Driving evaluation method based on pattern recognition
Patent Information
- Application Number
- CN202310633653.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-31
- Publication Date
- 2024-12-03
Smart Images

Figure CN119066352A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of predictive driving evaluation of heavy trucks, and specifically to a driving evaluation method based on pattern recognition. Background Art
[0002] Predictive driving refers to the driver's predictive driving operation while driving. It is necessary to analyze and judge the road conditions, traffic flow and related route information at any time, so as to make a preliminary estimate of the possible situation ahead and take corresponding measures in time. The main application scenarios of predictive driving include: early deceleration near traffic lights, early downshifting near uphill, control of sudden acceleration and deceleration, and the smoothness of vehicle driving. Good and effective predictive driving can effectively improve the economy, comfort and safety of the vehicle, and is extremely helpful for vehicle fuel saving, emission reduction and accident reduction.
[0003] The current in-vehicle driving scoring system does not comprehensively consider the impact of predictive driving on the score, and there is no unified standard for evaluating drivers' predictive driving. Since the evaluation results are subject to the influence of personal experience and subjective judgment, the reliability and accuracy of the predictive driving evaluation are low.
[0004] Pattern recognition refers to the process of processing and analyzing various forms of information that represent things or phenomena in order to describe, identify, classify and explain them. Pattern recognition-related algorithms can be used to establish a predictive driving evaluation system. Summary of the invention
[0005] In order to solve the problems existing in the background technology, the present invention aims to provide a driving evaluation method based on pattern recognition, which aims to evaluate whether the driver's predictive driving operation is reasonable, improve the reliability of the evaluation of predictive driving, avoid misjudgment caused by subjective factors, and solve the above-mentioned situation.
[0006] Solutions for solving problems
[0007] A driving evaluation method based on pattern recognition, the method comprising:
[0008] S1: Select sample vehicles, collect sample vehicle operation data, and select valid vehicle data after preprocessing to calculate feature values and determine the operation status;
[0009] S2, using the classification algorithm in pattern recognition to train the feature values and operating states in step 1 (classifier design), obtain the key classification parameters through parameter setting and the number of iterations, and establish a classification model; that is, complete the classification decision (recognition), and divide the levels according to the number of classifications, which can be divided into excellent, good, medium, poor and other levels;
[0010] S3, collecting the operating data of the target vehicle in the target time period, calculating the characteristic value and identifying the operating state using the processing method in step 1, inputting the obtained one-dimensional characteristic vector and the operating state into the classification model established in step 2, and determining the classification level by combining the weight calculation results of each factor;
[0011] S4, scoring the classification level obtained in step 3, and proposing improvement suggestions for the disadvantages of the target vehicle, that is, obtaining a rationality evaluation of the predictive driving operation.
[0012] Furthermore, in step 1, all vehicles in the vehicle cluster are numbered, random integers of no less than one third of the total number are generated within the numbering range, and vehicles corresponding to the random numbers are selected as sample vehicles.
[0013] Furthermore, in step 1, multiple characteristic values are actively selected as factors for identifying the operating state, and multiple operating states are preset. The characteristic values are used as characteristic parameters in the classification process to form a one-dimensional characteristic vector group.
[0014] Furthermore, in step 2, a classification algorithm is selected to classify the feature vector groups of the sample vehicles, and the classification results are sorted, and the preset multiple operating states are matched with the classification cluster closest to it, and the parameter weights finally obtained are used as the design parameters of the classifier.
[0015] Furthermore, in step 3, the characteristic values of the target vehicle data are calculated to form a one-dimensional characteristic vector, which is classified using a designed classifier, and the operating status is matched according to the classification results. The obtained classification level is used for comprehensive evaluation, and improvement suggestions are proposed for the disadvantageous items in the characteristics.
[0016] Furthermore, the characteristic value parameters include: slope conditions, sudden acceleration conditions, sudden deceleration conditions and gear usage conditions under different operating conditions.
[0017] Furthermore, the vehicle operation data includes: vehicle speed data, rotation speed data, brake switch data and GPS data.
