Traffic information flow construction method and device
A traffic information flow and construction method technology, applied in the field of autonomous driving, can solve the problems of inaccurate results and extremely difficult scene distribution collection, and achieve the effect of adapting to a wide range of scenes and accurate simulation results
Pending Publication Date: 2020-05-15
BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
View PDF6 Cites 6 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] At this stage, it is extremely difficult to obtain a complete set of scene distribution through field sampling
In addition, the scene obtained by using field sampling is generally more suitable for the collection place. The scene in other places may be different, and the results obtained by simulation may not be accurate
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 Construction
[0056] In the following, only some exemplary embodiments are briefly described. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. Accordingly, the drawings and descriptions are to be regarded as illustrative in nature and not restrictive.
[0057] figure 1 A flowchart showing a method for constructing a traffic information flow according to an embodiment of the present invention. like figure 1 As shown, the construction method of the traffic information flow may include:
[0058] Step S11, loading the area map of the simulation scene to be constructed.
[0059] Step S12, placing multiple agents in the area map.
[0060] Step S13, controlling each of the agents to run in the area map, and collecting the running status of each of the agents.
[0061]Step S14 , using traffic rules to carry out reinforcement learning on the running state of e...
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 embodiment of the invention provides a traffic information flow construction method and device. The method comprises the steps that a regional map of a simulation scene needing to be constructed is loaded; placing a plurality of intelligent agents in the regional map; controlling each intelligent agent to run in the regional map, and collecting the running state of each intelligent agent; andperforming reinforcement learning on the operation state of each agent by using a traffic rule to obtain a traffic information flow of the simulation scene. According to the embodiment of the invention, the traffic information flow of the simulation scene is trained by adopting the reinforcement learning method, and enough and complex automatic driving simulation scenes can be provided, so that the simulation result of the automatic driving vehicle is more accurate.
Description
technical field [0001] The present invention relates to the technical field of automatic driving, in particular to a method and device for constructing a traffic information flow. Background technique [0002] Statistically speaking, unmanned vehicles need to travel 200 million miles on a complete road network to ensure their performance surpasses that of humans. This goal is difficult to achieve in actual scenarios. Simulation can accelerate this goal. In the simulation, the road network traffic information is discretized to form discrete scenes. According to Bayesian theory, at least 3M (mega) scenes are needed to achieve this goal. However, the 3M scene is required to fully describe the scene distribution in the derivation. [0003] If you want to obtain a complete scene distribution, you can drive in a scene with relatively developed traffic information to collect. When constructing the simulation scene, the captured video of the real scene is played back frame by fra...
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): G06F30/20G06K9/62
Inventor 张俊飞孙庆瑞毛继明董芳芳
Owner BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD