Driving behavior detection method and device and medium

By performing multi-level detection area division and parameter comparison of driving data, the subjectivity problem of traditional driving behavior analysis is solved, more accurate and objective driving behavior evaluation is achieved, the application capabilities of traffic management and vehicle control systems are enhanced, and the application capabilities of traffic management and vehicle control systems are reduced. Safety hazards and efficiency losses.

CN118977720APending Publication Date: 2024-11-19WEICHAI POWER CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411152818.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

Traditional driving behavior analysis relies on subjective feelings, resulting in large subjectivity and differences in evaluation results, making it difficult to form an objective and comparable evaluation system, limiting its application in traffic management and safety assessment, and being difficult to match modern vehicles. The control system is effectively connected, resulting in potential safety hazards and efficiency losses that are difficult to detect and adjust in a timely manner.

Method used

By obtaining the driving data of the vehicle to be tested, multi-level detection areas are divided based on the driving data and the preset vehicle speed step length, acceleration detection parameters and emergency acceleration parameters are determined, combined with the initial detection curve chart and acceleration curve to comprehensively judge driving behavior. Reduce the impact of subjectivity and improve the accuracy and objectivity of detection.

Benefits of technology

It realizes a more refined analysis of driving behavior, can objectively and accurately reflect the driver's driving behavior characteristics, reduce misjudgment and misjudgment, improves the accuracy and safety of driving behavior evaluation, and enhances the vehicle control system Guidance ability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118977720A_ABST
    Figure CN118977720A_ABST
Patent Text Reader

Abstract

The embodiment of the invention discloses a driving behavior detection method and device and a medium, belongs to the technical field of driving data processing, and solves the problem that potential safety hazards and efficiency loss are difficult to find and adjust in time due to the fact that subjective feeling is difficult to directly guide a vehicle control system. Acquiring driving data of the to-be-tested vehicle; based on the driving data and a preset vehicle speed step length, dividing a multi-stage detection area; determining an acceleration detection parameter and a rapid acceleration parameter respectively corresponding to each stage of detection area; comparing the acceleration detection parameter corresponding to each level of detection area with a preset driving behavior initial detection curve graph to obtain a first driving behavior detection result; an acceleration curve corresponding to each level of detection area is drawn based on the driving data, and the rapid acceleration parameters are compared with the acceleration curves to obtain a second driving behavior detection result; and determining a final driving behavior detection result according to the first driving behavior detection result and the second driving behavior detection result.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of driving data processing, and in particular to a driving behavior detection method, device and medium. Background Art

[0002] In the modern transportation system, driving behavior is the link between vehicles, road environment and traffic safety, and its importance is self-evident. It is not only directly related to the life safety of road users, but also has a profound impact on fuel economy, vehicle emission levels and overall traffic flow. With the growing demand for energy conservation and emission reduction, green travel, and the vigorous development of intelligent driving technology, in-depth analysis and optimization of driving behavior has become an important research direction in the field of transportation science and technology.

[0003] However, traditional driving behavior analysis relies heavily on people's subjective feelings and experience, which makes the analysis results highly subjective and inconsistent. Driving behavior detection based on subjective feelings is often difficult to form an objective and comparable evaluation system, which limits its application in traffic management, safety assessment, etc.

[0004] Secondly, traditional subjective evaluation methods are difficult to effectively connect with modern vehicle control systems. With the continuous improvement of automobile electronics and intelligence, the demand for real-time feedback and optimization of driving behavior by vehicle control systems is becoming increasingly urgent. However, since subjective feelings are difficult to directly use to guide the adjustment and optimization of vehicle control systems, many potential safety hazards and efficiency losses are difficult to be discovered and adjusted in a timely manner. Summary of the invention

[0005] The embodiments of the present application provide a driving behavior detection method, device and medium for solving the following technical problems: Due to subjective feelings, it is difficult to directly use them to guide the adjustment and optimization of the vehicle control system, resulting in many potential safety hazards and efficiency losses being difficult to be discovered and adjusted in a timely manner.

[0006] The present application embodiment adopts the following technical solutions:

[0007] The embodiment of the present application provides a driving behavior detection method, which includes obtaining driving data of a vehicle to be tested; dividing a multi-level detection area based on the driving data and a preset vehicle speed step; determining the acceleration detection parameters and the sudden acceleration parameters corresponding to each level of the detection area; comparing the acceleration detection parameters corresponding to each level of the detection area with a preset driving behavior initial detection curve diagram to obtain a first driving behavior detection result; drawing an acceleration curve corresponding to each level of the detection area based on the driving data, comparing the sudden acceleration parameter with the acceleration curve, and obtaining a second driving behavior detection result; and determining the final driving behavior detection result according to the first driving behavior detection result and the second driving behavior detection result.

[0008] The embodiment of the present application divides the multi-level detection area by driving data and vehicle speed step length, and performs more refined driving behavior analysis for different speed intervals to improve the accuracy of detection. Using acceleration and sudden acceleration parameters as quantitative indicators can objectively and accurately reflect the driving behavior characteristics of the driver, achieve a unified evaluation of the driving behavior of different drivers, and reduce subjective influences. Combining the two test results, it is possible to more accurately judge whether the driver's driving behavior is standardized and safe, reducing the possibility of misjudgment and missed judgment.