[0018] Beneficial effects:
[0019] The beneficial effects of the above technical solution are:
[0020] First, by extracting sample vehicle operation data for feature calculation, a feature library related to predictive driving is obtained; then pattern recognition is performed on the feature library, and the evaluation criteria and grades of predictive driving are established based on the results to form an evaluation system model; then, by processing the target vehicle operation data and substituting the calculation results into the evaluation system model, the evaluation grade and improvement index of the vehicle's predictive driving are obtained. The evaluation model and implementation method described in the patent of this invention are more scientific and more reasonable, and solve the problem of low reliability of existing predictive driving evaluation. The final evaluation result is the evaluation grade, and reasonable improvement suggestions are output. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] Figure 1 It is a schematic diagram of the predictive driving evaluation system involved in the patent of this invention;
[0022] Figure 2 This is a processing flow chart of the predictive driving evaluation system involved in the patent of this invention. DETAILED DESCRIPTION
[0023] The specific implementation mode of the present invention is described below in conjunction with embodiments:
[0024] It should be noted that the structures, proportions, sizes, etc. shown in this specification are only used to match the contents disclosed in the specification so that people familiar with this technology can understand and read them, and are not used to limit the conditions under which the present invention can be implemented. Any structural modification, change in proportional relationship or adjustment of size should still fall within the scope of the technical content disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.
[0025] At the same time, the terms such as "upper", "lower", "left", "right", "middle" and "one" cited in this specification are only for the convenience of description and are not used to limit the scope of implementation of the present invention. Changes or adjustments to their relative relationships should be regarded as the scope of implementation of the present invention without substantially changing the technical content.
[0026] Embodiment 1:
[0027] The patent of this invention mainly includes four parts: sample data feature extraction, sample feature pattern recognition and evaluation system establishment, target vehicle data extraction and data processing, and target vehicle driving evaluation.
[0028] The evaluation system refers to obtaining sample vehicle operation-related data through data acquisition equipment; classifying the vehicle's predictive driving-related operations and identifying the vehicle's operating status based on a pattern recognition algorithm, establishing a classification model and designing an evaluation level based on the classification results; calculating characteristic values for the target vehicle's driving data, using the classification model to evaluate the target vehicle's predictive driving operations, and assigning levels (excellent, good, medium, and poor), and proposing improvement suggestions for disadvantageous items, thereby guiding the driver to perform predictive driving reasonably and improving vehicle safety and economy.
[0029] The present invention patent adopts the following technical scheme to realize the above system: a predictive driving evaluation method and system based on pattern recognition. It includes the following steps: sample vehicle operation data feature calculation and state recognition, classification model construction and verification, target vehicle feature calculation, and reasonable level assessment. The operation data input is the vehicle speed, driving speed, brake switch signal, GPS information and other data during the vehicle driving process.
[0030] The data sources of the evaluation system are mainly: vehicle ECU and other controllers, vehicle-mounted terminal equipment, etc., transmitted through the CAN bus. The required data are: vehicle speed, engine speed, brake switch signal, GPS data, etc.
[0031] The data processing process of the evaluation system is to analyze and process the collected vehicle driving data to obtain characteristic values. The characteristic parameters that need to be extracted include: sudden acceleration, sudden deceleration, uphill and downhill recognition, and gear position before uphill.
[0032] The classification model establishment process of the evaluation system extracts vehicle feature values to form a one-dimensional feature vector for each vehicle, classifies the feature values of all sample vehicles using a classification algorithm, identifies the current vehicle operating status, and matches the vehicle operating status with the classification result to form an excellent, good, medium, and poor grade classification.
[0033] The evaluation system also includes suggestions for improving disadvantages based on the rationality evaluation level obtained, including: accelerating or decelerating in advance according to the road conditions ahead, minimizing sudden acceleration and deceleration, and downshifting in advance before going uphill to avoid high throttle acceleration.
[0034] The present invention establishes a predictive driving evaluation system based on pattern recognition through data collection, characteristic value calculation, classifier design, and rationality evaluation. The system fully considers multiple factors related to predictive driving, such as: slope, slope length, gear, operating status, number of rapid accelerations, number of rapid decelerations, etc., obtains the weights of the relevant parameters through the pattern recognition classification algorithm, establishes a classification model, and forms a classification level; the evaluation level is obtained by bringing the characteristic value of the target vehicle into the classification model, and targeted rationality improvement suggestions are given. Compared with the prior art, the evaluation system described in the patent of the present invention is more reliable and accurate, and is not affected by personal experience and subjectivity when participating in the evaluation. The final evaluation result is the rationality evaluation level and improvement suggestions.
[0035] Embodiment 2:
[0036] A flow chart of a predictive evaluation method based on pattern recognition is as follows: Figure 1 and Figure 2 shown.
