CN112163285A - Modeling method of road surface type prediction model for simulating driving system - Google Patents
Modeling method of road surface type prediction model for simulating driving system Download PDFInfo
- Publication number
- CN112163285A CN112163285A CN202011206457.XA CN202011206457A CN112163285A CN 112163285 A CN112163285 A CN 112163285A CN 202011206457 A CN202011206457 A CN 202011206457A CN 112163285 A CN112163285 A CN 112163285A
- Authority
- CN
- China
- Prior art keywords
- road surface
- prediction model
- driving system
- database
- surface type
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000012360 testing method Methods 0.000 claims abstract description 49
- 238000004088 simulation Methods 0.000 claims abstract description 20
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 12
- 238000012549 training Methods 0.000 claims abstract description 12
- 238000002474 experimental method Methods 0.000 claims abstract description 7
- 238000012216 screening Methods 0.000 claims abstract description 5
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 230000001133 acceleration Effects 0.000 claims description 6
- 230000037396 body weight Effects 0.000 claims description 3
- 238000013515 script Methods 0.000 claims description 3
- 238000011161 development Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Geometry (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Civil Engineering (AREA)
- Structural Engineering (AREA)
- Architecture (AREA)
- Automation & Control Theory (AREA)
- Aviation & Aerospace Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a modeling method of a road surface type prediction model for a simulation driving system, which comprises the following steps: carrying out real vehicle tests, and acquiring vehicle dynamic parameter data under various road surface types and various working conditions to obtain an original test database with road surface type identification; screening an original experiment database to obtain a modeling database, and training a pavement type prediction model by using a K-Means algorithm; embedding the road surface prediction model obtained by training into a simulation driving system; carrying out the real vehicle test again, collecting test data to obtain a model test database, and importing the model test database into the simulation driving system; and (3) carrying out reliability test on the road surface prediction model by using the simulated driving system with the model test database. The method is based on real vehicle test data, and uses the K-Means algorithm to train the road surface type prediction model, thereby solving the problems of the prior art that the limitation to complex road surfaces and the robustness in the using process are difficult to ensure.
Description
Technical Field
The invention relates to a road surface type prediction model, in particular to a modeling method of a K-Means-based road surface type prediction model for a simulation driving system.
Background
In the present day when automobiles are becoming an indispensable part of people's lives, the automobile industry is increasingly competitive. How to rapidly develop automobiles meeting the requirements of customers becomes an important issue for manufacturers in the automobile industry. With the gradual popularization and application of artificial intelligence, automobiles are gradually developing towards the direction of no humanization and intellectualization. The driving simulation system can realize vehicle dynamics simulation with higher reliability on the premise of occupying less resources, and is adopted by most enterprises, so that the development process is shortened, and the development cost is reduced.
The road surface information is an indispensable part for vehicle simulation. When the simulated driving system (also called virtual driving system) is used for simulation, the reliability of a simulation result can be enhanced by a real road signal, the overall performance of the simulated driving system is improved, and the simulated driving experience of a user is improved. However, considering the complexity of the driving environment of the vehicle, how to set the correct road surface type for the vehicle in the simulation environment becomes an engineering problem with large workload and complicated operation. Patent CN201810584408.6 "a road surface identification system and method" proposes a road surface identification method based on physical attributes of road surfaces, but the method is complex to operate, and has certain limitations on identification of complex road surfaces; patent CN201810824707.2 "a road surface identification method and apparatus" proposes a road surface identification method based on image data, which is susceptible to image data acquisition environment, such as concern, and robustness in the using process is difficult to guarantee.
Disclosure of Invention
In order to solve the above technical problems, an object of the present invention is to provide a modeling method for a road surface type prediction model for simulating a driving system, which uses a K-Means algorithm to establish a relationship between a vehicle dynamic parameter and a road surface type based on real vehicle test data.
In order to achieve the above object, the present invention provides a modeling method of a road surface type prediction model for simulating a driving system, comprising the steps of:
carrying out real vehicle tests, and acquiring vehicle dynamic parameter data under various road surface types and various working conditions to obtain an original test database with road surface type identification;
screening an original experiment database to obtain a modeling database;
training a pavement type prediction model by using a K-Means algorithm on the basis of a modeling database;
embedding the road surface prediction model obtained by training into a simulation driving system;
carrying out real vehicle test again, collecting continuous real vehicle test data containing various road surface types and various working conditions to obtain a model test database, and importing the model test database into a simulation driving system;
and (3) carrying out reliability test on the road surface prediction model by using the simulated driving system with the model test database.
