Intelligent driving real vehicle test system

Abstract

The invention discloses an intelligent driving real vehicle test system, relates to the technical field of intelligent driving, and solves the problems that in the prior art, whether a vehicle can return to a set track within a set time after deviating in the driving process is usually considered during intelligent driving of the vehicle; the possibility of vehicle accidents is increased in practical application due to the fact that the deviation distance of the vehicle is too large is not considered; the method comprises the following steps: judging whether a vehicle exceeds a traffic marking center range in a corresponding driving path or not; if yes, starting timing; if not, continuing to judge; performing to-be-qualified classification on the corresponding vehicle intelligent drivers; judging whether the corresponding vehicle intelligent driving is marked as qualified or not; if yes, the real vehicle test is completed; if not, calling the vehicle driving information, and optimizing the corresponding vehicle according to the vehicle driving information; it is ensured that the vehicle can be kept on a correct path as much as possible in the driving process, the traffic accident risk caused by deviation is reduced, and the safety of road driving is improved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent driving, relates to intelligent driving testing technology, and specifically is an intelligent driving real vehicle testing system. Background Art

[0002] With the rapid development of science and technology, intelligent driving technology is gradually becoming a research hotspot and development trend in the automotive field. Intelligent driving aims to enable vehicles to achieve partial or full autonomous driving under various road conditions through advanced sensors, algorithms and control systems, thereby improving traffic safety, reducing traffic congestion, and providing drivers with a more convenient and comfortable travel experience; building a real vehicle test system can verify the performance of the intelligent driving system in a real road environment to ensure that it can operate safely in various complex and emergency situations; through actual testing, potential safety hazards can be discovered, and timely measures can be taken to improve them and reduce the risk of accidents; the real vehicle test system can quickly verify the effects of new algorithms and control strategies, and accelerate technology iteration and optimization; by collecting a large amount of actual operation data, in-depth analysis can be performed to discover the performance characteristics and existing problems of the system in different scenarios, providing strong support for technology research and development;

[0003] When performing intelligent driving of a vehicle, the existing technology often only considers whether the vehicle can return to the set trajectory within a specified time after deviating during driving; it does not take into account that if the vehicle deviates too far, the possibility of vehicle accidents will increase in actual applications.

[0004] The present invention provides an intelligent driving real vehicle testing system to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes an intelligent driving real vehicle testing system, which is used to solve the problem that the prior art often only considers whether the vehicle can return to the set trajectory within a specified time after being offset during driving; it does not consider that the vehicle offset distance is too large, which will increase the possibility of vehicle accidents in actual applications.

[0006] To achieve the above-mentioned object, a first aspect of the present invention provides an intelligent driving real vehicle testing system, comprising: a data acquisition module, an offset determination module and an intelligent driving optimization module;

[0007] Data acquisition module: obtain the road image in front of the vehicle in real time through the camera, identify and extract road information from the road image based on image processing technology; obtain the center range of traffic markings based on the road information;

[0008] Install several sensors and data recording devices to obtain vehicle driving information during the test; store the collected vehicle driving information in a database;

[0009] Offset judgment module: call the set traffic line center range to determine whether the vehicle exceeds the traffic line center range in the corresponding driving path; if yes, start timing until the vehicle deviates from the center of the traffic line and stops timing at the center of the traffic line; if no, continue to judge; classify the corresponding vehicle as pending qualification;

[0010] Intelligent driving optimization module: determines whether the corresponding vehicle's intelligent driving is marked as qualified; if yes, completes the actual vehicle test; if not, retrieves the vehicle driving information and optimizes the corresponding vehicle based on the vehicle driving information.

[0011] Preferably, the identifying and extracting road information from the road image according to the image processing technology includes:

[0012] Retrieve the road image and preprocess the image; extract edge information in the image through edge detection algorithm, and use region growing method to find the area where the road is located; automatically learn image features through convolutional neural network and extract road information; preprocessing includes: denoising, enhancement and grayscale; road information includes: location, shape, direction and color of traffic markings.

[0013] The present invention is beneficial to improving image quality and highlighting road features by preprocessing road images; and improving the accuracy and efficiency of road information extraction by automatically learning image features.

[0014] Preferably, storing the collected vehicle driving information in a database includes:

[0015] Retrieve vehicle driving information, pre-process the vehicle driving information, and import the vehicle driving information into the database using data import API technology; determine whether the vehicle driving information is correctly imported; if yes, complete the storage of the vehicle driving information and establish an information index; if no, send a reminder signal; wherein the vehicle driving information includes: vehicle speed, acceleration, steering angle, and the distance between the vehicle and the obstacle; the pre-processing steps include: cleaning the vehicle driving information and formatting the vehicle driving information that does not meet the database field requirements.