[0037] In this embodiment, the specific process includes: sample vehicle data input 1, data processing 2, eigenvalue calculation 3, pattern recognition based on the results obtained in 2 and 3 to establish a classification model 4, target vehicle data collection 5, extraction of eigenvalues of the target vehicle 6, establishment of evaluation model input parameters 7 based on the obtained classification model 4, classification using the evaluation model 8, evaluation of the classification level based on the classification results obtained in 8 9, and proposal of improvement suggestions 10.
[0038] The embodiment of the system comprises the following specific steps:
[0039] Step 1: Data preparation and feature calculation
[0040] Screen valid vehicles and randomly select vehicles nationwide; collect various required vehicle operation data; process and analyze the data (pre-processing such as data cleaning), calculate characteristic values and judge the operation status. Specific characteristic value parameters include slope conditions, sudden acceleration conditions, sudden deceleration conditions, gear usage conditions, etc. under different operation conditions; the required data include vehicle speed data, speed data, brake switch data, GPS data, etc.
[0041] Step 2: Model building and validation
[0042] According to the vehicle feature values obtained in step 1, classification is performed using the classification algorithm in pattern recognition. The key classification parameters are obtained by setting parameters and the number of iterations, and a classification model is established. Some vehicle feature data obtained in step 1 are selected for model verification.
[0043] Step 3: Target vehicle data input
[0044] Select the target vehicle, collect the operating data of the target vehicle within the specified time period according to the data required in step 1, calculate the characteristic value and identify the operating status using the processing method in step 1, input the obtained one-dimensional feature vector into the classification model established in step 2, and determine the classification level by combining the weight calculation results of each factor.
[0045] Step 4: Reasonable evaluation
[0046] Using the classification level obtained in step 3, a specific rationality evaluation of the target vehicle is given, and improvement suggestions are made for the disadvantages. Finally, the rationality evaluation of the predictive driving operation is obtained.
[0047] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0048] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0049] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
Claims
1. A driving evaluation method based on pattern recognition, characterized in that: The method comprises: S1: Select sample vehicles, collect sample vehicle operation data, and select valid vehicle data after preprocessing to calculate feature values and determine the operation status; S2, using the classification algorithm in pattern recognition to train the feature values and operating states in step 1, obtain key classification parameters through parameter setting and number of iterations, and establish a classification model; S3, collecting the operating data of the target vehicle in the target time period, calculating the characteristic value and identifying the operating state using the processing method in step 1, inputting the obtained one-dimensional characteristic vector and the operating state into the classification model established in step 2, and determining the classification level by combining the weight calculation results of each factor; S4, scoring the classification level obtained in step 3, and proposing improvement suggestions for the disadvantages of the target vehicle, that is, obtaining a rationality evaluation of the predictive driving operation.
2. The predictive driving evaluation system based on pattern recognition according to claim 1, characterized in that: In the step 1, all vehicles in the vehicle cluster are numbered, random integers of no less than one third of the total number are generated within the numbering range, and vehicles corresponding to the random numbers are selected as sample vehicles.
3. The predictive driving evaluation system based on pattern recognition according to claim 1, characterized in that: In step 1, multiple characteristic values are actively selected as factors for identifying the operating state, and multiple operating states are preset. The characteristic values are used as characteristic parameters in the classification process to form a one-dimensional characteristic vector group.
4. The predictive driving evaluation system based on pattern recognition according to claim 1, characterized in that: In step 2, a classification algorithm is selected to classify the feature vector groups of sample vehicles, and the classification results are sorted. The preset multiple operating states are matched with the classification cluster closest to them, and the parameter weights finally obtained are used as the design parameters of the classifier.
5. The predictive driving evaluation system based on pattern recognition according to claim 1, characterized in that: In step 3, the characteristic value of the target vehicle data is calculated to form a one-dimensional characteristic vector, which is classified using a designed classifier, and the operating status is matched according to the classification result. The obtained classification level is used for comprehensive evaluation, and improvement suggestions are proposed for the disadvantageous items in the characteristics.
6. The predictive driving evaluation system based on pattern recognition according to claim 1, characterized in that: The characteristic value parameters include: slope conditions, sudden acceleration conditions, sudden deceleration conditions and gear usage conditions under different operating conditions.
7. The predictive driving evaluation system based on pattern recognition according to claim 1, characterized in that: The vehicle operation data includes: vehicle speed data, rotation speed data, brake switch data and GPS data.
Citation Information
Patent Citations
Driving risk evaluation method and system
CN107909678A
Vehicle drivability evaluation method and system
CN111951430A
Systems and methods for utilizing machine learning and feature selection to classify driving behavior
US10990837B1
