Car-following behavior intelligent modeling method and system capable of sensing real-time state of driver

Abstract

The invention discloses a car-following behavior intelligent modeling method and system capable of sensing the real-time state of a driver, and the method specifically comprises the following steps: (1) data acquisition: collecting the driving data of a vehicle in an actual road, the data needing to be acquired comprising the speed and position change information of a current vehicle and a front vehicle, and the change information of the relative distance; (2) data screening and processing: screening the data obtained in the step (1), and selecting historical data of vehicles with following characteristics on one road; (3) constructing a corresponding simulation environment in the Matlab through a method; and (4) simulating the real driving condition of the driver through continuous state change in the Matlab analog simulation environment by using the following state conversion method. According to the method, human factors are considered in the car-following model, so that the car-following accuracy is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic flow research, and in particular to a method and system for intelligent modeling of car-following behavior based on real-time state perception of a driver. Background Art

[0002] In recent years, the modeling of traffic flow following behavior has been a hot topic in the field of traffic flow theory, and the problem of following cars has become increasingly important in traffic engineering and safety research. The development of car following theory is to simulate the movement of vehicles following each other on a one-way street without overtaking. As a pillar of traffic flow theory, by analyzing and predicting vehicle acceleration and modeling car following behavior, the longitudinal interaction between cars can be quantified, so as to understand the characteristics of traffic flow and reveal the internal mechanism of traffic congestion and other traffic phenomena. The human factors of the driver directly affect the driving trajectory, so it is of great significance to consider the human factors in the simulation process. The driver's vehicle driving behavior consists of mechanical control of the vehicle and the driver's random human factors, which include reaction delay, error in distance perception, etc. The traditional car following model oversimplifies the consideration of human factors and largely ignores the human factors of the driver. However, in large-scale urban traffic simulation, especially in the construction of intelligent transportation systems, the study of driver human factors is very important, and it is also necessary to model the human factors in vehicle following movement. Summary of the invention

[0003] Aiming at the problem of lack of human factors in the design of traditional car-following models, the present invention proposes a car-following behavior intelligent modeling method and system with real-time driver status perception, and adds consideration of human factors to the car-following model to ensure the accuracy of car-following.

[0004] In order to solve the above technical problems, the technical solution of the present invention is: a method and system for intelligent modeling of car-following behavior based on real-time state perception of the driver, which specifically includes the following steps:

[0005] Step (1) data acquisition is used to initialize simulation parameters and compare results. The driving data of the vehicle on the actual road is collected. The data to be acquired include the speed, position change information and relative distance change information of the current vehicle and the vehicle in front.

[0006] Step (2) data screening and processing, screening the data obtained in step (1) to select historical data of vehicles with following characteristics on a road.

[0007] Specifically include:

[0008] 1) Select vehicle driving data of longer sections and longer following time, with normal road conditions, moderate traffic, no congestion, good weather conditions, clear vision, and general traffic environment, with large vehicles accounting for no more than 30%.

[0009] 2) When selecting data, the following vehicle and the followed vehicle are considered as a whole and the data is processed as a following unit.

[0010] 3) During the driving process of the selected vehicles, the following vehicle and the followed vehicle are always in the same lane. The following process only involves changes in vehicle speed, position and distance, and no lane change or overtaking behavior is performed.

[0011] 4) When the distance between the vehicle and the vehicle in front is too large (the distance exceeds 2.5 times the speed limit), it is considered that there is no longer a following behavior, and it is regarded as a free acceleration behavior under the speed limit condition.

[0012] Step (3) constructs a corresponding simulation environment in Matlab by using the method. Inputs initial simulation data, sets the driver's initial state according to the initial simulation data, and sets the initial state to a long-distance following car attention concentration state if the initial distance is greater than 40 meters, and sets the initial state to a short-distance following car attention concentration state if the initial distance is less than 40 meters.

[0013] Step (4) uses the following state transformation method to simulate the driver's actual driving situation and verify the model effect through continuous state changes in the Matlab simulation environment.

