A vehicle simulation trajectory fitting calculation method based on information completion

Abstract

The present invention discloses a method for calculating the fitting degree of a vehicle simulation trajectory based on information completion, the method comprising: collecting road traffic micro information, determining all methods for completing the vehicle lane change trajectory, including six methods of half-completing the starting and ending points, starting point completion, and ending point completion from the perspectives of time and space, respectively, and then determining the completion method of the vehicle lane change trajectory according to the length, time, speed and other information of the simulated lane change trajectory and the real lane change trajectory, and finally updating the simulated lane change trajectory and the real lane change trajectory, and calculating the fitting degree of the vehicle simulation trajectory from the perspective of time / space. The method can make the simulated lane change trajectory consistent with the real lane change trajectory in terms of time or space dimension, complete the refined evaluation of the whole process, obtain a more scientific and reasonable fitting degree, and then help select high-quality vehicle trajectories, and provide a high-quality sample basis for intelligent vehicle learning or simulation model calibration.

Description

Technical Field

[0001] The present invention relates to the field of intelligent traffic control, and in particular to a vehicle simulation trajectory fitting degree calculation method based on information completion. Background Art

[0002] Vehicle trajectories are divided into longitudinal following trajectories and lateral lane-changing trajectories. Usually, lane-changing behavior involves multiple lanes and multiple vehicles, so its risk factor is higher than that of following a vehicle. Traffic simulation can be used for traffic prediction, and then to warn dangerous driving behaviors in advance to avoid collision accidents. The premise of simulation is the accurate imitation of real behavior. Therefore, how to evaluate the fit of vehicle trajectories is one of the key technologies. High fit means high restoration, which is of great significance to improving the scientific nature of the simulation platform and also has great guiding significance for improving the safety of the traffic system.

[0003] The patent "A method and device for lane-changing decision-making of an autonomous vehicle" (CN201810556054.4) solves the optimal lane-changing decision strategy for an autonomous vehicle based on the motion state at each moment, including following and lane-changing behaviors. Its evaluation function includes the distance-to-collision time at each moment, while the position and acceleration parameters are certain values; the patent "A method and device for lane-changing control" (CN201711064941.1) focuses on the environmental data when the lane-changing behavior is executed. When controlling the lane-changing behavior, the parameters in the calculation report are all within the preset time period, that is, the behavior of the entire process is ignored. Therefore, the temporal and spatial granularity of traffic parameters considered in the published literature and patents is relatively low in detail, and none of them involves the fit analysis of the entire process of the lane-changing trajectory from a temporal or spatial perspective, and the inconsistency between the simulated trajectory and the real trajectory in the temporal and spatial scale has not yet been solved. Summary of the invention

[0004] Purpose of the invention: In order to overcome the deficiencies of the above-mentioned prior art, the purpose of the present invention is to propose a vehicle simulation trajectory fitting calculation method based on information completion, collect the real running trajectory and simulated running trajectory of the target vehicle, and in order to handle the lane change trajectory consistency of the target vehicle from the perspective of time or space, for the trajectory with a short lane change duration or a short lane change length, half-complete, start-point complete, and end-point complete are completed at its starting and end points, and finally the trajectory fitting degree consistent in time / space is calculated. This method can process the real lane change trajectory and the simulated lane change trajectory into a unified dimension in space or time, and can be used to compare the deviations between the two in a one-to-one correspondence. The accumulated speed deviation of the entire lane change process is used as a fitting judgment condition, which can more scientifically and accurately reflect the simulation level.

[0005] Technical solution: To solve the above technical problems, the technical solution adopted by the present invention is:

[0006] Step 1: Collect road traffic micro information;

[0007] Step 2: Determine all ways to complete the vehicle's lane change trajectory;

[0008] Step 3: Calculate the length error rate between the simulated lane-changing trajectory and the actual lane-changing trajectory. When the length error rate is greater than the critical value P L_th , choose to complete the lane-changing trajectory from the time perspective and go to step 4; otherwise, choose to complete the lane-changing trajectory from the space perspective and go to step 6; where the critical value P L_th The default value is 20%;