[0009] In one implementation of the present application, the driving data of the vehicle to be tested is obtained, specifically including: obtaining real-time driving reference data corresponding to the vehicle to be tested; based on a preset driving data threshold, determining first abnormal data in the real-time driving reference data, and screening out the first abnormal data; based on time continuity and a preset continuous data difference threshold, determining second abnormal data in the real-time driving reference data, and performing difference processing on the second abnormal data; based on the data after screening and difference processing, obtaining the driving data of the vehicle to be tested.

[0010] In one implementation of the present application, multi-level detection areas are divided based on driving data and preset vehicle speed steps, specifically including: determining an overall detection area based on the full driving area in the driving data; comparing the driving position in the driving data with a preset working condition traffic area map to determine multiple working condition detection areas; wherein the working condition detection area includes at least an urban area, a suburban area and a highway area; based on the preset vehicle speed step, the vehicle speed in the driving data is divided to determine multiple segment detection areas; wherein the difference between the initial vehicle speed and the terminal vehicle speed corresponding to a segment detection area is a preset vehicle speed step.

[0011] In one implementation of the present application, the acceleration detection parameters corresponding to each level of detection area are determined, specifically including: constructing a first speed parameter set based on the product of the real-time vehicle speed and the real-time acceleration corresponding to the overall detection area, sorting multiple speed parameters in the first speed parameter set, and selecting the speed parameter corresponding to the preset percentile value as the acceleration detection parameter of the overall area; constructing a second speed parameter set corresponding to multiple working condition detection areas based on the product of the real-time vehicle speed and the real-time acceleration corresponding to multiple working condition detection areas, sorting multiple speed parameters in each second speed parameter set, and selecting the speed parameter corresponding to the preset percentile value as the acceleration detection parameter corresponding to each working condition detection area; constructing a third speed parameter set corresponding to multiple fragment detection areas based on the product of the real-time vehicle speed and the real-time acceleration corresponding to multiple fragment detection areas, sorting multiple speed parameters in each third speed parameter set, and selecting the speed parameter corresponding to the preset percentile value as the acceleration detection parameter corresponding to each fragment detection area.

[0012] In one implementation of the present application, the acceleration detection parameters corresponding to each level of detection area are compared with a preset driving behavior initial detection curve diagram to obtain a first driving behavior detection result, which specifically includes: determining the average vehicle speed corresponding to each level of detection area, and determining the horizontal coordinate value in the initial detection curve diagram based on the average vehicle speed; determining the vertical coordinate value in the initial detection curve diagram based on the acceleration detection parameters corresponding to each level of detection area; determining the calibration points in the initial detection curve diagram based on the horizontal coordinate value and the vertical coordinate value, and calibrating them through different markers; comparing the calibration points with multiple reference curves in the initial detection curve diagram, and obtaining the first driving behavior detection result based on the positional relationship between the calibration points and the reference curves; wherein the multiple reference curves correspond to different driving behaviors, respectively.

[0013] In one implementation of the present application, after obtaining the first driving behavior detection result, the method also includes: obtaining the latest driving behavior detection data; re-fitting the initial detection curve graph based on the latest driving behavior detection data; and determining multiple new reference curves based on the fitting results to optimize the initial detection curve graph.

[0014] In one implementation of the present application, an acceleration curve corresponding to each level of detection area is drawn based on the driving data, and the sudden acceleration parameter is compared with the acceleration curve to obtain a second driving behavior detection result, which specifically includes: determining the driving time corresponding to each level of detection area; constructing the acceleration curve corresponding to each level of detection area based on the real-time acceleration and driving time in the driving data; determining a sudden acceleration parameter point in the acceleration curve whose absolute value is greater than a preset acceleration threshold; and obtaining the second driving behavior detection result based on the proportion of the sudden acceleration parameter point in the acceleration curve.

[0015] In one implementation of the present application, a final driving behavior detection result is determined based on a first driving behavior detection result and a second driving behavior detection result, specifically including: comparing the first driving behavior detection result with the second driving behavior detection result; when the first driving behavior detection result corresponding to any detection area is greater than an upper limit reference curve, or the second driving behavior detection result is that the proportion of rapid acceleration parameter points is greater than a preset first proportion value, determining that the driving behavior corresponding to the detection area is aggressive driving; when the first driving behavior detection result corresponding to any detection area is less than the upper limit reference curve and greater than the lower limit reference curve, and the second driving behavior detection result is that the proportion of rapid acceleration parameter points is less than the preset first proportion value and greater than the second proportion value, determining that the driving behavior corresponding to the detection area is normal driving; when the first driving behavior detection result corresponding to any detection area is less than the lower limit reference curve, or the second driving behavior detection result is that the proportion of rapid acceleration parameter points is less than the preset second proportion value, determining that the driving behavior corresponding to the detection area is gentle driving.

