present invention is based on Ultra-wideband (UWB) technology, which is a carrierless wireless communication technology with characteristics such as fast transmission rate, low power consumption, and strong anti-multipath interference ability. In the 1960s, UWB was initially used for military purposes, and it was not until 2002 that the Federal Communications Commission (FCC) of the United States issued commercialization specifications. The FCC gave two definitions of ultra-wideband signals: the absolute bandwidth is greater than 500 MHz or the relative bandwidth meets the following conditions.

[0042]

[0043] In the formula, f h , f l are the upper cut-off frequency and the lower cut-off frequency when the signal attenuation is 10 dB. To avoid mutual interference with other radio wave communication systems, the FCC stipulates that the standard operating spectrum of ultra-wideband signals is 3.1 GHz to 10.6 GHz (the available spectrum is 7.5 GHz), and the radiation power does not exceed 41.25 dBm / MHz. The technologies applied to target section positioning mainly include infrared positioning technology, ultrasonic positioning technology, Radio Frequency Identification (RFID) positioning technology, Bluetooth positioning technology, Zigbee positioning technology, WiFi positioning technology, etc. However, most of these positioning technologies have problems such as short transmission distance, poor anti-multipath effect, and low positioning accuracy. The UWB positioning technology uses a working method of transmitting information with extremely short narrow pulses, becoming an ideal method for precise ranging, and it has the following characteristics:

[0044] (1) Fast transmission rate. Since the ultra-wideband pulse signal has a very large bandwidth, according to Shannon's second theorem, when the signal-to-noise ratio is constant, the ultra-wideband signal has a very high data transmission rate.

[0045]

[0046] In the formula, C is the maximum speed supported by the channel or the channel capacity, B is the signal transmission bandwidth, S is the average signal power, N is the average noise power, and S / N is the signal-to-noise ratio.

[0047] (2) High ranging accuracy and strong anti-multipath effect interference ability of the signal. Since ultra-wideband communication uses extremely short narrow pulses to propagate messages, its signal has good time-domain resolution properties.

[0048] (3) Low cost and low power efficiency. To avoid interfering with other communication signals, the FCC stipulates that ultra-wideband needs to use an extremely low signal transmission power (-41.25 dBm / MHz), so the ultra-wideband signal has low power consumption.

[0049] To address these issues, a vehicle behavior recognition method based on the UWB positioning system is adopted, which makes full use of the advantages of UWB technology, such as strong anti-interference performance, low power consumption, high transmission rate, long transmission distance, large spatial transmission capacity, high penetration, simple hardware structure, and low system complexity, enabling good coexistence with other systems. It realizes the precise positioning of personnel, vehicles, and equipment on the target section.

[0050] Please refer to Figure 1 , which shows the schematic structural diagram of a vehicle behavior recognition device provided by an embodiment of the present invention. The device mainly includes a monitor, a fixed base station, an on-vehicle tag, a data processing center, and a remote control center.

[0051] The monitor mainly realizes vehicle detection, traffic statistics, and time slot allocation. The fixed base station and the on-vehicle tag are mainly used to complete the ranging function between two points, providing the original data for solving the vehicle position coordinates. The base station covers the target section in a cell-based manner, and the mobile tag is connected to the on-vehicle computer through a USB interface. Ethernet transmission is used for data transmission. Considering the power supply of the target section, the PoE scheme is adopted, including data frames encapsulating messages such as the distance and timestamp between the fixed base station and the reference base station to the local location engine (LLE), and data frames encapsulating the position coordinates of the vehicle to be located from the local location engine to the remote monitoring center (such as the Municipal Transportation Bureau). In addition, the on-vehicle tag communicates with the on-vehicle ECU through Micro USB. The location of the device deployment is not restricted by power supply, with flexible wiring and easy management. The data processing center, that is, the local location engine, parses the data frames from the fixed base station according to the TCP / IP protocol, stores the useful information, eliminates errors through Kalman filtering, then uses the trilateration algorithm to calculate the position coordinates of the tag relative to each reference base station and converts them into WSG84 coordinates, and refreshes the positioning result on the console interface. At the same time, it stores the historical trajectory for trajectory prediction, accident detection, and post-accident investigation and evidence collection.