[0009] Step 4: Determine the lane-changing duration between the simulated lane-changing trajectory and the actual lane-changing trajectory, and the absolute difference between them is greater than the critical value T T_th When , the completion method is to complete the start and end points in half from the time perspective, and go to step 8; otherwise, go to step 5; where the critical value T T_th The default value is 2s;

[0010] Step 5: Calculate the time index of the simulated lane-changing trajectory and the real lane-changing trajectory by segment. When the time index is greater than 0, the completion method is to complete the starting point from the time perspective, and go to step 8; otherwise, the completion method is to complete the end point from the time perspective, and go to step 8;

[0011] Step 6: Calculate the absolute error of the speed standard deviation between the simulated lane-changing trajectory and the actual lane-changing trajectory. When the absolute error is greater than the critical value S V_th When , the completion method is to complete the starting and ending points in half from a spatial perspective, and go to step 9; otherwise, go to step 7; where the critical value S V_th The default value is 10km / s;

[0012] Step 7: Calculate the speed standard deviation index of the simulated lane-changing trajectory and the actual lane-changing trajectory. When the speed standard deviation index is greater than 0, the completion method is to complete the starting point from a spatial perspective, and go to step 9; otherwise, the completion method is to complete the end point from a spatial perspective, and go to step 9;

[0013] Step 8: Update the simulated lane-changing trajectory and the real lane-changing trajectory, calculate the vehicle simulation trajectory fitting degree from the time perspective, and the step ends;

[0014] Step 9: Update the simulated lane-changing trajectory and the actual lane-changing trajectory, calculate the vehicle simulation trajectory fitting degree from a spatial perspective, and the step ends.

[0015] Preferably, in step (1), the road traffic microscopic information includes the speed of the target vehicle recorded in unit time scale and unit space scale, the unit time scale is Δt, and the unit space time scale is Δd; the simulated lane change trajectory length L of the target vehicle sim and the actual lane-changing trajectory length L real, which is the distance along the driving direction of the road; the simulated lane change duration of the target vehicle is T sim and the actual lane change duration T real ; The duration of the simulated lane-changing trajectory of the target vehicle in the current lane and the target lane is and The actual lane-changing trajectory of the target vehicle lasts for two periods of time in the current lane and the target lane respectively. and The average speeds of the target vehicle during the entire simulated lane-changing process and the entire real lane-changing process are and The average speeds of the target vehicle in the simulated lane changing process and the actual lane changing process in the current lane are and The average speeds of the target vehicle in the simulated lane changing process and the actual lane changing process in the target lane are and

[0016] Preferably, in step (2), the vehicle lane change trajectory completion method includes the following six categories:

[0017] From the perspective of time, the starting and ending points are completed in half: the simulated lane change trajectory is recorded as trajectory A, and the real lane change trajectory is recorded as trajectory B; the absolute difference in the lane change duration between trajectory A and trajectory B is T dif =|T sim -T real |, then 0.5T is added at the start and end of the track with a shorter lane change time dif The following trajectory of the lane change duration is then converted into a new trajectory C. The trajectory with a longer lane change duration remains unchanged and is recorded as trajectory D.

[0018] Completing the starting point from the perspective of time: Add T at the starting point of the trajectory with a shorter lane change duration dif The following trajectory of the lane change duration is then converted into a new trajectory C. The trajectory with a longer lane change duration remains unchanged and is recorded as trajectory D.

[0019] Completing the endpoint from a time perspective: Add T at the end of the trajectory with a shorter lane change duration dif The following trajectory of the lane change duration is then converted into a new trajectory C. The trajectory with a longer lane change duration remains unchanged and is recorded as trajectory D.

[0020] From a spatial perspective, the starting and ending points are completed in half: the difference in the lane change trajectory length between trajectory A and trajectory B is L dif =|L sim -L real |, then add 0.5L at the start and end of the shorter lane change trajectory difThe following track of the same length is then formed into a new track C. The track with the longer lane-changing track remains unchanged and is recorded as track D.

[0021] Completing the starting point from a spatial perspective: Add L at the starting point of the shorter lane change trajectory. dif The following track of the same length is then formed into a new track C. The track with the longer lane-changing track remains unchanged and is recorded as track D.