[0016] An embodiment of the present application provides a driving behavior detection device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can: obtain driving data of a vehicle to be tested; divide a multi-level detection area based on the driving data and a preset vehicle speed step; determine acceleration detection parameters and sudden acceleration parameters corresponding to each level of the detection area; compare the acceleration detection parameters corresponding to each level of the detection area with a preset driving behavior initial detection curve diagram to obtain a first driving behavior detection result; draw an acceleration curve corresponding to each level of the detection area based on the driving data, and compare the sudden acceleration parameter with the acceleration curve to obtain a second driving behavior detection result; and determine a final driving behavior detection result based on the first driving behavior detection result and the second driving behavior detection result.

[0017] A non-volatile computer storage medium provided in an embodiment of the present application stores computer executable instructions, wherein the computer executable instructions are configured to: obtain driving data of a vehicle to be tested; divide a multi-level detection area based on the driving data and a preset vehicle speed step; determine acceleration detection parameters and sudden acceleration parameters corresponding to each level of detection area; compare the acceleration detection parameters corresponding to each level of detection area with a preset driving behavior initial detection curve diagram to obtain a first driving behavior detection result; draw an acceleration curve corresponding to each level of detection area based on the driving data, compare the sudden acceleration parameter with the acceleration curve, and obtain a second driving behavior detection result; and determine a final driving behavior detection result based on the first driving behavior detection result and the second driving behavior detection result.

[0018] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects: The embodiments of the present application divide the multi-level detection area by driving data and vehicle speed step length, and conduct more refined driving behavior analysis for different speed intervals to improve the accuracy of detection. Using acceleration and sudden acceleration parameters as quantitative indicators can objectively and accurately reflect the driving behavior characteristics of the driver, realize unified evaluation of the driving behavior of different drivers, and reduce subjective influence. Combining the two detection results, it is possible to more accurately judge whether the driver's driving behavior is standardized and safe, and reduce the possibility of misjudgment and missed judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. In the drawings:

[0020] Figure 1 A flow chart of a driving behavior detection method provided in an embodiment of the present application;

[0021] Figure 2 A schematic diagram of an initial curve of driving behavior provided in an embodiment of the present application;

[0022] Figure 3 A schematic diagram of a first driving behavior detection result corresponding to a multi-level detection area provided in an embodiment of the present application;

[0023] Figure 4 A schematic diagram of a driving behavior detection control device provided in an embodiment of the present application;

[0024] Figure 5 A schematic diagram of the structure of a driving behavior detection device provided in an embodiment of the present application.

[0025] Reference numerals:

[0026] 1: parameter setting module, 2: evaluation module, 3: signal receiving module, 4: signal processing module, 5: calculation and analysis module, 6: signal output module, 7: self-learning server, 8: subjective evaluation module; 200: driving behavior detection device, 201: processor, 202: memory. DETAILED DESCRIPTION

[0027] The embodiments of the present application provide a driving behavior detection method, device and medium.

[0028] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0029] The technical solution proposed in the embodiment of the present invention is described in detail below with reference to the accompanying drawings.

[0030] Figure 1 A flow chart of a driving behavior detection method provided in an embodiment of the present application is as follows: Figure 1 As shown, the driving behavior detection method includes the following steps:

[0031] S101, obtaining driving data of a vehicle to be tested.

[0032] In one embodiment of the present application, an initial detection curve graph is first constructed, specifically:

[0033] Initial parameter settings:

[0034] The initial parameters in the embodiments of the present application are set values, and can be self-learning optimized in the application, including:

[0035] (1) Speed ​​setting: The speed points for turning from the city to the suburbs and from the suburbs to the highway can be set. For example, 40 km / h and 60 km / h can be selected as turning points.

[0036] (2) Acceleration setting: including acceleration and deceleration, and rapid acceleration and deceleration settings. For example, according to common experience, the acceleration can be set to > 0.1 m / s 2 , deceleration is <-0.1m / s 2 , rapid acceleration is>1.55m / s 2 , rapid deceleration is <-1.65m / s 2 .

[0037] (3) Driving behavior evaluation curve: It mainly gives the initial curve for the normal value and upper and lower limits of v×a. Figure 2 A schematic diagram of an initial curve of driving behavior provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the horizontal axis is the vehicle speed, and the vertical axis is the product of v×a. The figure contains three curves, namely the upper limit reference curve, the conventional reference curve and the lower limit reference curve. The formula corresponding to each reference curve is as follows:

[0038] Upper limit reference curve formula: y = -0.0025 × v × v + 0.4509 × v - 3.5729;

[0039] Conventional reference value curve formula: y = -0.0036 × v × v + 0.493 × v - 4.8739;

[0040] The formula of the lower limit reference curve is: y = -0.0032 × v × v + 0.4198 × v - 4.7298;

[0041] Each reference curve is obtained by fitting multiple sets of historical driving behavior detection data within a certain period. Based on the fitted initial driving behavior curve diagram, the obtained driving data of the vehicle to be tested can be compared and analyzed to determine the corresponding driving behavior detection results.

[0042] In one embodiment of the present application, real-time driving reference data corresponding to the vehicle to be tested is obtained. Based on a preset driving data threshold, first abnormal data is determined in the real-time driving reference data, and the first abnormal data is screened out. Based on time continuity and a preset continuous data difference threshold, second abnormal data is determined in the real-time driving reference data, and difference processing is performed on the second abnormal data. Based on the data after screening and difference processing, the driving data of the vehicle to be tested is obtained.