[0014] On the other hand, the present invention also discloses a car-following behavior intelligent modeling system for real-time state perception of a driver, comprising:

[0015] The data collection base station is used to collect the driving data of vehicles on the actual road. The data to be obtained include the speed, position change information and relative distance change information of the current vehicle and the vehicle in front;

[0016] The data screening module is used to screen valid data from the acquired data, including the following behavior data of the distance between the front and rear vehicles within the effective following distance, and the free acceleration behavior data under the highway speed limit; the Matlab simulator is used to build the corresponding simulation environment. Input the initial simulation data and set the driver's initial state according to the initial simulation data;

[0017] The model evaluation module is used to simulate the driver's actual driving situation through continuous state changes in the Matlab simulation environment, and then verify the model effect.

[0018] In another aspect, the present invention further discloses a computer device, comprising a processor and a computer-readable storage medium; the processor is adapted to execute a computer program;

[0019] The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the above-mentioned intelligent modeling method for following behavior based on real-time state perception of the driver is implemented.

[0020] On the other hand, the present invention also provides a computer-readable storage medium, which stores a computer program, and the computer program is suitable for being loaded by a processor and executing the above-mentioned method for intelligent modeling of following behavior based on real-time state perception of the driver.

[0021] On the other hand, the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the above-mentioned intelligent modeling method of following behavior based on real-time state perception of the driver.

[0022] The present invention has the following characteristics and beneficial effects:

[0023] The above technical solution is adopted to divide the driver's driving process into multiple different states according to the distance of the following car and whether the driver is concentrating. This method uses the following car state as a discretization condition and the continuous motion process of the vehicle as a continuous condition, and models the following car process as a piecewise continuous process through a state machine. The results show that this technical solution can simulate the real driving process more effectively and solve the problem that the traditional following car model has a large deviation from the real driving situation in the simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0025] Figure 1 This is a flow chart of a system logic framework and a method for intelligent modeling of car-following behavior based on real-time driver status perception according to an embodiment of the present invention.

[0026] Figure 2 This is a state change diagram in this embodiment. DETAILED DESCRIPTION

[0027] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0028] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships 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. In addition, the terms "first", "second", and the like are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", and the like may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0029] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood by specific circumstances.

[0030] Example 1

[0031] This embodiment provides a method for intelligent modeling of car-following behavior based on real-time state perception of the driver.

[0032] In this embodiment, based on the long following distance and the short following distance, the following eight states are created to distinguish the driver's following state:

[0033] 1. Pay attention when following a vehicle at a long distance: refers to the normal driving situation when the distance between the vehicle and the vehicle is long.

[0034] 2. Distraction during long-distance following: This means that the distance between the following vehicles is long, the driver is somewhat distracted and has difficulty maintaining attention.

[0035] 3. Long distance following and short reaction time: It means that the distance between the following vehicles is long and there is a normal reaction delay.

[0036] 4. Long reaction time when following a vehicle at a long distance: It means that the distance between the following vehicles is longer, and the reaction time becomes longer due to distraction and other conditions.

[0037] 5. Pay attention when following a vehicle at a short distance: refers to normal driving conditions when the distance between the vehicle and the vehicle is short.

[0038] 6. Distraction during short-distance following: This means that the distance between the following vehicles is short, the driver is somewhat distracted and has difficulty maintaining attention.

[0039] 7. Short reaction time when following a vehicle at short distance: It means that the distance between the following vehicle and the normal reaction delay is very short.

[0040] 8. Short-distance following vehicle reaction time: It means that the distance between the following vehicles is shorter, and the reaction time becomes longer due to distraction and other conditions.

[0041] This method is based on short-distance following and long-distance following, and is combined with other states. Each state transition process is a response to changes in one or more parameters in the model. Taking short-distance following as an example, when the estimated error with the front vehicle exceeds the threshold, the driver's state will be changed from the short-distance following state of concentration to the short-distance short reaction time state. When the driver's concentration time exceeds a certain period of time (this article sets the driver's concentration time to 8 seconds), it will be changed from the short-distance following state of concentration to the short-distance distraction state.