[0022] Completing the end point from a spatial perspective: Add L at the end point of the shorter lane change trajectory. dif The following trajectory of the same length is then used to form a new trajectory C. The longer lane-changing trajectory remains unchanged and is recorded as trajectory D.

[0023] Preferably, in step (3), the length error rate P between the simulated lane-changing trajectory and the actual lane-changing trajectory is L The calculation method is

[0024] Preferably, in step (5), the time index of the simulated lane-changing trajectory and the real lane-changing trajectory segmented statistics is The calculation method is

[0025] Preferably, in step (6), the absolute error S of the speed standard deviation of the simulated lane-changing trajectory and the actual lane-changing trajectory is V , which is calculated as:

[0026] S V =|S sim -S real |

[0027]

[0028]

[0029]

[0030]

[0031] Where S sim , S real are the speed standard deviations of the target vehicle in the entire simulated lane-changing process and the entire real lane-changing process, and v sim (i) v real (i) is the instantaneous speed of the target vehicle in the entire simulated lane changing process and the entire real lane changing process in the i-th unit time interval, N sim 、N real are the unit time intervals of the entire simulated lane changing process and the entire real lane changing process respectively.

[0032] Preferably, in step (7), the speed standard deviation index of the simulated lane-changing trajectory and the real lane-changing trajectory is segmented statistically The calculation method is

[0033]

[0034]

[0035]

[0036]

[0037]

[0038]

[0039]

[0040]

[0041]

[0042] In the formula are the speed standard deviations of the target vehicle in the current lane and the target lane during the simulated lane change process, are the speed standard deviations of the target vehicle in the current lane and the target lane during the actual lane change process, are the instantaneous speed of the target vehicle in the simulated lane changing process and the real lane changing process in the i-th unit time interval, respectively. are the instantaneous speed of the target vehicle in the i-th unit time interval during the simulated lane changing process and the real lane changing process in the target lane, are the unit time intervals in the simulated lane change process of the current lane and the target lane, are the unit time intervals in the actual lane changing process in the current lane and the target lane, respectively.

[0043] Preferably, in step (8), the vehicle simulation trajectory fitting degree J is calculated from the time perspective T , which is calculated as:

[0044] T max =max{T real ,T sim}

[0045]

[0046]

[0047]

[0048] In the formula is the maximum number of parts that the lane-changing trajectory is divided into according to the time angle after the time length is unified. is the speed of the adjusted simulated lane-changing trajectory at time (t0+i·Δt), is the speed of the adjusted real lane-changing trajectory at the time (t0+i·Δt), t0 is the starting time of the adjusted lane-changing trajectory, α T is the time integrity factor.

[0049] Preferably, in step (9), the vehicle simulation trajectory fitting degree J is calculated from a spatial perspective L , which is calculated as:

[0050] L max =max{L real ,L sim}

[0051]

[0052]

[0053]

[0054] In the formula is the maximum number of equal parts of the lane-changing trajectory divided by spatial angle after the time length is unified, is the speed of the adjusted simulated lane-changing trajectory at the position (t0+i·Δd), is the speed of the adjusted real lane-changing trajectory at the position (t0+i·Δd), d0 is the starting position of the adjusted lane-changing trajectory, α L is the spatial integrity coefficient.

[0055] Beneficial effect: The present invention discloses a method for calculating the fitting degree of vehicle simulation trajectory based on information completion, which fully considers the inconsistency between the real lane changing trajectory and the simulated lane changing trajectory in lane changing time and trajectory length, introduces the idea of ​​information completion based on key parameters such as length, time, speed, etc., and completes the shorter trajectory in time / space. The completion methods include half completion of the starting and ending points, starting point completion, and end point completion, so that the real lane changing trajectory and the simulated lane changing trajectory are consistent in time / space, and then the two can be compared in a refined and full process to determine the fitting degree of the simulated lane changing trajectory, which provides guidance for evaluating the effectiveness and scientificity of the simulation model, helps to improve the performance of the traffic virtual simulation platform, and provides strong support for the verification of traffic plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a flow chart of the method of the present invention.

[0057] Figure 2 It is a schematic diagram of an implementation case in the present invention.

[0058] Figure 3 It is a data statistical diagram of the implementation case of the present invention, wherein (a) is the starting and ending points being filled in half, (b) is the starting point being filled, and (c) is the ending point being filled. DETAILED DESCRIPTION

[0059] In order to make the contents of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0060] In one embodiment, the simulated lane-changing trajectory A and the real lane-changing trajectory B are statistically analyzed as shown in Table 1:

[0061] Table 1

[0062]

[0063]

[0064] In Table 1, t0 and t end denote the starting time and ending time of the initial vehicle lane-changing trajectory, d0 and d end They represent the starting position and the end position of the initial vehicle lane-changing trajectory respectively. The unit time scale Δt and the unit space time scale Δd are 0.1s and 1m respectively.

[0065] In one embodiment, Figure 1 As shown, a vehicle simulation trajectory fitting calculation method based on information completion is provided, and the specific steps are as follows:

[0066] Step 1: Collect road traffic micro information

[0067] Simulated lane-changing trajectory length L sim and the actual lane-changing trajectory length L real , which is the distance along the driving direction of the road; the simulated lane change duration of the target vehicle is T sim and the actual lane change duration T real ; The duration of the simulated lane-changing trajectory of the target vehicle in the current lane and the target lane is and The actual lane-changing trajectory of the target vehicle lasts for two periods of time in the current lane and the target lane respectively. and The average speeds of the target vehicle during the entire simulated lane-changing process and the entire real lane-changing process are and The average speeds of the target vehicle in the simulated lane changing process and the actual lane changing process in the current lane are and The average speeds of the target vehicle in the simulated lane changing process and the actual lane changing process in the target lane are and

[0068] Step 2: Determine the vehicle lane change trajectory completion method

[0069] From the perspective of time and space, it can be divided into two categories, each of which is subdivided into starting and ending half completion, starting point completion, and end point completion, such as Figure 2 There are 6 ways to complete the task, as shown in (a), (b) and (c) in the figure:

[0070] ①Complete the start and end points in half from the perspective of time: the simulated lane-changing trajectory is recorded as trajectory A, and the real lane-changing trajectory is recorded as trajectory B; the absolute difference between the lane-changing duration of trajectory A and trajectory B is T dif =|T sim -T real |, then 0.5T is added at the start and end of the track with a shorter lane change time dif The following trajectory of the lane change duration is then converted into a new trajectory C. The trajectory with a longer lane change duration remains unchanged and is recorded as trajectory D.

[0071] ②Complete the starting point from the perspective of time: Add T at the starting point of the trajectory with a shorter lane change duration dif The following trajectory of the lane change duration is then converted into a new trajectory C. The trajectory with a longer lane change duration remains unchanged and is recorded as trajectory D.

[0072] ③Complete the endpoint from the perspective of time: Add T at the end of the trajectory with a shorter lane change duration dif The following trajectory of the lane change duration is then converted into a new trajectory C. The trajectory with a longer lane change duration remains unchanged and is recorded as trajectory D.

[0073] ④Complete the starting and ending points in half from a spatial perspective: The difference in the lane change track length between track A and track B is L dif =|L sim -L real |, then add 0.5L at the start and end of the shorter lane change trajectory dif The following track of the same length is then formed into a new track C. The track with the longer lane-changing track remains unchanged and is recorded as track D.

[0074] ⑤Complete the starting point from a spatial perspective: Add L at the starting point of the shorter lane change trajectory. dif The following track of the same length is then formed into a new track C. The track with the longer lane-changing track remains unchanged and is recorded as track D.

[0075] ⑥Complete the end point from a spatial perspective: Add L at the end point of the shorter lane change trajectory. difThe following trajectory of the same length is then used to form a new trajectory C. The longer lane-changing trajectory remains unchanged and is recorded as trajectory D.