[0016] The present invention helps to ensure the accuracy and consistency of data by cleaning data; formatting data to make it meet the requirements of database fields; establishing indexes for commonly used query fields to improve query efficiency; and ensuring data integrity and consistency by checking whether data is correctly imported; and facilitating data extraction during subsequent operations by storing vehicle driving information in a database.

[0017] Preferably, the determining whether the vehicle exceeds the center range of the traffic marking in the corresponding driving path includes:

[0018] Set the coordinate system, mark the center point of the vehicle as (X1, Y1), and the coordinates of any point within the center of the traffic marking as (X2, Y2);

[0019] By formula Calculate the deviation distance;

[0020] Retrieve the center range of the traffic marking and mark it as C;

[0021] Determine whether d(t) is less than and equal to |C|; if yes, the vehicle does not exceed the center range of the traffic marking in the corresponding driving path; if no, the vehicle exceeds the center range of the traffic marking in the corresponding driving path.

[0022] It should be noted that the coordinate system is set by the staff based on actual work experience; the positive and negative values ​​of C are set by the staff according to the direction of vehicle deviation.

[0023] The present invention can timely detect the vehicle driving deviation by real-time monitoring of the deviation between the vehicle and the center range of the traffic marking, thereby preventing the vehicle from deviating from the road or causing a collision accident.

[0024] Preferably, the step of classifying the corresponding vehicle intelligent driving as pending for qualification includes:

[0025] The time when the vehicle starts to exceed the center range of the traffic line is retrieved and marked as t, and the deviation threshold time is set as T;

[0026] If t<T and d(t)≤|C|, the corresponding vehicle intelligent driving mark is qualified;

[0027] If t≥T and d(t)≤|C|, the corresponding vehicle intelligent driving mark is pending qualification;

[0028] If t≥T and d(t)>|C|, the corresponding vehicle intelligent driving mark is unqualified.

[0029] It should be noted that the specific deviation threshold time is set by the staff based on a large number of simulation tests and actual road test data.

[0030] The present invention records the time when the vehicle starts to exceed the center range of the traffic marking, monitors the vehicle's driving status in real time, and discovers potential problems in time; by setting different conditions, the vehicle's intelligent driving performance can be divided into three levels, which is helpful for more detailed evaluation and classification of problems; by recording and analyzing data, it helps staff to discover the performance characteristics and existing problems of the vehicle in different scenarios, and provide data support for algorithm optimization and system improvement.

[0031] Preferably, the optimizing the corresponding vehicle according to the vehicle driving information includes:

[0032] Retrieve vehicle driving information that is not marked as qualified from the database; use the corresponding vehicle driving information as input to construct a historical data optimization model; adjust the strategy for the vehicle driving information that is not marked as qualified according to the historical data optimization model; determine whether the vehicle intelligent driving is qualified after the strategy adjustment; if yes, complete the corresponding vehicle intelligent driving test; if not, re-adjust the strategy; wherein the strategy adjustment includes: optimizing algorithm parameters, replacing or calibrating sensors, improving control logic, and enhancing scenario adaptability.

[0033] The present invention retrieves the driving information of vehicles that are not marked as qualified, makes full use of existing data for analysis and optimization, and avoids waste of resources; based on the historical data optimization model, the performance of the intelligent driving system can be continuously improved; according to the driving information of each vehicle to be qualified, targeted strategy adjustments are made to improve the accuracy and effectiveness of the adjustments.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. In the intelligent driving real-car test system, the vehicle path adjustment unit obtains the image of the road in front of the vehicle in real time through the camera, and identifies the center range of the traffic markings from the acquired road images through image processing technology. According to the identified traffic markings and the current position of the vehicle, the ideal driving path that the vehicle should maintain is calculated to ensure that the vehicle is located in the center range of the markings, so that the vehicle always stays on the correct driving path, reducing the risk of collision caused by lane deviation, and at the same time, the vehicle travels more standardized and orderly, reducing traffic congestion caused by lane deviation.