[0052] Among them, the TOA particle swarm algorithm is used to measure the optical propagation time between a UWB positioning terminal and multiple UWB positioning base stations. At least three positioning base stations are required to accurately locate the position of the terminal using the trilateration method. The layout spacing is set according to the minimum coverage distance of the fixed base station and is arranged at intervals on both sides of the target section.

[0053] Furthermore, the functions of the above device are further introduced below in combination with the vehicle behavior recognition method provided by the present invention.

[0054] Please refer to Figure 2 , which shows the schematic flow diagram of a vehicle behavior recognition method provided by an embodiment of the present invention. The method includes:

[0055] Step 1: Construct the lane coordinate system of the target road section.

[0056] Step 2: Based on the target vehicle entering the target road section, allocate communication time slots to the target vehicle according to the traffic flow of the target road section, so that the target vehicle can range with the fixed base station and return distance data.

[0057] The monitor broadcasts a Beacon message frame to detect whether there is a vehicle entering the target road section. When it is detected that there is a vehicle entering the target road section, the base station is awakened from the sleep state, and the monitor allocates communication time slots to the on-vehicle tags of the vehicles entering the target road section, and at the same time establishes a reference coordinate system. The reference coordinate system is used to determine the position parameters of the vehicle.

[0058] The on-vehicle tag is awakened when the allocated communication time slot arrives, broadcasts a Poll frame, completes ranging with the fixed base station, and the position engine completes the vehicle position coordinate calculation, trajectory prediction, and accident detection based on the data received from the base station and transmits them to the remote monitoring center, and feeds back to the on-vehicle tag in the next communication time slot to inform the vehicle position coordinates and perform correction. The on-vehicle ECU will also perform position coordinate prediction through an improved Kalman model to achieve real-time vehicle positioning and monitoring.

[0059] When the vehicle exits the target road section, the monitor releases the communication time slot for the positioning of later vehicles. At the same time, it is judged whether there are vehicles in the target road section. When there are no vehicles in the target road section, the fixed base station enters the sleep state to save power consumption until the monitor detects that there is a vehicle about to enter the target road section, and then the base station switches from the sleep state to the working state to continue positioning the vehicle.

[0060] Step 3: Calculate the coordinates of the target vehicle based on the distance data, and perform trajectory prediction to obtain the trajectory of the target vehicle.

[0061] In a TOA-based positioning system, the ranging accuracy determines the accuracy of target positioning, but the improvement of ranging accuracy is limited. Therefore, choosing a suitable coordinate position estimation method is also a necessary prerequisite for improving positioning accuracy.

[0062] The embodiment of the present invention uses a geometric positioning method to obtain the estimated coordinates of the target vehicle; the weighted least squares method and the weighted least squares method based on Taylor are used to eliminate the errors of the estimated coordinates to obtain the coordinates of the target vehicle.

[0063] Specifically, please refer to Figure 3, which shows a schematic diagram of the principle of the geometric positioning method provided by the embodiments of the present invention; taking trilateral positioning as an example, the problem of solving the coordinates of the node to be measured is converted into the problem of finding the intersection point of three anchor nodes with known coordinates, forming a circle with itself as the center and the distance from the node to be measured as the radius. Given the coordinates of three base station anchor nodes A(x1, y1), B(x2, y2), C(x3, y3), and the target node to be measured (x, y), the distances are as follows:

[0064]

[0065] From the above formula, the coordinates of the unknown node can be obtained:

[0066]

[0067] In the geometric positioning method, multi-lateral positioning is too ideal and there is NLOS error, resulting in several circles not intersecting at the real point, large error in solving the equations, and inaccurate estimation of the position coordinates. The multi-angle positioning method will also bring deviation to the angle under the multipath effect of NLSO propagation, and the coordinates cannot be accurately located.