[0043] Specifically, after obtaining the driving data of the vehicle to be tested, the embodiment of the present application records and analyzes the vehicle driving data and processes the abnormal data. For example, interpolation processing can be performed for the case where the vehicle speed is discontinuous due to the signal-affected section such as the tunnel. In addition, >5m / s 2 and<-5m / s 2 The acceleration is eliminated.

[0044] Furthermore, the real-time driving reference data in the embodiments of the present application can usually be obtained through a variety of channels, including but not limited to navigation applications, traffic websites, and Internet of Things devices. The obtained real-time driving reference data usually includes vehicle speed, acceleration, location, driving time, altitude, and vehicle emission coefficient.

[0045] Furthermore, the real-time driving reference data is compared with the preset driving data threshold, and the data exceeding the threshold range is the first abnormal data. For example, on the highway, the maximum speed of a private car cannot exceed 120km / h, so the data above this speed should be regarded as abnormal. By comparing the difference between adjacent data points with the preset continuous data difference threshold, the data whose difference exceeds the threshold is the second abnormal data. For example, when a vehicle travels to a signal-free area such as a tunnel, the speed changes sharply in a short period of time, exceeding the reasonable acceleration range, which is regarded as the second abnormal data. For the second abnormal data, the difference processing method can be used for adjustment. Based on the data after screening and difference processing, the driving data of the vehicle to be tested is obtained, and the driving data obtained is more accurate and reliable.

[0046] S102: Divide the vehicle into multiple detection areas based on the driving data and the preset vehicle speed step.

[0047] In one embodiment of the present application, the overall detection area is determined based on the entire driving area in the driving data. The driving position in the driving data is compared with the preset working condition traffic area map to determine multiple working condition detection areas; wherein the working condition detection area at least includes an urban area, a suburban area, and a highway area. Based on the preset vehicle speed step, the vehicle speed in the driving data is divided to determine multiple segment detection areas; wherein the difference between the initial vehicle speed and the terminal vehicle speed corresponding to a segment detection area is a preset vehicle speed step.

[0048] Specifically, all the position information of the vehicle to be tested during the entire driving process is extracted from the driving data, and these position information constitutes the full driving area. Based on the full driving area, it is regarded as an overall detection range. This range includes all the driving data that need to be detected.

[0049] Furthermore, the driving position in the driving data, that is, the real-time position information of the vehicle, is compared with the preset working condition traffic area map, and the driving position is mapped to the working condition traffic area map for comparison. The preset working condition traffic area map in the embodiment of the present application is a pre-drawn map with different types of traffic areas marked on it, such as urban areas, suburban areas, highway areas, etc. According to the comparison results, multiple working condition detection areas passed by the vehicle during driving can be determined. The working condition detection areas in the embodiment of the present application include urban areas, suburban areas and highway areas. For example, the vehicle may first drive in the urban area, then enter the suburban area, and finally enter the highway area.

[0050] Furthermore, based on the preset speed step, the vehicle speed in the driving data is divided. Specifically, the speed value at each time point is extracted from the driving data, and then grouped according to the speed step. The speed value in each group fluctuates within a specific speed range. Each speed group corresponds to a fragment detection area. The division of the fragment detection area is determined based on the continuity and change of the vehicle speed. For example, if the preset speed step is 20km / h, all speed data from the initial speed (such as 60km / h) to the terminal speed (such as 80km / h) will be divided into the same fragment detection area. The difference between the initial speed and the terminal speed of this area is exactly a preset speed step.

[0051] S103, determining acceleration detection parameters and rapid acceleration parameters corresponding to each level of detection area.

[0052] In one embodiment of the present application, based on the product of the real-time vehicle speed and the real-time acceleration corresponding to the overall detection area, a first speed parameter set is constructed, multiple speed parameters in the first speed parameter set are sorted, and the speed parameter corresponding to the preset percentile value is selected as the acceleration detection parameter of the overall area. Based on the product of the real-time vehicle speed and the real-time acceleration corresponding to multiple working condition detection areas, a second speed parameter set corresponding to multiple working condition detection areas is constructed, multiple speed parameters in each second speed parameter set are sorted, and the speed parameter corresponding to the preset percentile value is selected as the acceleration detection parameter corresponding to each working condition detection area. Based on the product of the real-time vehicle speed and the real-time acceleration corresponding to multiple segment detection areas, a third speed parameter set corresponding to multiple segment detection areas is constructed, multiple speed parameters in each third speed parameter set are sorted, and the speed parameter corresponding to the preset percentile value is selected as the acceleration detection parameter corresponding to each segment detection area.

[0053] Specifically, the embodiment of the present application calculates the product of vehicle speed and acceleration according to the following formula:

[0054] (v×a) i =v i ×a i / 3.6

[0055] Where:

[0056] v is the real-time vehicle speed, a is the real-time acceleration, (v×a) i ——The product of the actual vehicle speed and acceleration in time step i, in m 2 / s 3 .