[0036] 2. Based on the data analysis results, diagnose the reasons why the vehicle deviates from the ideal driving state, such as algorithm defects, sensor errors, improper control strategies, etc., and formulate corresponding adjustment strategies for the diagnosed problems. This may include optimizing algorithm parameters, replacing or calibrating sensors, improving control logic, and continuously improving the accuracy, stability and reliability of the vehicle's intelligent driving system, so that it can better adapt to various complex road environments and driving scenarios, and improve users' driving experience and satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 A schematic diagram showing the relationship between the modules included in the present invention;

[0039] Figure 2 It is a schematic diagram of the process of the intelligent driving test of the present invention;

[0040] Figure 3 Schematic diagram of specific steps of the offset determination module of the present invention. DETAILED DESCRIPTION

[0041] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] See also Figure 1-Figure 2 , the first aspect of the present invention provides an intelligent driving real vehicle testing system, including: a data acquisition module, an offset determination module and an intelligent driving optimization module;

[0043] Data acquisition module: obtain the road image in front of the vehicle in real time through the camera, identify and extract road information from the road image based on image processing technology; obtain the center range of traffic markings based on the road information;

[0044] Install several sensors and data recording devices to obtain vehicle driving information during the test; store the collected vehicle driving information in a database;

[0045] Offset judgment module: call the set traffic line center range to determine whether the vehicle exceeds the traffic line center range in the corresponding driving path; if yes, start timing until the vehicle deviates from the center of the traffic line and stops timing at the center of the traffic line; if no, continue to judge; classify the corresponding vehicle as pending qualification;

[0046] Intelligent driving optimization module: determines whether the corresponding vehicle's intelligent driving is marked as qualified; if yes, completes the actual vehicle test; if not, retrieves the vehicle driving information and optimizes the corresponding vehicle based on the vehicle driving information.

[0047] See also Figure 3 ,The specific steps of the offset determination module are to set the coordinate system, mark the center point coordinate of the vehicle as (X1, Y1), and the coordinate of any point within the center range of the traffic marking line as (X2, Y2);

[0048] By formula Calculate the deviation distance;

[0049] Retrieve the center range of the traffic marking and mark it as C;

[0050] Determine whether d(t) is less than and equal to |C|; if yes, the vehicle does not exceed the center range of the traffic marking in the corresponding driving path; if no, the vehicle exceeds the center range of the traffic marking in the corresponding driving path;

[0051] The time when the vehicle starts to exceed the center range of the traffic line is marked as t, and the deviation threshold time is set as T;

[0052] If t<T and d(t)≤|C|, the corresponding vehicle intelligent driving mark is qualified and the actual vehicle test is completed;

[0053] If t≥T and d(t)≤|C|, the corresponding vehicle intelligent driving mark is pending qualification;

[0054] If t≥T and d(t)>|C|, the corresponding vehicle intelligent driving is marked as unqualified;

[0055] The corresponding vehicle driving information is used as input to construct a historical data optimization model; the strategy is adjusted for the vehicle driving information that is not marked as qualified according to the historical data optimization model; it is determined whether the vehicle intelligent driving is qualified after the strategy adjustment; if so, the corresponding vehicle intelligent driving test is completed; if not, the strategy is adjusted again.

[0056] For example, there are three intelligent driving vehicles A, B, and C for testing; the point at the center of the traffic line is taken as the origin, and the coordinates are (0, 0); the horizontal and vertical coordinate units are both m;

[0057] Set the traffic line center range to 0-0.3m and the deviation threshold time to 5s;

[0058] According to the test, the coordinates of the center point of vehicle A farthest from the origin during driving are (0.4, 0.3); through the formula The calculated deviation distance is 0.5m; the deviation distance at the 3rd second is 0.2m, and no vehicle deviation occurs subsequently;

[0059] If vehicle A exceeds the center range of the traffic marking during driving, but returns to the center range of the traffic marking within the set deviation threshold time, vehicle A will be marked as pending;

[0060] According to the test, the coordinates of the center point of vehicle B farthest from the origin during driving are (0.2, 0.3); through the formula The calculated deviation distance is 0.36m; the deviation distance at the 2nd second is 0.1m, and no vehicle deviation occurs subsequently;

[0061] If vehicle B does not exceed the center range of the traffic line within the set deviation threshold time, vehicle B is marked as qualified and the actual vehicle test is completed;

[0062] According to the test, the coordinates of the center point of vehicle C farthest from the origin during driving are (0.4, 0.3); through the formula The calculated deviation distance is 0.5m; the deviation distance at the 5th second is 0.6m;

[0063] If vehicle C exceeds the center range of the traffic marking during driving and fails to return to the center range of the traffic marking within the set deviation threshold time, vehicle C will be marked as unqualified;

[0064] The driving information of vehicles A and C is used as input to construct a historical data optimization model; the strategies of vehicles A and C are adjusted according to the historical data optimization model;

[0065] The vehicles A and C after the strategy adjustment are tested again; if the vehicle A does not exceed the center range of the traffic line within the set deviation threshold time, the vehicle A is re-marked as qualified, and the actual vehicle test of the vehicle A is completed;

[0066] If vehicle C still exceeds the center range of the traffic marking during driving, but returns to the center range of the traffic marking within the set deviation threshold time, vehicle C will be marked as pending and the strategy of vehicle C will be adjusted again.