[0068] Furthermore, the non-linear equations of formula (1) cannot be accurately solved. The common positioning method to solve the above error problem is the least square algorithm (LS), and its principle is as Figure 4 shown. Assuming that the coordinates of n base station nodes are known in two-dimensional positioning, expressed as (x i , y i ), and the tag node to be measured is (x, y), according to the distance formula, we get:

[0069]

[0070] Subtracting the first n - 1 terms of the above formula from the nth term respectively to obtain the equations:

[0071]

[0072] Its matrix form is denoted as:

[0073] AX = B(4)

[0074] Let where

[0075]

[0076] Due to the ranging error, let its model be:

[0077] AX + N = B(5)

[0078] N is a random error vector, so the solution of X should minimize N = B - AX. The LS algorithm is a common method for solving non-linear equations and optimizing errors. Let the equation be N(X) = ||B - AX|| 2 , when the derivative of N(X) is 0, solve for the optimal solution of X:

[0079]

[0080] Simplify the above equation to obtain the least squares solution:

[0081] X = (A T A) -1 A T B (7)

[0082] Considering that the error between nodes is related to the measured distance, different weights are assigned to each node according to the position accuracy and distance accuracy at different positions, resulting in the Weighted Least Square algorithm (WLS). Therefore, formula (4) is replaced with:

[0083] WAX = WB(8)

[0084] In the formula, W is the weighting coefficient, which is jointly determined by the positioning accuracy of the anchor node itself and the distance accuracy corresponding to each tag node. Assume that the tag node receives information from the i-th anchor node S, and the positioning accuracy of the anchor node S i is δ i , and the distance accuracy from the tag node X is μ i . Therefore, the weighting coefficient of the anchor node S for the tag node X is:

[0085]

[0086] So the weighted least squares solution is:

[0087] X = (A T WA) -1 A T WB (10)

[0088] Eliminating the quadratic term by subtracting the two equations of the system of equations in the above formula will cause loss of coordinate information. Therefore, the Taylor formula expansion method is introduced for linearization to solve the above problem. The Taylor algorithm is a method that uses the initial value to iterate repeatedly and outputs the optimal coordinate result of the target node when the positioning error is lower than the threshold value.

[0089]

[0090] Let the initial position be (x0, y0), and expand the Taylor formula for f(x, y):

[0091]

[0092] The error quantity can be solved according to the weighted least squares solution formula (10):

[0093]

[0094] The set threshold is:[[]]

[0095] |Δx| + |Δy| ≤ ε (14)

[0096] Judge whether formula (14) holds. If it holds, stop the iteration and output the result; otherwise, perform the next iteration. Let x0 = x0 + Δx0 and y0 = y0 + Δy0 be used as the initial values and substitute them into formula (13) to recalculate until formula (14) is satisfied. The finally obtained iteration result coordinates (x0, y0) are the estimated coordinates of the target node.

[0097] The defect of the Taylor algorithm is that the initial iteration value needs to be as close as possible to the true value. If the initial value deviates too much, it will not only increase the number of iterations of the algorithm but also cause the algorithm to have local convergence, and the algorithm cannot perform global optimization. Therefore, the Taylor algorithm and WLS are fused and improved, and the result obtained by WLS is used as the initial value of the Taylor algorithm to ensure the rapid convergence of Taylor iteration.

[0098] In summary, different positioning methods can obtain position coordinate values. Combining with special positioning scenarios, in order to reduce the error of the position coordinates of the target node, the algorithm is continuously optimized, and at the same time, the algorithm complexity will also increase. Theoretically speaking, the direct geometric positioning method uses geometric relationships to solve, which is simple and fast to calculate, but it is too idealized and cannot be directly used in actual applications. The weighted least squares method based on Taylor is more in line with the actual environment. The weighted initial value can accelerate global convergence, reduce the number of recursions, and can also obtain higher accuracy. Currently, the weighted least squares method based on Taylor is more in line with the positioning system. Therefore, the weighted least squares method based on Taylor is selected as the position estimation algorithm for the UWB positioning system.

[0099] Step 4: Judge the deviation of the trajectory of the target vehicle in the lane coordinate system, and identify the behavior of the target vehicle based on the deviation; the behaviors of the target vehicle include overtaking behavior and lane-changing behavior.

[0100] Specifically, judging the deviation of the trajectory of the target vehicle in the lane coordinate system includes:[[]]

[0101] Based on the preset trajectory, obtain the lateral distance between each trajectory point in the trajectory of the target vehicle and each trajectory point in the preset trajectory;

[0102] Judge the deviation based on the lateral distance.

