CN114511827B - Intelligent driving-assisted vehicle cloud sensing closed-loop processing method - Google Patents
Intelligent driving-assisted vehicle cloud sensing closed-loop processing method Download PDFInfo
- Publication number
- CN114511827B CN114511827B CN202111443566.8A CN202111443566A CN114511827B CN 114511827 B CN114511827 B CN 114511827B CN 202111443566 A CN202111443566 A CN 202111443566A CN 114511827 B CN114511827 B CN 114511827B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- model
- data
- closed
- training
- 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.)
- Active
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 11
- 238000012549 training Methods 0.000 claims abstract description 27
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 23
- 238000011156 evaluation Methods 0.000 claims abstract description 23
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000004458 analytical method Methods 0.000 claims abstract description 6
- 238000000034 method Methods 0.000 claims description 16
- 230000007613 environmental effect Effects 0.000 claims description 9
- 230000008447 perception Effects 0.000 claims description 7
- 238000013145 classification model Methods 0.000 claims description 6
- 230000004927 fusion Effects 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 8
- 238000000605 extraction Methods 0.000 description 7
- 238000011161 development Methods 0.000 description 6
- 238000002372 labelling Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 241001465754 Metazoa Species 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- HPTJABJPZMULFH-UHFFFAOYSA-N 12-[(Cyclohexylcarbamoyl)amino]dodecanoic acid Chemical compound OC(=O)CCCCCCCCCCCNC(=O)NC1CCCCC1 HPTJABJPZMULFH-UHFFFAOYSA-N 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/182—Distributed file systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/40—Transformation of program code
- G06F8/41—Compilation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45504—Abstract machines for programme code execution, e.g. Java virtual machine [JVM], interpreters, emulators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a vehicle cloud sensing closed-loop processing method for intelligent auxiliary driving, which comprises the following steps: 1) The vehicle end acquires environment data and uploads the environment data to the FTP server; 2) Mounting a distributed storage system in an FTP server, and dividing a storage space of a cloud sensing closed-loop system in the distributed storage system; 3) Cloud data processing: A. data preprocessing: preprocessing the environment video, automatically pre-marking and manually marking; B. model training: training tasks are carried out by combining training algorithms and corresponding training AI engines; C. model evaluation: pulling a model and a data set required by model evaluation by a vehicle cloud sensing closed-loop system; 4) And (3) vehicle-end data application: and finishing model compiling work through a compiler, and issuing the model compiling work to a vehicle-end chip through OTA. According to the invention, through multi-scene fusion analysis, a plurality of vehicle-mounted chips are supported, and the full-flow opening of the service deployed at the vehicle end and the automatic serial connection of the system can be realized.
Description
Technical Field
The invention relates to the technical field of intelligent auxiliary driving, in particular to a cloud sensing closed-loop processing method for intelligent auxiliary driving.
Background
In recent years, research in the field of intelligent auxiliary driving is continuously developed, and the intelligent auxiliary driving is an important component of future intelligent traffic and is one of current research hotspots. The current auxiliary scheme generally comprises a sensing system and a control system (automatic driving system), wherein the surrounding environment can be sensed by the automatic driving vehicle in the driving process through the sensing system, and the obstacles such as other vehicles, figures, animals, road signs and the like are included; the control system processes and controls the data collected by the sensing system to achieve the effect of driving assistance.
