Check patentability & draft patents in minutes with Patsnap Eureka AI!

Pedestrian avoidance method and system for realizing vehicle L3-level automatic driving based on DQN

A technology for automatic driving and vehicle movement, applied in the direction of vehicle components, external condition input parameters, control devices, etc., can solve the problems that vehicles and pedestrians cannot be avoided, achieve good pedestrian avoidance performance, reduce training samples, and improve iteration efficiency.

Pending Publication Date: 2022-02-18
CHINA NANHU ACAD OF ELECTRONICS & INFORMATION TECH
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method uses a fuzzy controller to imitate the pedestrian behavior and response in the process of collision avoidance. Experiments show that this fuzzy controller-based approach can achieve avoidance with a reliability of 75%, but this method has an obvious disadvantage: see figure 1 As shown in (a)-(b), it can only work in a specific state, that is, vehicles and pedestrians are driving in the same lane, and at the same time, the left lane needs to be empty and have enough distance to overtake pedestrians
None of these existing technologies can achieve pedestrian avoidance in L3 autonomous driving of vehicles

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Pedestrian avoidance method and system for realizing vehicle L3-level automatic driving based on DQN
  • Pedestrian avoidance method and system for realizing vehicle L3-level automatic driving based on DQN
  • Pedestrian avoidance method and system for realizing vehicle L3-level automatic driving based on DQN

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0063] For ease of understanding, this specific embodiment is a preferred embodiment of the pedestrian avoidance method and device based on DQN to realize L3 automatic driving proposed by the present invention, to describe the structure and invention points of the present invention in detail, but not as a claim of the present invention limited scope of protection.

[0064] The invention innovatively applies reinforcement learning to pedestrian avoidance in vehicle automatic driving. Reinforcement learning can obtain digital rewards by evaluating the interaction between the agent (autonomous driving vehicle in this invention) and the state of the environment (road environment in this invention), and use the greedy strategy to find the behavior that maximizes the reward. The goal-directed learning in the interaction between the agent and the environment is abstracted through the reinforcement learning framework, such as assuming that it has Markov properties. When an agent trie...

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

No PUM Login to View More

Abstract

The invention provides a pedestrian avoidance method for realizing vehicle L3-level automatic driving based on DQN. The pedestrian avoidance method comprises the following steps: detecting whether pedestrians exist in front and vehicles exist around through a sensor of an automatic driving vehicle; and determining a pedestrian avoidance strategy after DQN training according to the detected pedestrian and vehicle information and whether other vehicles exist in an overtaking lane of the driving route of the autonomous vehicle. According to the invention, the L3-level automatic driving pedestrian avoidance strategy is realized with low training cost by using the DQN.

Description

【Technical field】 [0001] The present invention relates to the technical field of automatic driving, in particular to a pedestrian avoidance method and system for realizing vehicle L3 automatic driving based on DQN. 【Background technique】 [0002] For ease of understanding, the following terms used in this application are explained: [0003] L2, L3: The American Society of Automotive Engineers formulated the automatic driving classification standard (L0~L5) in 2014. L2 means that in order to realize partial automatic driving, it provides driving support for multiple operations in the steering wheel and acceleration and deceleration through the driving environment. Other driving Actions are all operated by human drivers; L3 means conditional automation, where all driving operations are completed by the unmanned driving system, and human drivers provide appropriate responses according to system requests. [0004] AEB: Automatic emergency braking technology. [0005] TensorFlo...

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
no application Login to View More
IPC IPC(8): B60W30/095B60W40/02B60W60/00
CPCB60W30/0956B60W60/0017B60W40/02B60W2554/4029B60W2552/50B60W2554/80
Inventor 管达志徐庶刘庆杰马金艳高爽
Owner CHINA NANHU ACAD OF ELECTRONICS & INFORMATION TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com