[0103] In this embodiment, by collecting various trajectory data, the type of the trajectory is analyzed. If multiple trajectories belong to the same category, it can be considered that they have similar driving behaviors. At this time, the driving condition can be considered stable and safe. At the same time, through the analysis of the trajectory data, it is also possible to identify whether the vehicle has overtaking or lane-changing behaviors, and whether the two vehicles are driving in separate lanes or the same lane.

[0104] The trajectory data is generated by connecting the vehicle positions obtained by recognition and calculation.

[0105] Please refer to Figure 5 , which shows a schematic diagram of an overtaking trajectory provided by an embodiment of the present invention. First, two trajectories are classified. The main idea of the classification algorithm is that each trajectory corresponds to a set of trajectory points, and the classification of the trajectories is achieved by relying on the lateral distance between each trajectory curve. Taking trajectories A and B as an example, the specific calculation method is as follows: First, take a point a1 on trajectory A, calculate its Euclidean distance to all points on trajectory B and obtain the minimum value d1 among them. This distance is the lateral distance from a1 to trajectory B. And so on, calculate the lateral distances d2, d3,..., d of the remaining trajectory points a2, a3,..., an on trajectory A to trajectory B n . Through the above method, the distances between all trajectories can be obtained, and a distance matrix can be obtained through the lateral distances. Together with the set time window and threshold, they are used as the discriminant conditions for curve classification.

[0106] Furthermore, based on the deviation, the behavior of the target vehicle is identified, including:

[0107] Based on the fact that within the first time window, there are consecutive lateral distances greater than a preset threshold reaching a preset ratio in the lateral distances, it is identified that the behavior of the target vehicle is a lane-changing behavior.

[0108] Furthermore, based on the fact that the behavior of the target vehicle is a lane-changing behavior, by the lateral distance within the second time window returning to be less than the preset threshold, it is identified that the behavior of the target vehicle is an overtaking behavior.

[0109] It should be noted that if the lateral distance is greater than the preset threshold, it is classified into a separate category; if it is less than the preset threshold, the two curves are classified into one category. For the identification of the lane-changing behavior, on this basis, if the relative distance in the trajectory is greater than this threshold for a certain proportion of points, then there is a lane-changing behavior in this trajectory. For the identification of the overtaking behavior, it is to identify that the lateral distance returns to the preset threshold on the basis that the target vehicle has a lane-changing behavior. Please refer to Figure 6 , which shows a schematic diagram of the trajectory classification algorithm flow provided by an embodiment of the present invention.

[0110] It can be understood that the first time window and the second time window can be adjusted according to the sampling frequency and the actual scenario, and are usually set when the lateral distance starts to increase or decrease. The preset threshold is set to half of the lane width in the target vehicle driving scenario. The preset ratio can be adjusted according to the required recognition sensitivity and is generally set to 50%.

[0111] By allocating communication time slots and using a fixed base station for ranging, the present invention can obtain the distance data of the vehicle in real time and reliably, and adopts the behavior of the vehicle based on deviation recognition, simplifying the data processing flow from distance data to behavior recognition. The present invention can adapt to different traffic sections and environments and has good adaptability and flexibility.

[0112] It should be noted that the terms "first", "second", etc. in the description embodiments, claims and drawings of the present invention are only used for the purpose of distinguishing descriptions, and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0113] On the other hand, an embodiment of the present invention further provides an electronic device, which includes a memory and a processor. Among them, the memory is used to store a computer program; the processor is used to call and execute the computer program so that the device executes a vehicle behavior recognition method as described in any one of the above.

[0114] In the embodiment of the present invention, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), a field programmable gate array (FPGA for short), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0115] The various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor, or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present invention can be directly embodied as being executed and completed by a hardware decoding processor, or can be executed and completed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The processor reads the information in the storage medium and combines its hardware to complete the steps of the above method.

[0116] The storage medium can be a memory, for example, it can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories.

[0117] Among them, the non-volatile memory can be a read-only memory (ROM for short), a programmable read-only memory (PROM for short), an erasable programmable read-only memory (EPROM for short), an electrically erasable programmable read-only memory (EEPROM for short), or a flash memory.