Further, the vehicle dynamic parameters comprise vehicle speed, yaw rate, vehicle body acceleration, vehicle body weight, vehicle body pitch angle speed, vehicle body pitch angle, vehicle body roll angle speed and vehicle body dynamic deflection.
Further, the plurality of road surface types include an expressway, an urban road, a suburban road, an off-road surface, and the like; the multiple working conditions comprise straight running, turning, rapid acceleration, rapid deceleration and parking and warehousing.
Further, when the original experiment database is screened, the abnormal data and the invalid data need to be deleted: taking data outside the range of plus or minus 3 times of standard deviation of the whole data as abnormal data; null data is considered invalid data.
Further, the screened original experimental database must be normalized to serve as a modeling database to avoid failure in modeling due to the difference in magnitude between different variables.
Furthermore, when the K-Means algorithm is used for carrying out road surface type prediction model training, at least 100 iterations are needed to ensure that the parameters of the obtained model are optimal.
Further, the model test database should contain all the conditions and road types to be predicted.
Furthermore, when the working conditions in the model test database are reproduced, the scripts must be used for automatic operation so as to ensure that the working conditions in the model test database are completely reproduced.
Due to the adoption of the technical scheme, the invention achieves the following technical effects: the method is based on real vehicle test data, and establishes the relationship between the vehicle dynamic parameters and the road surface type by using the K-Means algorithm, so that a high-accuracy road surface type prediction model can be obtained, various road surface types can be accurately predicted, and the robustness of the road surface type prediction model used in a simulated driving system is ensured.
Drawings
Fig. 1 is a flow chart illustrating a modeling method of a road surface type prediction model based on a K-Means algorithm for simulating a driving system according to the present invention.
Detailed Description
In order to make the technical solution of the embodiments of the present invention better understood, the technical solution of the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by equivalent changes and modifications by one skilled in the art based on the embodiments of the present invention, shall fall within the scope of the present invention.
Referring to fig. 1, the present embodiment provides a modeling method for a K-Means-based road surface type prediction model for a driving simulation system, including the following steps:
s1, collecting original data
And carrying out real vehicle tests, and acquiring vehicle dynamic parameter data under various types and various working conditions to obtain an original test database with road surface type identification, wherein the acquired vehicle dynamic parameters comprise vehicle speed, yaw rate, vehicle body acceleration, vehicle body weight, vehicle body pitch angle rate, vehicle body pitch angle, vehicle body roll angle rate, vehicle body dynamic deflection and the like. The original experimental data comprises data of various road surface types and various working conditions; the types of the road surfaces for testing comprise expressways, urban roads, suburban roads, off-road roads and the like; the working conditions comprise straight running, turning, rapid acceleration, rapid deceleration, parking and warehousing and the like.
S2, screening original data
And screening the original experiment database, and deleting abnormal data and invalid data. Taking data outside the range of plus or minus 3 times of standard deviation of the whole data as abnormal data; null data is considered invalid data. The screened original experimental database can be used as a modeling database after standard normalization is carried out, so that the problem that modeling fails due to the difference of orders of magnitude between different variables is avoided.
S3, training a K-Means model
Training a pavement type prediction model by using a K-Means algorithm on the basis of a modeling database; when the K-Means algorithm is used for carrying out the road surface type prediction model training, at least 100 iterations are needed to ensure that the parameters of the obtained model are optimal.
S4. model embedded simulation driving system
And embedding the road surface prediction model obtained by training into a simulated driving system.
S5, collecting test data
Carrying out real vehicle test again, collecting continuous real vehicle test data containing various road surface types and working conditions to obtain a model test database, and importing the model test database into a simulation driving system; the model test database should contain all the conditions and road types that need to be predicted.
S6, testing model credibility
Carrying out reliability test on the road surface prediction model by using a simulation driving system with a model test database; when the working conditions in the model test database are reproduced, the scripts are required to be used for automatically performing so as to ensure that the working conditions in the model test database are completely reproduced.