[0042] Specifically, Figure 1 As shown, the following steps are included:

[0043] Step (1) data acquisition is used to initialize simulation parameters and compare results. The driving data of the vehicle on the actual road is collected. The data to be acquired include the speed, position change information and relative distance change information of the current vehicle and the vehicle in front.

[0044] Step (2) data screening: in the acquired data, valid data is screened, including following behavior data when the distance between the front and rear vehicles is within the effective following distance, free acceleration behavior data under highway speed limit conditions (the distance between the following vehicle and the followed vehicle is too large), and other information.

[0045] In this embodiment, data screening includes:

[0046] 1) Select vehicle driving data of longer sections and longer following time, with normal road conditions, moderate traffic, no congestion, good weather conditions, clear vision, and general traffic environment, with large vehicles accounting for no more than 30%.

[0047] 2) When selecting data, the following vehicle and the followed vehicle are considered as a whole and the data is processed as a following unit.

[0048] 3) During the driving process of the selected vehicles, the following vehicle and the followed vehicle are always in the same lane. The following process only involves changes in vehicle speed, position and distance, and no lane change or overtaking behavior is performed.

[0049] 4) When the distance between the vehicle and the vehicle in front is too large (the distance exceeds 2.5 times the speed limit), it is considered that there is no longer a following behavior, and it is regarded as a free acceleration behavior under the speed limit condition.

[0050] Step (3) constructs a corresponding simulation environment in Matlab, inputs initial simulation data, sets the driver's initial state according to the initial simulation data, and sets the initial state to a long-distance following car attention concentration state if the initial distance is greater than s, s=40 meters in this embodiment, and sets the initial state to a short-distance following car attention concentration state if the initial distance is less than 40 meters.

[0051] Step (4) uses the following state transformation method to simulate the driver's actual driving situation and verify the model effect through continuous state changes in the Matlab simulation environment.

[0052] Specifically, Figure 2 As shown, the state transformation method in this embodiment is as follows:

[0053] 1) In the state of short-distance following with concentrated attention, when the estimated error of the position of the preceding vehicle is greater than k 1 The driver needs to react. In this embodiment, k 1 =1m, at this time, because t<8, the driver's state changes to short-distance following and short reaction time, the estimated distance of the front vehicle is updated, and the timer t=0 is reset.

[0054] When When t=0,

[0055] Among them, The calculation process is as follows:

[0056] First calculate the estimated acceleration:

[0057]

[0058] In the formula represents the start of the i-th reaction of vehicle n, Indicates the end of the reaction by Represents the duration of the ith reaction. When the first reaction begins, the estimated acceleration of the vehicle ahead is the value input at time zero. Each subsequent calculation of the estimated acceleration is calculated by dividing the speed difference of the previous reaction time by the reaction duration.

[0059] By calculating the estimated acceleration of the preceding vehicle, the estimated speed of the preceding vehicle can be calculated.

[0060]

[0061] The calculation equation for the estimated position can be obtained by estimating the speed of the preceding vehicle:

[0062]

[0063] 2) In the state of short-distance following with concentration, within the time when attention can be concentrated, the error of the leading vehicle position estimation has been within the threshold range, and the driver has not responded. At this time, due to limited attention, the driver cannot concentrate for a long time. When t>=8, the driver switches to the short-distance following distracted state.

[0064] When .

[0065] 3) When the driver is in a state of short-distance following and short reaction time, after making a corresponding response to the change, he will return to the state of short-distance following and concentration.

[0066] That is, when t≥0.5, let t=0,

[0067] 4) In the case of distracted following at short distances, the estimated error of the position of the vehicle in front is greater than k 1 , the driver's state is switched to the short-distance following vehicle length reaction time, and the driver makes a corresponding response within the limited reaction time, and updates the timer t=0.

[0068] When When t=0,

[0069] 5) When following a vehicle at a short distance and having a long reaction time, after responding to the changes, the attention will be refocused and will return to the state of concentrated attention when following a vehicle at a short distance.

[0070] That is, when t≥0.5, let t=0,

[0071] 6) When the distance to the vehicle ahead is too large (greater than 40 meters), the driver cannot perceive the small distance error and needs to update the threshold of the distance estimation error. At this time, the threshold will become k 2 In this embodiment, k 2 =2.5m, and then set the state to the concentration state at a long following distance.