[0057] The average vehicle speed in each area is taken as the horizontal coordinate value in the preset driving behavior initial detection curve. There are multiple products of real-time vehicle speed and real-time acceleration in the area, and the multiple products are sorted, and the percentile value of the preset percentile value number corresponding to (v×a) in the area is taken as the vertical coordinate value in the preset driving behavior initial detection curve to perform driving behavior comparison analysis.

[0058] For example, the embodiment of the present application may preferentially select the 95th percentile value of (v×a), or may change it as needed, and the embodiment of the present application is not limited to this.

[0059] Furthermore, when determining the acceleration detection parameter corresponding to the overall detection area, the real-time vehicle speed and real-time acceleration are extracted from the driving data corresponding to the overall detection area. Then, the product of these two parameters is calculated, and the product value at each time point is used as a speed parameter, which is sorted in order to construct a first speed parameter set. According to the preset percentile value, the corresponding speed parameter is selected from the sorted speed parameter set as the acceleration detection parameter. If it is the "95th percentile", the maximum value of the speed parameters in the top 95% after sorting is selected.

[0060] Furthermore, when determining the acceleration detection parameters corresponding to the multiple working condition detection areas, for each working condition detection area, such as urban area, suburbs, and highways, the product of the real-time vehicle speed and the real-time acceleration in the corresponding driving data is calculated. The product value at the time point of each working condition area is used as a speed parameter, sorted in order, and constructed into a separate second speed parameter set. According to the preset percentile value, the corresponding speed parameter is selected from the sorted speed parameter set as the acceleration detection parameter. If it is the "95th percentile", the maximum value of the speed parameters in the top 95% after sorting is selected.

[0061] Furthermore, when determining the acceleration detection parameters corresponding to the multiple segment detection areas, for each segment detection area, the product of the real-time vehicle speed and the real-time acceleration in the corresponding driving data is calculated. The product value at the time point of each segment area is used as a speed parameter, sorted in order, and constructed into a separate third speed parameter set. According to the preset percentile value, the corresponding speed parameter is selected from the sorted speed parameter set as the acceleration detection parameter. If it is the "95th percentile", the maximum value of the speed parameters in the top 95% after sorting is selected.

[0062] In one embodiment of the present application, the rapid acceleration parameters corresponding to each level of detection area are determined based on the acquired driving data of the vehicle to be tested. For example, the rapid acceleration in the embodiment of the present application can be set to >1.55m / s 2, the rapid deceleration can be set to <-1.65m / s 2 .

[0063] S104: Compare the acceleration detection parameters corresponding to each level of detection area with a preset driving behavior initial detection curve diagram to obtain a first driving behavior detection result.

[0064] In one embodiment of the present application, the average vehicle speed corresponding to each level of detection area is determined, and the horizontal coordinate value is determined in the initial detection curve diagram based on the average vehicle speed. The vertical coordinate value is determined in the initial detection curve diagram based on the acceleration detection parameters corresponding to each level of detection area. Based on the horizontal coordinate value and the vertical coordinate value, the calibration point is determined in the initial detection curve diagram, and calibrated by different markers. The calibration point is compared with multiple reference curves in the initial detection curve diagram, and based on the positional relationship between the calibration point and the reference curve, a first driving behavior detection result is obtained; wherein the multiple reference curves correspond to different driving behaviors.

[0065] Specifically, according to the driving data corresponding to each level of detection area, the average vehicle speed corresponding to each level of detection area is determined. According to the determined average vehicle speed, the corresponding horizontal coordinate point can be determined in the initial detection curve diagram. Secondly, according to the acceleration detection parameters corresponding to each level of detection area, the vertical coordinate value corresponding to each level of detection area is determined. The corresponding calibration point can be obtained through the determined horizontal coordinate point and vertical coordinate value.

[0066] Furthermore, in order to distinguish the driving behavior detection results corresponding to different detection areas, the embodiments of the present application use different colors, shapes or symbols to mark these points. For example, the overall detection area corresponds to square calibration points, the working condition detection area corresponds to circular calibration points, and the fragment detection area corresponds to triangular calibration points.

[0067] Furthermore, the calibration point is compared with the predefined reference curve in the initial detection curve diagram. These reference curves represent different driving behaviors. For example, if the calibration point is above the upper limit reference curve, the first driving behavior detection result of the detection area corresponding to the calibration point is aggressive driving. If the calibration point is near the conventional reference curve, the first driving behavior detection result of the detection area corresponding to the calibration point is normal driving. If the calibration point is below the lower limit reference curve, the first driving behavior detection result of the detection area corresponding to the calibration point is gentle driving.

[0068] In one embodiment of the present application, the latest driving behavior detection data is obtained, and the initial detection curve graph is re-fitted based on the latest driving behavior detection data. Multiple new reference curves are determined based on the fitting results to optimize the initial detection curve graph.

[0069] Specifically, the latest collected driving behavior detection data is cleaned to remove outliers, missing values ​​or erroneous data to ensure the accuracy and reliability of the data. The initial detection curve graph is re-fitted according to the newly acquired data and historical data. On the initial detection curve graph, multiple new reference curves are drawn according to the determined threshold or range. These curves should be able to clearly reflect the characteristics of different driving behaviors. The new reference curve is added to the initial detection curve graph. Through curve optimization, the initial detection curve graph can more clearly reflect the characteristics of different driving behaviors, making driving behavior detection more accurate.