In order to ensure the driving safety, the accuracy of a sensing system is required to be tested; at present, when accuracy test is carried out on a sensing system, a mode of matching and evaluating the obstacle output by the sensing system with labeling data is generally adopted; the method is to carry out overall evaluation on the overall result output by the sensing system and the marked overall result, and iterate the sensing system based on the evaluation index of the overall evaluation output. Chinese patent CN202011568410.8 discloses a method, apparatus, device and storage medium for testing an automatic driving vehicle sensing system, which can sense the surrounding environment during driving, including other vehicles, figures, animals, road signs and other obstacles, but the sensing environment related to the scheme does not meet the actual situation of the existing vehicle road multi-scene environment; the existing large-vehicle enterprises or Internet companies are subjected to strong coupling of data pre-marking and marking platforms during data processing, and the existing data service modularization concept is not met, and the pre-marking and marking cannot be well adapted, so that the quality of the data service is low; meanwhile, with the establishment of more and more vehicle-mounted chip companies, the variety of chips in the future is various, the key of how a vehicle enterprise performs multi-chip collocation is the adaptation of an algorithm model to the chips, but at present, most of vehicle enterprises are bound with the algorithm model and a chip-processing mode; the whole intelligent auxiliary driving system is complicated, has poor self-adaptability and cannot meet the actual conditions of the existing vehicle road multi-scene environment.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide an intelligent driving-assisted vehicle cloud sensing closed-loop processing method, which supports various vehicle-mounted chips through multi-scene fusion analysis and algorithm data customization, realizes vehicle-end data acquisition, picture and video data preprocessing, data labeling, model training and model compiling, and can realize the purposes of full-process opening of service deployed at a vehicle end of a compiled model and automatic series connection of the system.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows: a vehicle cloud perception closed-loop processing method for intelligent auxiliary driving is characterized by comprising the following steps of: the method comprises the following steps:
1) The vehicle end acquires environment data comprising environment pictures and environment videos and uploads the environment data to the FTP server;
2) Mounting a distributed storage system in an FTP server, and dividing a storage space of a vehicle cloud sensing closed-loop system in the distributed storage system, wherein the vehicle cloud sensing closed-loop system is stored in the storage space of the vehicle cloud sensing closed-loop system, and the uploaded environmental data in the step 1) are stored in the storage space of a non-vehicle cloud sensing closed-loop system;
3) Cloud data processing:
A. data preprocessing: preprocessing an environmental video uploaded by a vehicle end to obtain an image data packet, and automatically pre-marking the image data packet by utilizing a picture classification model provided by a vehicle cloud perception closed-loop system; then, the user manually screens the pre-marked image data packet according to the requirement to realize manual marking;
B. model training: forming a data set by the image data packets after automatic pre-marking and manual marking, and carrying out training tasks by combining a training algorithm and a corresponding training AI engine;
C. model evaluation: after the training task is finished, the vehicle cloud perception closed-loop system pulls the model and the data set required by model evaluation, and compares the model and the data set with the set evaluation standard, and if the model evaluation is not passed, the step 3) is repeated;
4) And (3) vehicle-end data application:
and 3) after the model evaluation is passed, finishing model compiling work through a compiler, and sending the model compiling work to a vehicle end chip through OTA to finish model application, namely finishing intelligent auxiliary driving of the vehicle.
Further, in step 2), the environment data is set up and run to the FTP server, where the springboard machine is directly mounted in the directory of the distributed storage system by using a fuse protocol in a mount manner.
Further, in step 3), the vehicle cloud sensing closed-loop system provides a set of built-in picture classification models for implementing multi-scene sensing analysis.
Further, in step 4), when the model compilation needs to be performed on a certain vehicle-mounted chip, a compiler corresponding to the vehicle-mounted chip is called for compilation.
Compared with the prior art, the invention has the following advantages:
1. the vehicle end realizes the data acquisition and uploading server, the system performs image and video data preprocessing, and the system performs cooperative processing of pre-marking and manual marking on the contact, so that multi-scene fusion analysis can be realized, and the scene environment of the vehicle road can be sensed more comprehensively.
2. In the model training process, users can realize algorithm model self-definition, so that the requirements of different users are met; meanwhile, the evaluation standard can be trained by the self-defined model according to the requirements of the user, so that the intelligent driving assistance adaptability is higher.
3. After the model evaluation is passed, the model compiling work is completed through a compiler, the compiler operates in a mode of a Docker mirror image, and after the Docker mirror image is manufactured, the model is compiled into a corresponding vehicle-mounted chip model, so that the purposes of supporting various vehicle-mounted chips and finally realizing the full-flow opening of the service deployed at the vehicle end and the automatic serial connection of the system of the model can be realized.