[0072] That is, when x n -x n-1 ≥s, let t=0,

[0073] 7) When the distance to the vehicle ahead is small (less than 40 meters), the sensitivity to the distance increases, and the threshold of the distance estimation error needs to be updated to set the state to the focused state under a short following distance.

[0074] That is, when x n -x n-1 When <s, let t=0,

[0075] 8) When the estimated error of the position of the vehicle ahead is greater than k 1 The driver needs to react. At this time, because t<8, the driver's state changes to long-distance following with short reaction time, the estimated distance to the vehicle in front is updated, and the timer t=0 is reset.

[0076] When When t=0,

[0077] 9) In the state of long-distance following with concentration, within the time when attention can be concentrated, the error of the leading vehicle position estimation has been within the threshold range, and the driver has not responded. At this time, due to limited attention, the driver cannot concentrate for a long time. When t>=8, the driver switches to the long-distance following distracted state.

[0078] When .

[0079] 10) When the driver is in a state of short reaction time and long distance following, after making a corresponding response to the change, he will return to the state of concentration and attention in long distance following.

[0080] That is, when t≥0.5, let t=0,

[0081] 11) In the case of long-distance distracted following, the estimated position error of the preceding vehicle is greater than k 1 , the driver's state is switched to the long-distance following vehicle reaction time, and the driver makes a corresponding response within the limited reaction time, and updates the timer t=0.

[0082] When When t=0,

[0083] 12) When following a vehicle for a long distance and having a long reaction time, after responding to the changes, the attention will be refocused and will return to the state of concentrated attention when following a vehicle for a long distance.

[0084] That is, when t≥0.5, let t=0,

[0085] In order to verify the effect of the above method, the present invention models the driver's state changes through a state machine and analyzes and evaluates the experimental results.

[0086] Specifically, the present invention uses two well-known car-following models, the Intelligent Driver Model (IDM) and the Full Velocity Difference model (FVD), for expansion and comparison. These two models have fewer parameters, are widely usable, and have fewer additional influencing factors, so they can well test the method of adding human factors to improve the efficiency of driver driving trajectory simulation.

[0087] The model uses the public NGSIM (Next Generation Simulation) trajectory dataset as experimental data. The NGSIM dataset is a traffic dataset developed by the Federal Highway Administration of the United States for studying and simulating traffic mobility. This dataset mainly contains the running trajectory data of vehicles on highways, which can be used for research on traffic flow simulation, behavior analysis, and vehicle interaction. Data with longer following time and more obvious following characteristics are extracted from the dataset as the source of experimental data.

[0088] Firstly, the spacing change data of the original IDM and FVD were recorded and compared with the actual data. Then, human factors were added to the model. By comparing the spacing error and root mean square error (RMSE), it was found that compared with the original simulation, the simulation results after adding human factors were obviously close to the actual data curve. The spacing error decreased significantly during driving, making the spacing more realistic and the simulation effect significantly improved.

[0089] Example 2

[0090] This embodiment discloses a car-following behavior intelligent modeling system for real-time driver status perception, including:

[0091] The data collection base station is used to collect the driving data of vehicles on the actual road. The data to be obtained include the speed, position change information and relative distance change information of the current vehicle and the vehicle in front;

[0092] The data screening module is used to screen valid data from the acquired data, including the following behavior data of the distance between the front and rear vehicles within the effective following distance, and the free acceleration behavior data under the highway speed limit; the Matlab simulator is used to build the corresponding simulation environment. Input the initial simulation data and set the driver's initial state according to the initial simulation data;

[0093] The model evaluation module is used to simulate the driver's actual driving situation through continuous state changes in the Matlab simulation environment, and then verify the model effect.

[0094] In this embodiment, the specific working process details of each module refer to the process in Embodiment 1, which will not be repeated here.

[0095] It is understandable that the above modules can be separately or completely combined into one or more other units, or one (or more) of the units can be further divided into multiple smaller units in function, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application. The above units are divided based on logical functions. In practical applications, the functions of one unit can also be implemented by multiple units, or the functions of multiple units can be implemented by one unit.