[0070] Figure 3 A schematic diagram of a first driving behavior detection result corresponding to a multi-level detection area provided in an embodiment of the present application, such as Figure 3 As shown, the overall detection area corresponds to square calibration points, the working condition detection area corresponds to circular calibration points, and the fragment detection area corresponds to triangular calibration points. The top curve is the upper limit reference curve, the middle curve is the regular reference curve, and the bottom curve is the lower limit reference curve. If the square calibration point is between the upper limit reference curve and the lower limit reference curve, it means that the first driving behavior detection result corresponding to the overall detection area is normal driving. Two circular calibration points fall between the upper limit reference curve and the lower limit reference curve, and one circular calibration point falls above the upper limit reference curve, which means that in the working condition detection area, there are two detection areas for normal driving and one detection area for intense driving. The fragment detection area corresponds to the triangular calibration points. According to the position of the triangular calibration points, it can be seen that multiple fragment detection areas correspond to intense driving, normal driving, and gentle driving, respectively.

[0071] S105 . Draw acceleration curves corresponding to each level of detection area based on the driving data, compare the sudden acceleration parameter with the acceleration curve, and obtain a second driving behavior detection result.

[0072] In one embodiment of the present application, the driving time corresponding to each level of detection area is determined. Based on the real-time acceleration and driving time in the driving data, an acceleration curve corresponding to each level of detection area is constructed. A sudden acceleration parameter point whose absolute value is greater than a preset acceleration threshold is determined in the acceleration curve. Based on the proportion of the sudden acceleration parameter point in the acceleration curve, a second driving behavior detection result is obtained.

[0073] Specifically, in the embodiment of the present application, for each level of detection area, the driving time required for the vehicle to pass through the area is determined. Based on the real-time acceleration and driving time in the driving data, an acceleration curve is constructed for each level of detection area. The acceleration curve is a graphical representation of the acceleration changing over time, which can intuitively show the acceleration behavior of the vehicle in the area. In the acceleration curve, the point whose absolute value is greater than the preset acceleration threshold is determined, that is, the rapid acceleration parameter point. For example, in the embodiment of the present application, the rapid acceleration can be set to >1.55m / s 2 , the rapid deceleration can be set to <-1.65m / s 2 The thresholds for rapid acceleration and rapid deceleration may also be adjusted as required.

[0074] Furthermore, the proportion of sudden acceleration parameter points in the acceleration curve is calculated by dividing the number of sudden acceleration parameter points by the total number of time points. This proportion reflects the frequency of sudden acceleration behavior of the vehicle in the detection area of ​​this level. Based on the proportion of sudden acceleration parameter points, a second driving behavior detection result can be obtained. For example, it can be set that the proportion in the detection area exceeds 1%, and the second driving behavior detection result corresponding to the detection area is intense driving. The proportion value in the embodiment of the present application can be adjusted according to demand, and the embodiment of the present application is not limited to this.

[0075] S106: Determine a final driving behavior detection result according to the first driving behavior detection result and the second driving behavior detection result.

[0076] In one embodiment of the present application, the first driving behavior detection result is compared with the second driving behavior detection result. When the first driving behavior detection result corresponding to any detection area is greater than the upper limit reference curve, or the second driving behavior detection result is that the proportion of the rapid acceleration parameter point is greater than the preset first proportion value, it is determined that the driving behavior corresponding to the detection area belongs to fierce driving. When the first driving behavior detection result corresponding to any detection area is less than the upper limit reference curve and greater than the lower limit reference curve, and the second driving behavior detection result is that the proportion of the rapid acceleration parameter point is less than the preset first proportion value and greater than the second proportion value, it is determined that the driving behavior corresponding to the detection area belongs to normal driving. When the first driving behavior detection result corresponding to any detection area is less than the lower limit reference curve, or the second driving behavior detection result is that the proportion of the rapid acceleration parameter point is less than the preset second proportion value, it is determined that the driving behavior corresponding to the detection area belongs to gentle driving.

[0077] Specifically, the first driving behavior detection result is compared with the second driving behavior detection result. If, in any detection area, the first driving behavior detection result shows that the driving behavior is greater than the upper limit reference curve, or the second driving behavior detection result shows that the proportion of the rapid acceleration parameter point is greater than the preset first proportion value, then the driving behavior in the detection area is determined to be aggressive driving.

[0078] If, in any detection area, the first driving behavior detection result shows that the driving behavior is between the upper reference curve and the lower reference curve, and the second driving behavior detection result shows that the proportion of the rapid acceleration parameter points is neither greater than the preset first proportion value nor less than the preset second proportion value, it is determined to be normal driving. If, in any detection area, the first driving behavior detection result shows that the driving behavior is less than the lower reference curve, or the second driving behavior detection result shows that the proportion of the rapid acceleration parameter points is less than the preset second proportion value, it is determined to be gentle driving. Among them, the preset first proportion value is greater than the preset second proportion value.

[0079] The embodiment of the present application can improve the accuracy of driving behavior detection by taking the proportion of time of sudden acceleration as an auxiliary evaluation item.