Drawings
Fig. 1 is a flow chart of the operation of the present invention.
Fig. 2 is a flow chart of data upload.
FIG. 3 is a flow chart of environmental data preprocessing.
Fig. 4 is a flow chart of model compilation.
Detailed Description
The invention will be further described with reference to the drawings and examples.
Examples: referring to fig. 1 to 4, a vehicle cloud sensing closed-loop processing method for intelligent auxiliary driving includes the following steps:
1) The vehicle end is provided with an environment sensing system (such as an environment sensor, a camera and the like) so as to acquire environment data, wherein the environment data mainly comprises environment pictures and environment videos and is uploaded to an FTP server.
2) And (3) mounting a distributed storage system (HDFS) in the FTP server, and dividing a storage space of the vehicle cloud sensing closed-loop system in the distributed storage system, wherein the vehicle cloud sensing closed-loop system is stored in the storage space of the vehicle cloud sensing closed-loop system, and the uploaded environmental data in the step 1) is stored in the storage space of the non-vehicle cloud sensing closed-loop system. Because the environmental data required by the sensing closed-loop algorithm is mainly batch non-real-time data, the data volume is huge, video streams, pictures and the like reach TB level, and the real-time requirement of the data is not high; therefore, in the implementation process, the environment data is constructed and run to the FTP server by adopting the jump board machine, wherein the jump board machine is directly mounted in the directory of the HDFS by using a fuse protocol in a mount mode. Therefore, the operation on the FTP directory and the file is the operation on the HDFS; the HDFS root directory is divided into a special directory serving as a vehicle cloud sensing closed-loop system exclusive directory. And the user uses the FTP client to upload the file to be uploaded to the exclusive directory of the HDFS non-vehicle cloud sensing closed-loop system, and when the file is processed by using the vehicle cloud sensing closed-loop system, the file to be uploaded needs to be imported into the exclusive directory of the vehicle cloud sensing closed-loop system from the exclusive directory of the HDFS non-vehicle cloud sensing closed-loop system.
3) Cloud data processing (performed by the vehicle cloud-aware closed-loop system):
A. data preprocessing: preprocessing an environmental video uploaded by a vehicle end to obtain an image data packet, and directly forming the image data packet by an environmental picture uploaded by the vehicle end; the specific process is as follows: the video data collected by the user terminal is subjected to an image extraction function, so that the subsequent image classification preprocessing and warehousing are convenient; the video frame extraction tool extracts images of the input video files and generates image data packets in a system adaptation format, and the main functions supported by the video frame extraction tool are as follows:
supporting frame extraction period parameters;
supporting frame extraction frequency parameters;
picture resolution, format, etc., default JPEG format, resolution: 1920x1080;
support the frame selection function according to the time range of second;
video format: MP4;
supporting the starting time of the output frame extraction task and the progress of the frame extraction task;
support setting picture quality settings.
And then, automatically pre-labeling the image data packet by using a picture classification model provided by the vehicle cloud sensing closed-loop system. Specifically, the vehicle cloud sensing closed-loop system provides an image target detection algorithm to automatically pre-label the image; the vehicle cloud sensing closed-loop system provides a group of built-in picture classification models for realizing multi-scene sensing analysis, and specifically comprises the steps of identifying scenes such as vehicles, pedestrians, traffic signs, signal lamps, weather, light conditions and the like.
Then, the user manually screens the pre-marked image data packets according to the needs to realize manual marking. The image data packets after automatic pre-marking and manual marking are stored in a sample warehouse for standby. Specifically, the user can upload the model for image preprocessing by himself; and according to the provided example classification algorithm, the algorithm core part is modified, so that the self algorithm can be conveniently and rapidly uploaded and used on a vehicle cloud sensing closed-loop system platform.