[0096] In other embodiments of the present application, the construction system may also include other units. In practical applications, these functions may also be implemented with the assistance of other units, and may be implemented by collaboration of multiple units.

[0097] According to another embodiment of the present application, the program can be executed by running on a general computing device such as a computer including a central processing unit (CPU), a random access memory (RAM), a read-only memory (ROM), and other processing elements and storage elements.

[0098] The computer program (including program code) of each step involved in the corresponding method described in Example 1 is used to construct the system described in this embodiment and to implement the construction method of the embodiment of the present application. The computer program can be recorded on, for example, a computer-readable recording medium, and loaded into the above-mentioned computing device through the computer-readable recording medium and run therein.

[0099] Example 3

[0100] This embodiment also discloses a computer device, including a processor and a computer-readable storage medium; the processor is adapted to execute a computer program;

[0101] The computer-readable storage medium stores a computer program, and the computer program is executed by the processor.

[0102] Specifically, the processor is configured to perform the following process:

[0103] Step 1: Collecting driving data of vehicles on the actual road, wherein the collected driving data includes speed, position change information, and relative distance change information between the current vehicle and the vehicle ahead;

[0104] Step 2: According to the preset screening conditions, historical data of vehicles with car-following behavior on a road are selected from the driving data;

[0105] Step 3: Based on the selected vehicle historical data, a corresponding simulation environment is constructed in Matlab, initial simulation data is input, and the initial state of the driver is set according to the initial simulation data;

[0106] Step 4: Through state transformation, simulate the driver's actual driving situation in the Matlab simulation environment and verify the model effect.

[0107] Example 4

[0108] This embodiment discloses a computer-readable storage medium, which stores a computer program. It is understandable that the computer-readable storage medium here can include both a built-in storage medium in an electronic device and an extended storage medium supported by the electronic device. The computer-readable storage medium provides a storage space, which stores a processing system of the electronic device.

[0109] Furthermore, the storage space also stores one or more instructions suitable for being loaded and executed by the processor, and these instructions may be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here may be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as at least one disk memory; optionally, it may also be at least one computer-readable storage medium located away from the aforementioned processor.

[0110] Specifically, the computer program is suitable for being loaded by a processor and executing the following process:

[0111] Step 1: Collecting driving data of vehicles on the actual road, wherein the collected driving data includes speed, position change information, and relative distance change information between the current vehicle and the vehicle ahead;

[0112] Step 2: According to the preset screening conditions, historical data of vehicles with car-following behavior on a road are selected from the driving data;

[0113] Step 3: Based on the selected vehicle historical data, a corresponding simulation environment is constructed in Matlab, initial simulation data is input, and the initial state of the driver is set according to the initial simulation data;

[0114] Step 4: Through state transformation, simulate the driver's actual driving situation in the Matlab simulation environment and verify the model effect.

[0115] Example 5

[0116] This embodiment discloses a computer program product, which includes a computer program. When the computer program is executed by a processor, the computer instructions are stored in a computer-readable storage medium. The processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes:

[0117] Step 1: Collecting driving data of vehicles on the actual road, wherein the collected driving data includes speed, position change information, and relative distance change information between the current vehicle and the vehicle ahead;

[0118] Step 2: According to the preset screening conditions, historical data of vehicles with car-following behavior on a road are selected from the driving data;

[0119] Step 3: Based on the selected vehicle historical data, a corresponding simulation environment is constructed in Matlab, initial simulation data is input, and the initial state of the driver is set according to the initial simulation data;

[0120] Step 4: Through state transformation, simulate the driver's actual driving situation in the Matlab simulation environment and verify the model effect.

[0121] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions and variations of these embodiments including components are made without departing from the principles and spirit of the present invention, and still fall within the scope of protection of the present invention.