[0080] In addition, the embodiment of the present application also obtains in real time the altitude corresponding to the driving path of the vehicle to be tested, and the emission correlation coefficient corresponding to the vehicle to be tested. In areas where the altitude changes greatly, in order to improve the accuracy of driving behavior detection, the corresponding initial driving behavior curve diagram can be set according to the altitude. For example, the initial driving behavior curve diagrams corresponding to different altitudes can be constructed based on the historical driving behavior data set. The corresponding initial driving behavior curve diagram is determined according to the altitude corresponding to the vehicle to be tested, so that a more accurate driving behavior comparison test can be performed on the vehicle to be tested. Among them, different driving behavior detection curve diagrams have different corresponding upper and lower limits.

[0081] The detection results obtained in the embodiment of the present application can be displayed on the vehicle dashboard, with normal driving displayed in green, intense driving displayed in red, and gentle driving displayed in yellow. In addition, the system can generate driving behavior analysis reports according to different time periods, and upload the final data to the database, and has self-learning, optimization and update functions. Combined with the driver's subjective evaluation and historical database, the existing benchmark data is updated within the set time, so that the curve can be more realistic and close to the optimal, thereby better guiding the driver to drive the vehicle.

[0082] Figure 4 A schematic diagram of a driving behavior detection control device provided in an embodiment of the present application is shown in FIG. Figure 4As shown, the driving behavior detection control device includes: a parameter setting module 1, which sets the initial parameters; an evaluation model 2, which sets and optimizes the driving behavior evaluation model curve; a signal receiving module 3, which receives vehicle operation data through an ECU or an external sensor, including time, speed, altitude, distance, etc.; a signal processing module 4, which cleans the signal and processes abnormal data; a calculation and analysis module 5, which compares the acceleration and v×a with the evaluation model to obtain the level of the driving behavior this time; a signal output module 6, which displays it to the driver through a dashboard or other means; and can generate an analysis report and upload the original data and result data to a database; a self-learning server 7; which accumulates historical data, combines subjective evaluation information, continuously iterates and corrects the actual results, and regularly optimizes the evaluation model; a subjective evaluation module 8; which interacts with the driver to let the driver make a subjective evaluation of each driving, and provides materials for model optimization.

[0083] Figure 5 This is a schematic diagram of the structure of a driving behavior detection device provided in an embodiment of the present application. Figure 5 As shown, the driving behavior detection device 200 includes: at least one processor 201; and a memory 202 that is communicatively connected to the at least one processor 201; wherein the memory 202 stores instructions that can be executed by the at least one processor 201, and the instructions are executed by the at least one processor 201 so that the at least one processor 201 can: obtain driving data of the vehicle to be tested; divide a multi-level detection area based on the driving data and a preset vehicle speed step; determine the acceleration detection parameters and the sudden acceleration parameters corresponding to each level of the detection area; compare the acceleration detection parameters corresponding to each level of the detection area with the preset driving behavior initial detection curve diagram to obtain a first driving behavior detection result; draw the acceleration curve corresponding to each level of the detection area based on the driving data, compare the sudden acceleration parameter with the acceleration curve, and obtain a second driving behavior detection result; determine the final driving behavior detection result according to the first driving behavior detection result and the second driving behavior detection result.

[0084] A non-volatile computer storage medium provided in an embodiment of the present application stores computer executable instructions, wherein the computer executable instructions are configured to: obtain driving data of a vehicle to be tested; divide a multi-level detection area based on the driving data and a preset vehicle speed step; determine acceleration detection parameters and sudden acceleration parameters corresponding to each level of detection area; compare the acceleration detection parameters corresponding to each level of detection area with a preset driving behavior initial detection curve diagram to obtain a first driving behavior detection result; draw an acceleration curve corresponding to each level of detection area based on the driving data, compare the sudden acceleration parameter with the acceleration curve, and obtain a second driving behavior detection result; and determine a final driving behavior detection result based on the first driving behavior detection result and the second driving behavior detection result.

[0085] Each embodiment in this application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0086] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the embodiments of the present application may have various modifications and variations. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A driving behavior detection method, characterized in that: The method comprises: Obtaining driving data of the vehicle to be tested; Based on the driving data and the preset vehicle speed step, a multi-level detection area is divided; Determine the acceleration detection parameters and the rapid acceleration parameters corresponding to each level of detection area; Comparing the acceleration detection parameters corresponding to each level of detection area with a preset driving behavior initial detection curve diagram to obtain a first driving behavior detection result; Based on the driving data, drawing acceleration curves corresponding to each level of detection area, comparing the rapid acceleration parameter with the acceleration curve, and obtaining a second driving behavior detection result; A final driving behavior detection result is determined based on the first driving behavior detection result and the second driving behavior detection result.

2. A driving behavior detection method according to claim 1, characterized in that: The step of obtaining the driving data of the vehicle to be tested specifically includes: Acquiring real-time driving reference data corresponding to the vehicle to be tested; Based on a preset driving data threshold, determining first abnormal data in the real-time driving reference data, and screening out the first abnormal data; Based on time continuity and a preset continuous data difference threshold, second abnormal data is determined in the real-time driving reference data, and difference processing is performed on the second abnormal data; Based on the data after the screening and difference processing, the driving data of the vehicle to be tested is obtained.