[0067] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0068] The working principle of the present invention: the present invention obtains the center range of traffic markings according to road information; obtains the vehicle driving information of the vehicle during the test; stores the collected vehicle driving information in a database; retrieves the set center range of traffic markings to determine whether the vehicle exceeds the center range of traffic markings in the corresponding driving path; if yes, starts timing until the vehicle that deviates from the center of the traffic markings is at the center of the traffic markings and stops timing; if no, continues to judge; classifies the corresponding vehicle intelligent driving as pending qualification; determines whether the corresponding vehicle intelligent driving is marked as qualified; if yes, completes the actual vehicle test; if no, retrieves the vehicle driving information, and optimizes the corresponding vehicle according to the vehicle driving information.

[0069] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent driving real vehicle testing system, characterized in that: include: Data acquisition module, deviation determination module and intelligent driving optimization module; Data acquisition module: obtain the road image in front of the vehicle in real time through the camera, identify and extract road information from the road image based on image processing technology; obtain the center range of traffic markings based on the road information; Install several sensors and data recording devices to obtain vehicle driving information during the test; store the collected vehicle driving information in a database; Offset judgment module: call the set traffic line center range to determine whether the vehicle exceeds the traffic line center range in the corresponding driving path; if yes, start timing until the vehicle deviates from the center of the traffic line and stops timing at the center of the traffic line; if no, continue to judge; classify the corresponding vehicle as pending qualification; Intelligent driving optimization module: determines whether the corresponding vehicle intelligent driving is marked as qualified; If yes, the actual vehicle test is completed; If not, retrieve the vehicle driving information and optimize the corresponding vehicle according to the vehicle driving information.

2. The intelligent driving real vehicle testing system according to claim 1, characterized in that: The method of identifying and extracting road information from a road image using image processing technology includes: Retrieve the road image and preprocess the image; extract edge information in the image through edge detection algorithm, and use region growing method to find the area where the road is located; automatically learn image features through convolutional neural network and extract road information; preprocessing includes: denoising, enhancement and grayscale; road information includes: location, shape, direction and color of traffic markings.

3. The intelligent driving real vehicle testing system according to claim 1, characterized in that: The step of storing the collected vehicle driving information in a database includes: Retrieve vehicle driving information, pre-process the vehicle driving information, and import the vehicle driving information into the database using data import API technology; determine whether the vehicle driving information is correctly imported; if yes, complete the storage of the vehicle driving information and establish an information index; if no, send a reminder signal; wherein the vehicle driving information includes: vehicle speed, acceleration, steering angle, and the distance between the vehicle and the obstacle; the pre-processing steps include: cleaning the vehicle driving information and formatting the vehicle driving information that does not meet the database field requirements.

4. The intelligent driving real vehicle testing system according to claim 1, characterized in that: The determining whether the vehicle exceeds the center range of the traffic marking in the corresponding driving path includes: Set the coordinate system, mark the center point of the vehicle as (X1, Y1), and the coordinates of any point within the center of the traffic marking as (X2, Y2); By formula Calculate the deviation distance; Retrieve the center range of the traffic marking and mark it as C; Determine whether d(t) is less than and equal to |C|; if yes, the vehicle does not exceed the center range of the traffic marking in the corresponding driving path; if no, the vehicle exceeds the center range of the traffic marking in the corresponding driving path.

5. The intelligent driving real vehicle testing system according to claim 1, characterized in that: The corresponding vehicle intelligent driving will be classified as pending, including: The time when the vehicle starts to exceed the center range of the traffic line is retrieved and marked as t, and the deviation threshold time is set as T; If t<T and d(t)≤|C|, the corresponding vehicle intelligent driving mark is qualified; If t≥T and d(t)≤|C|, the corresponding vehicle intelligent driving mark is pending qualification; If t≥T and d(t)>|C|, the corresponding vehicle intelligent driving mark is unqualified.

6. The intelligent driving real vehicle testing system according to claim 1, characterized in that: The optimizing of the corresponding vehicle according to the vehicle driving information includes: Retrieve vehicle driving information that is not marked as qualified from the database; use the corresponding vehicle driving information as input to construct a historical data optimization model; adjust the strategy for the vehicle driving information that is not marked as qualified according to the historical data optimization model; determine whether the vehicle intelligent driving is qualified after the strategy adjustment; if yes, complete the corresponding vehicle intelligent driving test; if not, re-adjust the strategy; wherein the strategy adjustment includes: optimizing algorithm parameters, replacing or calibrating sensors, improving control logic, and enhancing scenario adaptability.

Images