End cloud collaborative development: the cloud collaborative development of the terminal not only has the detailed content, but also can use the plug-in provided by the vehicle cloud sensing closed-loop system to register the debugged algorithm into the algorithm management of the vehicle cloud sensing closed-loop system. Or the development environment is packaged into a custom AI training frame for use in training tasks. The system provides a Pycharm professional plug-in and supports end cloud collaboration. The existing Docker user directly puts the Docker repo, and the vehicle cloud sensing closed-loop system can read the Docker repo.
B. Model training: and forming a data set by the image data packets after automatic pre-marking and manual marking, and carrying out training tasks by combining a training algorithm and a corresponding training AI engine. After the algorithm is created, the user is allowed to modify the algorithm entry path, run the used AI framework and parameter list, etc. The training task can be performed by adding an algorithm given by algorithm management to a data set formed by labeling the data uploaded by the vehicle end and additionally selecting a corresponding training AI engine. The model generated after training is completed can be used for intelligently assisting driving assistance on vehicle end data application.
The end cloud collaborative development provided by the scheme can well solve the problems that the existing algorithm is difficult, no adaptive dependency package exists, the algorithm debugging is difficult to complete and the like; the system provides a most basic AI engine, wherein the engine is only provided with a CUDA suite compatible with Nvidia GPU drive, and basic python and corresponding mini conda, and a developer can perform collaborative development in an end cloud environment: and (3) completing the AI training frame required by the user, the installation of the dependent package required by the algorithm, the code development, the algorithm debugging and the like.
C. Model evaluation: after the training task is finished, the vehicle cloud perception closed-loop system pulls the model and the data set required by model evaluation, and compares the model and the data set with the set evaluation standard, and if the model evaluation is not passed, the step 3) is repeated; in the implementation process, a user makes a script according to the required index, writes the result into a file and stores the result in a Docker fixed directory, so that the user-defined model evaluation can be realized.
4) And (3) vehicle-end data application:
and 3) after the model evaluation is passed, finishing model compiling work through a compiler, and sending the model compiling work to a vehicle end chip through OTA to finish model application, namely finishing intelligent auxiliary driving of the vehicle. When a model compiling is needed to be carried out on a certain vehicle-mounted chip, a compiler corresponding to the vehicle-mounted chip is called for compiling. For example, the compiler operates in a mode of a Docker mirror image, and when compiling, pulls the relevant Docker and invokes a compiling tool to compile; after the Docker mirror image is manufactured by a chip compiling tool, compiling the Nvidia model into a corresponding vehicle-mounted chip model through a compiler to realize vehicle-end data application; therefore, various vehicle-mounted chips can be supported, and applicability is improved.
When a user needs to compile a certain algorithm, the model to be compiled, the compiler used and as a version of which compiled model is pointed out. When the task is pulled up, the host process downloads the model identified by the NVIDIA GPU to be compiled, and after the downloading is completed, a compiling tool is called to compile the model to be compiled into a model related to the chip. After compiling is completed, uploading the chip related model to OTA. So far, the whole vehicle cloud sensing closed loop whole process comprises the following steps: the whole process is completed through the vehicle-end data acquisition and uploading, the cloud data multilayer processing and the vehicle-end data OTA issuing application.
Finally, it should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the technical solution, and those skilled in the art should understand that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the present invention, and all such modifications and equivalents are included in the scope of the claims.