3. A driving behavior detection method according to claim 1, characterized in that: The dividing of the multi-level detection areas based on the driving data and the preset vehicle speed step size specifically includes: Determining an overall detection area based on the entire driving area in the driving data; Comparing the driving position in the driving data with a preset working condition traffic area map to determine a plurality of working condition detection areas; wherein the working condition detection areas at least include an urban area, a suburban area, and a highway area; Based on the preset vehicle speed step, the vehicle speed in the driving data is divided to determine a plurality of segment detection areas; wherein the difference between the initial vehicle speed and the terminal vehicle speed corresponding to a segment detection area is one of the preset vehicle speed step.

4. A driving behavior detection method according to claim 3, characterized in that: Determining the acceleration detection parameters corresponding to each level of detection area specifically includes: Based on the product of the real-time vehicle speed and the real-time acceleration corresponding to the overall detection area, a first speed parameter set is constructed, a plurality of speed parameters in the first speed parameter set are sorted, and a speed parameter corresponding to a preset percentile value is selected as the acceleration detection parameter of the overall area; Based on the product of the real-time vehicle speed and the real-time acceleration respectively corresponding to the plurality of working condition detection areas, constructing the second speed parameter sets respectively corresponding to the plurality of working condition detection areas, sorting the plurality of speed parameters in each of the second speed parameter sets, and selecting the speed parameter corresponding to the preset percentile value as the acceleration detection parameter respectively corresponding to each of the working condition detection areas; Based on the product of the real-time vehicle speed and the real-time acceleration respectively corresponding to the multiple segment detection areas, third speed parameter sets respectively corresponding to the multiple segment detection areas are constructed, the multiple speed parameters in each of the third speed parameter sets are sorted, and the speed parameter corresponding to the preset percentile value is selected as the acceleration detection parameter respectively corresponding to each of the segment detection areas.

5. A driving behavior detection method according to claim 1, characterized in that: The step of comparing the acceleration detection parameters corresponding to each level of the detection area with a preset driving behavior initial detection curve diagram to obtain a first driving behavior detection result specifically includes: Determine the average vehicle speed corresponding to each level of detection area, and determine the horizontal coordinate value in the initial detection curve diagram based on the average vehicle speed; Based on the acceleration detection parameters corresponding to each level of detection area, a vertical coordinate value is determined in the initial detection curve graph; Based on the abscissa value and the ordinate value, a calibration point is determined in the initial detection curve graph, and calibration is performed using different markers; The calibration point is compared with a plurality of reference curves in the initial detection curve diagram, and the first driving behavior detection result is obtained based on the positional relationship between the calibration point and the reference curve; wherein the plurality of reference curves correspond to different driving behaviors respectively.

6. A driving behavior detection method according to claim 1, characterized in that: After obtaining the first driving behavior detection result, the method further includes: Get the latest driving behavior detection data; Based on the latest driving behavior detection data, re-performing data fitting on the initial detection curve graph; Based on the fitting results, a plurality of new reference curves are determined to optimize the initial detection curve graph.

7. A driving behavior detection method according to claim 1, characterized in that: The step of drawing the acceleration curves corresponding to each level of detection area based on the driving data, and comparing the sudden acceleration parameter with the acceleration curve to obtain the second driving behavior detection result specifically includes: Determine the driving time corresponding to each level of detection area; Based on the real-time acceleration in the driving data and the driving time, constructing acceleration curves corresponding to each level of detection area; Determining a rapid acceleration parameter point in the acceleration curve whose absolute value is greater than a preset acceleration threshold; The second driving behavior detection result is obtained based on the proportion of the rapid acceleration parameter point in the acceleration curve.

8. A driving behavior detection method according to claim 1, characterized in that: The determining of a final driving behavior detection result according to the first driving behavior detection result and the second driving behavior detection result specifically includes: comparing the first driving behavior detection result with the second driving behavior detection result; When the first driving behavior detection result corresponding to any detection area is greater than the upper limit reference curve, or the second driving behavior detection result is that the proportion of the rapid acceleration parameter points is greater than the preset first proportion value, it is determined that the driving behavior corresponding to the detection area is intense driving; When the first driving behavior detection result corresponding to any detection area is less than the upper limit reference curve and greater than the lower limit reference curve, and the second driving behavior detection result is that the proportion of the rapid acceleration parameter point is less than the preset first proportion value and greater than the second proportion value, it is determined that the driving behavior corresponding to the detection area is normal driving; When the first driving behavior detection result corresponding to any detection area is less than the lower limit reference curve, or the second driving behavior detection result is that the proportion of rapid acceleration parameter points is less than a preset second proportion value, it is determined that the driving behavior corresponding to the detection area is gentle driving.

9. A driving behavior detection device, characterized in that: The device comprises a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the device is triggered to execute the method according to any one of claims 1 to 8.

10. A non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions can execute the method according to any one of claims 1 to 8.

Citation Information

Cited By

  • Method for generating actual road driving condition of motor vehicle

    CN119397197A