S7, judging whether modeling needs to be carried out again
If the model credibility is found to be lower than 85% by the test, returning to the step S1; otherwise, the next step is carried out.
S8, outputting the model
And outputting a road surface type prediction model based on a K-Means algorithm.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention; also, the above description should be understood as being readily apparent to those skilled in the relevant art and can be implemented, and therefore, other equivalent changes and modifications without departing from the concept disclosed herein are intended to be included within the scope of the present invention.
Claims (8)
1. A modeling method for a road surface type prediction model for simulating a driving system, characterized by comprising the steps of:
carrying out real vehicle tests, and acquiring vehicle dynamic parameter data under various road surface types and various working conditions to obtain an original test database with road surface type identification;
screening a facility original experiment database to obtain a modeling database;
training a pavement type prediction model by using a K-Means algorithm based on a facility modeling database;
embedding the road surface type prediction model obtained by training into a simulation driving system;
carrying out real vehicle test again, collecting continuous real vehicle test data containing various road surface types and various working conditions to obtain a model test database, and importing the model test database into a simulation driving system;
and carrying out reliability test on the road surface type prediction model by using the simulated driving system with the model test database.
2. The modeling method of a road surface type prediction model for a simulated driving system of claim 1, wherein said vehicle dynamics parameters include vehicle speed, yaw rate, body acceleration, body weight, body pitch rate, body pitch angle, body roll rate, body moment.
3. The modeling method of a road surface type prediction model for a simulated driving system according to claim 1, wherein the plurality of road surface types include an expressway, an urban road, a suburban road, an off-road, and the like; the multiple working conditions comprise straight running, turning, rapid acceleration, rapid deceleration and parking and warehousing.
4. The modeling method of the road surface type prediction model for a simulation driving system according to claim 1, wherein when the original experiment database is screened, the abnormal data and the invalid data are deleted: taking data outside the range of plus or minus 3 times of standard deviation of the whole data as abnormal data; null data is considered invalid data.
5. The modeling method of the road surface type prediction model for simulating the driving system according to claim 4, characterized in that the screened original experiment database must be normalized to be used as the modeling database.
6. The modeling method of the road surface type prediction model for simulating the driving system according to claim 1, wherein the K-Means algorithm is applied to perform the training of the road surface type prediction model at least 100 times.
7. The modeling method of a road surface type prediction model for a simulated driving system according to claim 1, characterized in that the model test database should contain all the conditions and road surface types to be predicted.
8. The modeling method of a road surface type prediction model for a simulation driving system of claim 1, wherein in the step of performing the model confidence test, when the working conditions in the model test database are reproduced, it is necessary to use a script to automatically perform so as to ensure that the working conditions in the model test database are completely reproduced.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011206457.XA CN112163285B (en) | 2020-11-03 | 2020-11-03 | Modeling method of road surface type prediction model for simulating driving system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011206457.XA CN112163285B (en) | 2020-11-03 | 2020-11-03 | Modeling method of road surface type prediction model for simulating driving system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112163285A true CN112163285A (en) | 2021-01-01 |
CN112163285B CN112163285B (en) | 2024-01-23 |
Family
ID=73865491
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011206457.XA Active CN112163285B (en) | 2020-11-03 | 2020-11-03 | Modeling method of road surface type prediction model for simulating driving system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112163285B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113479177A (en) * | 2021-07-07 | 2021-10-08 | 的卢技术有限公司 | Vehicle brake control method, system, computer device and storage medium |
CN113515813A (en) * | 2021-07-16 | 2021-10-19 | 长安大学 | On-site verification method for simulation reliability of automobile dynamics simulation software |