[0118] The volatile memory can be a random access memory (RAM for short), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM for short), dynamic random access memory (DRAM for short), synchronous dynamic random access memory (SDRAM for short), double data rate synchronous dynamic random access memory (DDR SDRAM for short), enhanced synchronous dynamic random access memory (ESDRAM for short), synchronous link dynamic random access memory (SLDRAM for short), and direct rambus random access memory (DRRAM for short).

[0119] The storage medium described in the embodiments of the present invention is intended to include but not be limited to these and any other suitable types of memories.

[0120] It should be understood that the devices disclosed in the embodiments of the present invention can be implemented in other ways. For example, the division of the above modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the communication connections between the modules can be through some interfaces, and the indirect coupling or communication connection of the server or unit can be in an electrical or other form.

[0121] In addition, each functional module in the various embodiments of the present invention can be integrated in a processing unit, or each module can exist physically alone, or two or more modules can be integrated in a processing unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0122] If the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs that can store program codes.

[0123] The foregoing are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A vehicle behavior recognition method, characterized in that, The method includes: Constructing a lane coordinate system for the target road section; Based on the target vehicle entering the target road section, allocating communication time slots to the target vehicle according to the traffic flow of the target road section, so that the target vehicle can measure the distance from a fixed base station and return distance data; Calculating the coordinates of the target vehicle according to the distance data and performing trajectory prediction to obtain the trajectory of the target vehicle; Judging the deviation of the trajectory of the target vehicle in the lane coordinate system and identifying the behavior of the target vehicle based on the deviation; the behaviors of the target vehicle include overtaking behavior and lane-changing behavior.

2. The vehicle behavior recognition method according to claim 1, characterized in that At least three fixed base stations are set in the target road section, and the layout spacing is set according to the minimum coverage distance of the fixed base stations, and they are arranged at intervals on both sides of the target road section.

3. A vehicle behavior recognition method according to claim 2, characterized in that Obtaining the width of the target road section through the positions of the fixed base stations, and then combining the lane width of the target road section to construct a lane coordinate system for the target road section.

4. A vehicle behavior recognition method according to claim 1, characterized in that, Calculating the coordinates of the target vehicle according to the distance data includes: Using the geometric positioning method to obtain the estimated coordinates of the target vehicle; Using the weighted least squares method and the weighted least squares method based on Taylor to eliminate the errors of the estimated coordinates to obtain the coordinates of the target vehicle.

5. A vehicle behavior recognition method according to claim 1, characterized in that, Judging the deviation of the trajectory of the target vehicle in the lane coordinate system includes: Based on a preset trajectory, obtaining the lateral distance between each trajectory point in the trajectory of the target vehicle and each trajectory point in the preset trajectory; Judging the deviation based on the lateral distance.

6. A vehicle behavior recognition method according to claim 5, characterized in that Identifying the behavior of the target vehicle based on the deviation includes: Based on the fact that within a first time window, there is a continuous lateral distance greater than a preset threshold reaching a preset ratio in the lateral distances, identifying the behavior of the target vehicle as a lane-changing behavior; Based on the fact that the behavior of the target vehicle is a lane-changing behavior, and the lateral distance returns to be less than the preset threshold within a second time window, identifying the behavior of the target vehicle as an overtaking behavior.

7. A vehicle behavior recognition device, characterized in that, The device includes: A monitor for allocating communication time slots to the target vehicle according to the traffic flow of the target road section; A fixed base station for completing distance measurement with the target vehicle; An in-vehicle tag for receiving communication time slots, measuring the distance from the fixed base station, and returning distance data to the data processing center; A data processing center for receiving the distance data and using a vehicle behavior recognition method according to any one of claims 1 to 6 to obtain the behavior of the target vehicle; Wherein, the monitor, the fixed base station and the data processing center are communicatively connected.

8. The vehicle behavior recognition device according to claim 7, wherein The device further includes: A remote control center communicatively connected to the data processing center for displaying the coordinates and trajectory of the target vehicle and making predictions according to the trajectory of the target vehicle.

9. An electronic device, characterized in that, The device includes a memory and a processor, wherein, The memory is used for storing a computer program; The processor is used for calling and executing the computer program so that the device executes a vehicle behavior recognition method according to any one of claims 1 to 6.

Images