[0076] Step 3: Calculate the length error rate P between the simulated lane-changing trajectory and the actual lane-changing trajectory L , When P L Greater than the critical value P L_th , choose to complete the lane-changing trajectory from the time perspective and go to step 4; otherwise, choose to complete the lane-changing trajectory from the space perspective and go to step 6; where the critical value P L_th The default value is 20%;

[0077] Step 4: Determine the lane-changing duration between the simulated lane-changing trajectory and the actual lane-changing trajectory, and the absolute difference T dif =|T sim -T real |, when T dif Greater than the critical value T T_th When , the completion method is to complete the start and end points in half from the time perspective, and go to step 8; otherwise, go to step 5; where the critical value T T_th The default value is 2s;

[0078] Step 5: Calculate the time index of the simulated lane-changing trajectory and the actual lane-changing trajectory segment statistics when When it is greater than 0, the completion method is to complete the starting point from the time perspective, and go to step 8; otherwise, the completion method is to complete the end point from the time perspective, and go to step 8;

[0079] Step 6: Calculate the absolute error S of the speed standard deviation between the simulated lane-changing trajectory and the actual lane-changing trajectory V , which is calculated as:

[0080] S V =|S sim -S real |

[0081]

[0082]

[0083]

[0084]

[0085] Where S sim , S real are the speed standard deviations of the target vehicle in the entire simulated lane changing process and the entire real lane changing process, respectively. sim (i) v real(i) is the instantaneous speed of the target vehicle in the entire simulated lane changing process and the entire real lane changing process in the i-th unit time interval, N sim 、N real are the unit time intervals of the entire simulated lane changing process and the entire real lane changing process respectively.

[0086] When S V Greater than the critical value S V_th When , the completion method is to complete the starting and ending points in half from a spatial perspective, and go to step 9; otherwise, go to step 7; where the critical value S V_th The default value is 10km / s;

[0087] Step 7: Calculate the speed standard deviation index of the simulated lane-changing trajectory and the actual lane-changing trajectory The calculation method is

[0088]

[0089]

[0090]

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097] In the formula are the speed standard deviations of the target vehicle in the current lane and the target lane during the simulated lane change process, are the speed standard deviations of the target vehicle in the current lane and the target lane during the actual lane change process, are the instantaneous speed of the target vehicle in the simulated lane changing process and the real lane changing process in the i-th unit time interval, respectively. are the instantaneous speed of the target vehicle in the i-th unit time interval during the simulated lane changing process and the real lane changing process in the target lane, are the unit time intervals in the simulated lane change process of the current lane and the target lane, are the unit time intervals in the actual lane changing process in the current lane and the target lane, respectively.

[0098] when When it is greater than 0, the completion method is to complete the starting point from a spatial perspective, and go to step 9; otherwise, the completion method is to complete the end point from a spatial perspective, and go to step 9;

[0099] Step 8: Update the simulated lane-changing trajectory and the real lane-changing trajectory, and calculate the vehicle simulation trajectory fitting degree J from the time perspective T , which is calculated as:

[0100] T max =max{T real ,T sim}

[0101]

[0102]

[0103]

[0104] In the formula is the maximum number of parts that the lane-changing trajectory is divided into according to the time angle after the time length is unified. is the speed of the adjusted simulated lane-changing trajectory at time (t0+i·Δt), is the speed of the adjusted real lane-changing trajectory at the time (t0+i·Δt), α T is the time integrity factor.

[0105] Step 9: Update the simulated lane-changing trajectory and the real lane-changing trajectory, and calculate the vehicle simulation trajectory fitting degree J from a spatial perspective L , which is calculated as:

[0106] L max =max{L real ,L sim}

[0107]

[0108]

[0109]

[0110] In the formula is the maximum number of equal parts of the lane-changing trajectory divided by spatial angle after the time length is unified, is the speed of the adjusted simulated lane-changing trajectory at the position (t0+i·Δd), is the speed of the adjusted real lane-changing trajectory at the position (t0+i·Δd), α L is the spatial integrity coefficient.

[0111] In one embodiment, Figure 3 The actual lane-changing trajectory and the simulated lane-changing trajectory of the vehicle are shown. It is determined that the simulated lane-changing trajectory is a shorter trajectory. After the calculations in steps 3, 4, and 5, the end point of the simulated lane-changing trajectory is padded from a time perspective.

[0112] According to step 8, the fitting degree of the vehicle simulation trajectory is calculated from the time perspective, and the fitting degree is 81.54%. The average speeds of the vehicle simulation lane change trajectory and the real lane change trajectory are 82.1km / h and 87.6km / h respectively. The deviation between the two is That is, the degree of fit is 1-6.28%=93.72%.

[0113] It can be seen from this that evaluating the trajectory fitting degree with a single indicator and only considering the overall average value will ignore the process and result in an excessively high fitting degree. However, based on the idea of ​​information completion, considering the lane changing trajectory and then statistically analyzing the deviation between the simulated trajectory and the real trajectory during the whole process can better reflect its fitting degree.

[0114] In one embodiment, a vehicle simulation trajectory fitting degree calculation device based on information completion is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned vehicle simulation trajectory fitting degree calculation method based on information completion when executing the computer program.

[0115] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned vehicle simulation trajectory fitting degree calculation method based on information completion are implemented.

[0116] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0117] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A vehicle simulation trajectory fitting degree calculation method based on information completion, characterized in that: The steps include: Step 1: Collect road traffic micro information; Step 2: Calculate the length error rate between the simulated lane-changing trajectory and the actual lane-changing trajectory. When the length error rate is greater than the set first critical value P L_th , choose to complete the lane-changing trajectory from a temporal perspective and go to step 3; otherwise, choose to complete the lane-changing trajectory from a spatial perspective and go to step 5; Step 3: Calculate the absolute difference between the lane-changing duration of the simulated lane-changing trajectory and the actual lane-changing trajectory. When the absolute difference is greater than the set second threshold value T T_th When , the lane change trajectory is completed by completing the starting and ending points in half from the time perspective, and then go to step 7; otherwise, go to step 4; Step 4: Calculate the time index of the simulated lane-changing trajectory and the real lane-changing trajectory by segment. When the time index is greater than 0, the lane-changing trajectory is completed by completing the starting point from the time perspective, and go to step 7; otherwise, the lane-changing trajectory is completed by completing the end point from the time perspective, and go to step 7; Step 5: Calculate the absolute error of the simulated lane-changing trajectory and the actual lane-changing trajectory speed standard deviation. When the absolute error is greater than the set third critical value S V_th When , the lane change trajectory is completed by completing the starting and ending points in half from a spatial perspective, and then go to step 8; otherwise, go to step 6; Step 6: Calculate the speed standard deviation index of the simulated lane-changing trajectory and the actual lane-changing trajectory. When the speed standard deviation index is greater than 0, the lane-changing trajectory is completed by completing the starting point from a spatial perspective, and go to step 8; otherwise, the lane-changing trajectory is completed by completing the end point from a spatial perspective, and go to step 8; Step 7: Calculate the vehicle simulation trajectory fitting degree from the time perspective for the completed simulated lane-changing trajectory and the real lane-changing trajectory, and the step ends; Step 8: Calculate the vehicle simulation trajectory fitting degree from a spatial perspective for the completed simulated lane-changing trajectory and the actual lane-changing trajectory, and the step ends; In step 5, the absolute error S of the speed standard deviation between the simulated lane-changing trajectory and the actual lane-changing trajectory is V The calculation method is: S V =|S sim -S real | Where S sim , S real are the speed standard deviations of the target vehicle in the entire simulated lane changing process and the entire real lane changing process, respectively. sim (i) v real (i) is the instantaneous speed of the target vehicle in the entire simulated lane changing process and the entire real lane changing process in the i-th unit time interval, N sim 、N real are the unit time intervals of the entire simulated lane-changing process and the entire real lane-changing process respectively; In step 6, the speed standard deviation index of the simulated lane-changing trajectory and the real lane-changing trajectory is calculated by segmentation The calculation method is In the formula are the speed standard deviations of the target vehicle in the current lane and the target lane during the simulated lane change process, are the speed standard deviations of the target vehicle in the current lane and the target lane during the actual lane change process, are the instantaneous speed of the target vehicle in the simulated lane changing process and the real lane changing process in the i-th unit time interval, respectively. are the instantaneous speed of the target vehicle in the i-th unit time interval during the simulated lane changing process and the real lane changing process in the target lane, are the unit time intervals in the simulated lane change process of the current lane and the target lane, are the number of unit time intervals in the actual lane-changing process between the current lane and the target lane, respectively; In step 7, the vehicle simulation trajectory fitting degree J is calculated from the time perspective T The calculation method is: T max =max{T real ,T sim } In the formula is the maximum number of parts that the lane-changing trajectory is divided into according to the time angle after the time length is unified. is the speed of the completed simulated lane-changing trajectory at time (t0+i·Δt), is the speed of the completed lane-changing trajectory at the time (t0+i·Δt), t0 is the starting time of the completed lane-changing trajectory, α T is the temporal integrity coefficient; In step 8, the vehicle simulation trajectory fitting degree J is calculated from the spatial perspective L , which is calculated as: L max =max{L real ,L sim } In the formula is the maximum number of equal parts of the lane-changing trajectory divided by spatial angle after the time length is unified, is the speed of the completed simulated lane-changing trajectory at the position (d0+i·Δd), is the speed of the completed real lane-changing trajectory at the position (d0+i·Δd), d0 is the starting position of the completed lane-changing trajectory, α L is the spatial integrity coefficient.