Claims (4)
1. A vehicle cloud perception closed-loop processing method for intelligent auxiliary driving is characterized by comprising the following steps of: the method comprises the following steps:
1) The vehicle end acquires environment data comprising environment pictures and environment videos and uploads the environment data to the FTP server;
2) Mounting a distributed storage system in an FTP server, and dividing a storage space of a vehicle cloud sensing closed-loop system in the distributed storage system, wherein the vehicle cloud sensing closed-loop system is stored in the storage space of the vehicle cloud sensing closed-loop system, and the uploaded environmental data in the step 1) are stored in the storage space of a non-vehicle cloud sensing closed-loop system;
3) Cloud data processing:
A. data preprocessing: preprocessing an environmental video uploaded by a vehicle end to obtain an image data packet, and automatically pre-marking the image data packet by utilizing a picture classification model provided by a vehicle cloud perception closed-loop system; then, the user manually screens the pre-marked image data packet according to the requirement to realize manual marking;
B. model training: forming a data set by the image data packets after automatic pre-marking and manual marking, and carrying out training tasks by combining a training algorithm and a corresponding training AI engine;
C. model evaluation: after the training task is finished, the vehicle cloud perception closed-loop system pulls the model and the data set required by model evaluation, and compares the model and the data set with the set evaluation standard, and if the model evaluation is not passed, the step 3) is repeated;
4) And (3) vehicle-end data application:
and 3) after the model evaluation is passed, finishing model compiling work through a compiler, and sending the model compiling work to a vehicle end chip through OTA to finish model application, namely finishing intelligent auxiliary driving of the vehicle.
2. The intelligent driving-assisted vehicle cloud sensing closed-loop processing method according to claim 1, wherein the method comprises the following steps of: in the step 2), the environment data is constructed and run to the FTP server by adopting a jump board machine, wherein the jump board machine is directly mounted in a directory of the distributed storage system by using a fuse protocol in a mount mode.
3. The intelligent driving-assisted vehicle cloud sensing closed-loop processing method according to claim 1, wherein the method comprises the following steps of: in step 3), the vehicle cloud sensing closed-loop system provides a set of built-in picture classification models for realizing multi-scene sensing analysis.
4. The intelligent driving-assisted vehicle cloud sensing closed-loop processing method according to claim 1, wherein the method comprises the following steps of: in the step 4), when a certain vehicle-mounted chip needs to be subjected to model compiling, a compiler corresponding to the vehicle-mounted chip is called for compiling.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111443566.8A CN114511827B (en) | 2021-11-30 | 2021-11-30 | Intelligent driving-assisted vehicle cloud sensing closed-loop processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111443566.8A CN114511827B (en) | 2021-11-30 | 2021-11-30 | Intelligent driving-assisted vehicle cloud sensing closed-loop processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114511827A CN114511827A (en) | 2022-05-17 |
| CN114511827B true CN114511827B (en) | 2024-04-12 |
Family
ID=81548318
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111443566.8A Active CN114511827B (en) | 2021-11-30 | 2021-11-30 | Intelligent driving-assisted vehicle cloud sensing closed-loop processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114511827B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116665025B (en) * | 2023-07-31 | 2023-11-14 | 福思(杭州)智能科技有限公司 | Data closed-loop method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110070139A (en) * | 2019-04-28 | 2019-07-30 | 吉林大学 | Small sample towards automatic Pilot environment sensing is in ring learning system and method |
| CN112035640A (en) * | 2020-08-31 | 2020-12-04 | 重庆长安汽车股份有限公司 | Refined question-answering method based on intelligent question-answering robot, storage medium and intelligent equipment |
| CN112862874A (en) * | 2021-04-23 | 2021-05-28 | 腾讯科技（深圳）有限公司 | Point cloud data matching method and device, electronic equipment and computer storage medium |
| CN113297667A (en) * | 2021-04-30 | 2021-08-24 | 东风汽车集团股份有限公司 | Intelligent driving data closed-loop method and system |
| CN113449632A (en) * | 2021-06-28 | 2021-09-28 | 重庆长安汽车股份有限公司 | Vision and radar perception algorithm optimization method and system based on fusion perception and automobile |
| CN113641734A (en) * | 2021-08-12 | 2021-11-12 | 驭势科技（北京）有限公司 | Data processing method, device, equipment and medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10332320B2 (en) * | 2017-04-17 | 2019-06-25 | Intel Corporation | Autonomous vehicle advanced sensing and response |
| US11328219B2 (en) * | 2018-04-12 | 2022-05-10 | Baidu Usa Llc | System and method for training a machine learning model deployed on a simulation platform |
| US10755007B2 (en) * | 2018-05-17 | 2020-08-25 | Toyota Jidosha Kabushiki Kaisha | Mixed reality simulation system for testing vehicle control system designs |
-
2021
- 2021-11-30 CN CN202111443566.8A patent/CN114511827B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110070139A (en) * | 2019-04-28 | 2019-07-30 | 吉林大学 | Small sample towards automatic Pilot environment sensing is in ring learning system and method |
| CN112035640A (en) * | 2020-08-31 | 2020-12-04 | 重庆长安汽车股份有限公司 | Refined question-answering method based on intelligent question-answering robot, storage medium and intelligent equipment |
| CN112862874A (en) * | 2021-04-23 | 2021-05-28 | 腾讯科技（深圳）有限公司 | Point cloud data matching method and device, electronic equipment and computer storage medium |
| CN113297667A (en) * | 2021-04-30 | 2021-08-24 | 东风汽车集团股份有限公司 | Intelligent driving data closed-loop method and system |
| CN113449632A (en) * | 2021-06-28 | 2021-09-28 | 重庆长安汽车股份有限公司 | Vision and radar perception algorithm optimization method and system based on fusion perception and automobile |
| CN113641734A (en) * | 2021-08-12 | 2021-11-12 | 驭势科技（北京）有限公司 | Data processing method, device, equipment and medium |
Non-Patent Citations (3)
| Title |
|---|
| Semantic Segmentation of 3D LiDAR Data in Dynamic Scene Using Semi-Supervised Learning;Jilin Mei 等;IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS;20190610;第21卷(第6期);全文 * |
| 数据智能:趋势与挑战;吴俊杰;刘冠男;王静远;左源;部慧;林浩;;系统工程理论与实践;20200803(第08期);第195-228页 * |
| 虚实系统互驱的混合增强智能开放创新平台的架构与方案;张俊;李灵犀;林懿伦;张天韵;张科;许沛东;阮克羽;沈聃;;智能科学与技术学报;20191215(第04期);第64-76页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114511827A (en) | 2022-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107609502A (en) | Method and apparatus for controlling automatic driving vehicle | |
| CN110765661A (en) | Automatic driving simulation scene generation method and device, electronic equipment and storage medium | |
| CN106776294B (en) | Automatic android mobile phone testing method and system | |
| CN106066849B (en) | A kind of template page editing system and method | |
| CN114511827B (en) | Intelligent driving-assisted vehicle cloud sensing closed-loop processing method | |
| CN104765617A (en) | Stream application function interface distributing method and system realized on basis of HTML5 | |
| CN110807219B (en) | Three-dimensional simulation modeling method, device, terminal and storage medium for road network | |
| CN113259597B (en) | Image processing method, apparatus, device, medium, and program product | |
| CN108711152B (en) | Image analysis method and device based on AI technology and user terminal | |
| CN112291477B (en) | Multimedia information processing method, device, storage medium and electronic device | |
| CN111507325B (en) | Industrial visual OCR recognition system and method based on deep learning | |
| CN110928620A (en) | Method and system for evaluating distraction of automobile HMI design to attract driving attention | |
| Tung et al. | Large-scale object detection of images from network cameras in variable ambient lighting conditions | |
| CN111598006A (en) | Method and device for labeling objects | |
| CN112651358A (en) | Target detection method and device, network camera and storage medium | |
| CN113297667B (en) | Intelligent driving data closed-loop method and system | |
| CN114419502A (en) | Data analysis method and device and storage medium | |
| CN114415542A (en) | Automatic driving simulation system, method, server and medium | |
| CN115688496B (en) | Method for obtaining automatic driving simulation test script and related device | |
| CN110928762B (en) | Display screen debugging method and system | |
| CN1606864A (en) | Description generation in the form of metadata | |
| CN103778671A (en) | ETC device and information interaction method | |
| CN115684155A (en) | Smartphone high-throughput blood calcium concentration prediction system and method | |
| CN115422058A (en) | Method and device for testing face recognition application | |
| CN115103206A (en) | Video data processing method, device, equipment, system and storage medium |
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 |