Claims

1. An intelligent modeling method for car-following behavior based on real-time state perception of the driver, characterized in that: The steps include: Step 1: Collecting driving data of vehicles on the actual road, wherein the collected driving data includes speed, position change information, and relative distance change information between the current vehicle and the vehicle ahead; Step 2: According to the preset screening conditions, historical data of vehicles with car-following behavior on a road are selected from the driving data; Step 3: Based on the selected vehicle historical data, a corresponding simulation environment is constructed in Matlab, initial simulation data is input, and the initial state of the driver is set according to the initial simulation data; Step 4: Through state transformation, simulate the driver's actual driving situation in the Matlab simulation environment and verify the model effect.

2. The intelligent modeling method for following car behavior based on driver real-time state perception according to claim 1, characterized in that: In step 2, when screening the driving data, the following vehicle and the followed vehicle are regarded as a whole and the data are processed as a following unit.

3. The intelligent modeling method for following car behavior based on driver real-time state perception according to claim 1, characterized in that: The screening conditions include: Select the vehicle driving data of longer sections and longer following time, with normal road conditions, moderate traffic, no congestion, good weather conditions, clear vision, and general traffic environment, and the proportion of large vehicles does not exceed 30%; The following vehicle and the vehicle being followed are always in the same lane. The following process only involves changes in vehicle speed, position and distance, and there is no lane change or overtaking behavior. A threshold value of the distance between the current vehicle and the vehicle in front is set. If the distance between the current vehicle and the vehicle in front exceeds the threshold value, it is considered that there is no longer a following behavior, and it is regarded as a free acceleration behavior under the speed limit condition.

4. The intelligent modeling method for following car behavior based on driver real-time state perception according to claim 3 is characterized in that: The threshold of the distance between the current vehicle and the preceding vehicle is 2.5 times the current road speed limit.

5. The intelligent modeling method for following car behavior based on driver real-time state perception according to claim 1, characterized in that: The initial state is defined as: when the initial distance is greater than s, the initial state is set to a long-distance following vehicle with concentrated attention state; when the initial distance is less than s meters, the initial state is set to a short-distance following vehicle with concentrated attention state.

6. The intelligent modeling method for following car behavior based on driver real-time state perception according to claim 1, characterized in that: In step 4, the timer t is set to 0 at the beginning of the simulation, and the initial state is set to the short-distance following state.

7. The intelligent modeling method for following car behavior based on driver real-time state perception according to claim 6, characterized in that: The state transformation method comprises: In the state of short-distance following with concentration, when the estimated error of the position of the vehicle in front is greater than the threshold k1, the driver needs to react. At this time, because t<8, the driver's state changes to short-distance following with short reaction time, updates the estimated distance of the vehicle in front, and resets the timer t=0. When season in, is the estimated position of the vehicle in front of vehicle n, x n-1 is the actual position of the vehicle in front of vehicle n, and t represents the value of the timer. In the state of short-distance following with concentration, the position estimation error of the leading vehicle is always within the threshold k1 range within the time when the attention can be concentrated, and the driver has not responded. At this time, due to limited attention, the driver cannot concentrate for a long time. When t>=8, the driver switches to the state of short-distance following with distraction. When When converting; When the driver is in a short-distance following state with a short reaction time, after making a corresponding response to the change, he will return to the state of short-distance following with a focused attention. That is, when t≥0.5, let in, is the estimated speed of the vehicle in front of vehicle n, v n-1 is the actual speed of the vehicle in front of vehicle n, and t represents the value of the timer. In the distracted state of short-distance following, the estimated error of the position of the vehicle in front is greater than the threshold k1, and the driver's state is switched to the long reaction time state of short-distance following. The driver responds accordingly within the limited reaction time and updates the timer t=0. When season When following a car at a short distance and having a long reaction time, the attention will be refocused after responding to the change, and will return to the state of focusing attention when following a car at a short distance. That is, when t≥0.5, let When the distance to the vehicle ahead is greater than s meters, the driver cannot perceive a small distance error. The threshold of the distance estimation error needs to be updated. At this time, the threshold becomes k2, and the state is set to the state of concentration under long following distance. That is, when x n -x n-1 ≥s, let When the distance to the vehicle ahead is less than s meters, the sensitivity to the distance increases, and the threshold k1 of the distance estimation error needs to be updated, and the state is set to the state of attention concentration under short following distance. That is, when x n -x n-1 <s, let In the state of long-distance following with concentration, when the estimated error of the position of the vehicle in front is greater than the threshold k2, the driver needs to react. At this time, because t<8, the driver's state changes to long-distance following with short reaction time, updates the estimated distance of the vehicle in front, and resets the timer t=0. When season In the state of long-distance following with concentration, the position estimation error of the leading vehicle is always within the threshold k1 range within the time when the attention can be concentrated, and the driver has not responded. At this time, due to limited attention, the driver cannot concentrate for a long time. When t>=8, the driver switches to the state of long-distance following with distraction. When When converting; When the driver is in a state of short reaction time in long-distance following, he will return to the state of concentration in long-distance following after making corresponding responses. That is, when t≥0.5, let In the distracted state of long-distance following, the estimated error of the position of the vehicle in front is greater than the threshold k1, and the driver's state is switched to the long-distance following state with a long reaction time. The driver responds accordingly within the limited reaction time and updates the timer t=0. When season When following a car for a long distance and taking a long reaction time, the attention will be refocused after responding to the change, and will return to the state of long-distance following car with a long reaction time. That is, when t≥0.5, let 8. The intelligent modeling method for following car behavior based on driver's real-time state perception according to claim 7 is characterized in that: The estimated acceleration is calculated as follows: In the formula represents the estimated acceleration of the vehicle in front of vehicle n, v n-1 represents the actual speed of the vehicle in front of vehicle n, represents the start of the i-th reaction of vehicle n, Indicates the end of the reaction by represents the duration of the ith reaction, t represents the value of the timer; a n-1 (0) means that when the first reaction begins, the estimated acceleration of the vehicle in front of vehicle n uses the value input at time zero. Each subsequent calculation of the estimated acceleration is calculated by dividing the speed difference of the previous reaction time by the reaction duration.

9. The intelligent modeling method for following car behavior based on driver's real-time state perception according to claim 8, characterized in that: The estimation method of the front vehicle speed is: By calculating the estimated acceleration of the front vehicle, the estimated speed of the front vehicle can be calculated: In the formula represents the estimated speed of the vehicle in front of vehicle n, v n-1 represents the actual speed of the vehicle in front of vehicle n, represents the estimated acceleration of the vehicle in front of vehicle n, t represents the value of the timer, Indicates the end of the i-th reaction of vehicle n.

10. The intelligent modeling method for following car behavior based on driver real-time state perception according to claim 9, characterized in that: The estimation method of the position of the preceding vehicle is: The calculation equation for the estimated position can be obtained by estimating the speed of the preceding vehicle: In the formula represents the estimated position of the vehicle in front of vehicle n, x n-1 represents the actual position of the vehicle in front of vehicle n, represents the estimated speed of the vehicle in front of vehicle n, t represents the value of the timer, represents the start of the i-th reaction of vehicle n.

11. An intelligent modeling system for following car behavior based on real-time driver status perception, characterized in that: include The data collection base station is used to collect the driving data of vehicles on the actual road. The data to be obtained include the speed, position change information and relative distance change information of the current vehicle and the vehicle in front; A data screening module is used to screen valid data from the acquired data, including the following behavior data of the distance between the front and rear vehicles within the effective following distance, and the free acceleration behavior data under the highway speed limit condition; Matlab simulator, used to build a corresponding simulation environment, input initial simulation data, and set the driver's initial state according to the initial simulation data; The model evaluation module is used to simulate the driver's actual driving situation through continuous state changes in the Matlab simulation environment, and then verify the model effect.

12. A computer device, characterized in that: The device comprises a processor and a computer-readable storage medium; the processor is adapted to execute a computer program; The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the method for intelligent modeling of following behavior based on real-time state perception of a driver as described in any one of claims 1-10 is implemented.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing a method for intelligent modeling of following behavior of a driver's real-time state perception as described in any one of claims 1-10.

14. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, it implements the intelligent modeling method of following behavior of driver real-time state perception as described in any one of claims 1-10.

Images