2. The vehicle simulation trajectory fitting degree calculation method based on information completion according to claim 1 is characterized in that: In step 1, the road traffic microscopic information includes the speed of the target vehicle recorded according to the unit time scale Δt and the unit space scale Δd; the simulated lane change trajectory length L of the target vehicle sim and the actual lane-changing trajectory length L real ; The simulated lane change duration of the target vehicle is T sim and the actual lane change duration T real ; Duration of the simulated lane-changing trajectory of the target vehicle in the current lane and the target lane and The actual lane-changing trajectory of the target vehicle lasts for a certain period of time in the current lane and the target lane. and The average speed of the target vehicle during the entire simulated lane change process and the entire real lane change process and Average speed of the target vehicle during the simulated lane change process and the actual lane change process in the current lane and Average speed of the target vehicle during the simulated lane change process and the actual lane change process in the target lane and 3. The vehicle simulation trajectory fitting degree calculation method based on information completion according to claim 2 is characterized in that: In step 2, the simulated lane-changing trajectory is recorded as trajectory A, the actual lane-changing trajectory is recorded as trajectory B, and the absolute difference between the lane-changing durations of trajectory A and trajectory B is T dif =|T sim -T real |, the difference in lane change trajectory length between trajectory A and trajectory B is L dif =|L sim -L real |; The lane change trajectory completion methods include the following six categories: From the perspective of time, half of the starting and ending points are supplemented: 0.5T is added at the starting and ending points of the track with a shorter lane change duration. dif The following trajectory with longer duration, the trajectory with longer lane change duration remains unchanged; Completing the starting point from the perspective of time: Add T at the starting point of the trajectory with a shorter lane change duration dif The following trajectory with longer duration, the trajectory with longer lane change duration remains unchanged; Completing the endpoint from a time perspective: Add T at the end of the trajectory with a shorter lane change duration dif The following trajectory with longer duration, the trajectory with longer lane change duration remains unchanged; From a spatial perspective, half of the starting and ending points are supplemented: 0.5L is added at the starting and ending points of the shorter lane change trajectory dif The following trajectory with a longer length remains unchanged; Completing the starting point from a spatial perspective: Add L at the starting point of the shorter lane change trajectory. dif The following trajectory with a longer length remains unchanged; Completing the end point from a spatial perspective: Add L at the end point of the shorter lane change trajectory. dif The following trajectory with a longer length remains unchanged.

4. The method for calculating the vehicle simulation trajectory fitting degree based on information completion according to claim 3 is characterized in that: In step 4, the time index of the simulated lane-changing trajectory and the real lane-changing trajectory segmented statistics The calculation method is 5. A vehicle simulation trajectory fitting degree calculation device based on information completion, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the vehicle simulation trajectory fitting degree calculation method based on information completion described in any one of claims 1 to 4 are implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the vehicle simulation trajectory fitting degree calculation method based on information completion described in any one of claims 1 to 4 are implemented.

Images