CN114996114A (en) * | 2021-03-01 | 2022-09-02 | 浙江天行健智能科技有限公司 | Intelligent traffic simulation test system and method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140122045A1 (en) * | 2013-08-08 | 2014-05-01 | Iteris, Inc. | Pavement condition analysis from modeling impact of traffic characteristics, weather data and road conditions on segments of a transportation network infrastructure |
CN104331953A (en) * | 2014-10-29 | 2015-02-04 | 云南大学 | Car behavior data identification and management method based on internet of things |
CN107679557A (en) * | 2017-09-19 | 2018-02-09 | 平安科技(深圳)有限公司 | Driving model training method, driver's recognition methods, device, equipment and medium |
CN111507233A (en) * | 2020-04-13 | 2020-08-07 | 吉林大学 | Multi-mode information fusion intelligent vehicle pavement type identification method |
CN111815948A (en) * | 2020-06-12 | 2020-10-23 | 南京航空航天大学 | Vehicle running condition prediction method based on condition characteristics |
-
2020
- 2020-11-03 CN CN202011206457.XA patent/CN112163285B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140122045A1 (en) * | 2013-08-08 | 2014-05-01 | Iteris, Inc. | Pavement condition analysis from modeling impact of traffic characteristics, weather data and road conditions on segments of a transportation network infrastructure |
CN104331953A (en) * | 2014-10-29 | 2015-02-04 | 云南大学 | Car behavior data identification and management method based on internet of things |
CN107679557A (en) * | 2017-09-19 | 2018-02-09 | 平安科技(深圳)有限公司 | Driving model training method, driver's recognition methods, device, equipment and medium |
CN111507233A (en) * | 2020-04-13 | 2020-08-07 | 吉林大学 | Multi-mode information fusion intelligent vehicle pavement type identification method |
CN111815948A (en) * | 2020-06-12 | 2020-10-23 | 南京航空航天大学 | Vehicle running condition prediction method based on condition characteristics |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114996114A (en) * | 2021-03-01 | 2022-09-02 | 浙江天行健智能科技有限公司 | Intelligent traffic simulation test system and method |
CN113479177A (en) * | 2021-07-07 | 2021-10-08 | 的卢技术有限公司 | Vehicle brake control method, system, computer device and storage medium |
CN113479177B (en) * | 2021-07-07 | 2024-04-09 | 的卢技术有限公司 | Vehicle brake control method, system, computer device, and storage medium |
CN113515813A (en) * | 2021-07-16 | 2021-10-19 | 长安大学 | On-site verification method for simulation reliability of automobile dynamics simulation software |
CN113515813B (en) * | 2021-07-16 | 2023-03-14 | 长安大学 | On-site verification method for simulation reliability of automobile dynamics simulation software |
Also Published As
Publication number | Publication date |
---|---|
CN112163285B (en) | 2024-01-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112163285A (en) | Modeling method of road surface type prediction model for simulating driving system | |
US20210197851A1 (en) | Method for building virtual scenario library for autonomous vehicle | |
US20170316127A1 (en) | Method and apparatus for constructing testing scenario for driverless vehicle | |
CN112997060A (en) | Method and system for modifying a control unit of an autonomous vehicle | |
CN114077541A (en) | Method and system for validating automatic control software for an autonomous vehicle | |
CN109461342B (en) | Teaching system for unmanned motor vehicle and teaching method thereof | |
CN111832652B (en) | Training method and device for decision model | |
CN113343461A (en) | Simulation method and device for automatic driving vehicle, electronic equipment and storage medium | |
US11893840B2 (en) | Systems and methods for modeling and simulation in vehicle forensics | |
CN113777952A (en) | Automatic driving simulation test method for interactive mapping of real vehicle and virtual vehicle | |
US20220266854A1 (en) | Method for Operating a Driver Assistance System of a Vehicle and Driver Assistance System for a Vehicle | |
CN112698582A (en) | ADAS ECU simulation test method and system | |
CN115291515A (en) | Automatic driving simulation test system and evaluation method based on digital twinning | |
WO2021146906A1 (en) | Test scenario simulation method and apparatus, computer device, and storage medium | |
CN112597996B (en) | Method for detecting traffic sign significance in natural scene based on task driving | |
CN115689946B (en) | Image restoration method, electronic device and computer program product | |
CN113297667A (en) | Intelligent driving data closed-loop method and system | |
Li | A scenario-based development framework for autonomous driving | |
CN108460057B (en) | User travel mining method and device based on unsupervised learning | |
CN108459996B (en) | User travel mining method and device based on supervised learning | |
CN115857685A (en) | Perception algorithm data closed-loop method and related device | |
Coe et al. | Virtualized in situ software update verification: verification of over-the-air automotive software updates | |
CN115129027A (en) | Automatic evaluation method and device for intelligent driving | |
US20230273047A1 (en) | Method and Apparatus for Providing a High-Resolution Digital Map | |
US20230177234A1 (en) | Method and system for quantifying vehicle requirement |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |