Method for obtaining dynamic traffic information required by unmanned driving
A traffic information and driverless technology, applied in traffic control systems, traffic control systems of road vehicles, instruments, etc., can solve the problem of low accuracy, inability to accurately obtain the movement trend of dynamic things on the road, and cumbersome and complicated identification methods for dynamic traffic information. And other issues
Inactive Publication Date: 2019-05-28
TRINA SOLAR CO LTD
View PDF9 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] In the prior art, the recognition method for dynamic traffic information is cumbersome and complicated, and the accuracy is low, so unmanned vehicles cannot accurately obtain road dynamics and their movement trends, and unmanned vehicles cannot perform reasonable path planning and motion control decisions
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0024] In this embodiment, multiple wide-angle starlight-level cameras (3-4 mm focal length) are used, and the angle of view covers the front and left and right sides of the vehicle body. The smaller the focal length, the more serious the image distortion, so it is not recommended to use a camera with too small focal length. Calibrate the camera to eliminate image distortion to a certain extent, and obtain camera parameters for target spatial positioning estimation.
[0025] (1) Production of training samples: Use LabelImg software to mark objects in the real scene image through rectangular frames, add classification, and generate a sample database. The samples include various types of cars, trucks, bicycles, pedestrians, etc. Shooting under lighting conditions, such as collecting under different weather and time conditions on the application road section.
[0026] (2) Establish a Darknet neural network model based on yolo3. First, the current frame image is gridded, and the ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a method for obtaining dynamic traffic information required by unmanned driving. The method comprises the steps that a training sample is made; based on yolo3, a Darknet neuralnetwork model is built, an adjacent detection rectangular identification result is a part of a whole object or not is determined through an association probability, therefore, the area of a target inan image is repositioned, and the identification classification ID and the corresponding score of a sample are output; the identified whole target is used as a template, and template comparison withsubsequent multiple frames of images is conducted in a video frame number range through the template, so that the position where the target appears in the subsequent image is determined; through a comparison value and the position where the target appears, the confidence level is judged, and the possible displacement trend is predicted. Identification of a road dynamic thing and prediction of themovement trend of the road dynamic thing provide decision-making information of path planning and movement control of an unmanned vehicle, and a powerful reference is provided for value estimation ofpath planning.
Description
technical field [0001] The invention belongs to the technical field of unmanned driving, and in particular relates to a method for obtaining dynamic traffic information required by unmanned driving. Background technique [0002] Driverless car is a kind of smart car, also known as wheeled mobile robot, which mainly relies on the computer system-based smart driver in the car to achieve the purpose of driverless driving. Use on-board sensors to perceive the surrounding environment of the vehicle, and control the steering and speed of the vehicle according to the road, vehicle position and obstacle information obtained from the perception, so that the vehicle can drive safely and reliably on the road. [0003] In the prior art, the recognition method for dynamic traffic information is cumbersome and complicated, and the accuracy is low, so unmanned vehicles cannot accurately obtain road dynamics and their movement trends, and unmanned vehicles cannot perform reasonable path pla...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G08G1/0967
Inventor 陈剑徐涛
Owner TRINA SOLAR CO LTD