Path planning method based on Q-learning algorithm

A learning algorithm and path planning technology, applied in two-dimensional position/channel control, vehicle position/route/altitude control, instruments and other directions, can solve the problems of Q-learning algorithm staying in simulation, lack of combination of practical problems, etc. Efficiency improvement, fast speed, fast convergence effect

Active Publication Date: 2018-09-28
JILIN UNIV
View PDF6 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] At the same time, after reviewing a large number of papers, we found that the exploration

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
  • Path planning method based on Q-learning algorithm
  • Path planning method based on Q-learning algorithm
  • Path planning method based on Q-learning algorithm

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0030] See Figure 1 to Figure 6 Shown:

[0031] The path planning method based on the Q-learning algorithm provided by the present invention is as follows:

[0032] Step 1: Use an ordinary camera to collect images of our actual environment to obtain basic information;

[0033] The second step: MATLAB reads the picture information collected by the camera and performs binarization processing on the picture to determine the obstacle coordinates in the picture;

[0034] The third step: segmentation of the graphics. In order to simplify the learning process, we use the grid method to establish the environment model. We divide the picture in the previous step into 10×10 grids, judge in the program, if found in each grid Obstacles, the grid is defined as a grid with obstacles, the robot cannot pass, other grids are defined as grids without obstacles, and the robot can pass;

[0035] Step 4: Use the improved Q-learning algorithm to plan the path, first define the starting point and the end p...

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 discloses a path planning method based on a Q-learning algorithm. The method includes the following steps: firstly, obtaining basic information; secondly, determining the coordinate of an obstacle in an image; thirdly, carrying out segmentation on the image; fourthly, planning paths through an improved Q-learning algorithm; fifthly, obtaining an optimal path, and drawing the optimalpath through MATLAB according to a learning result; and sixthly, controlling a robot to walk for verification, and using a computer to control the robot to walk to verify the path on the basis of thelearning result. The beneficial effects are that a simulation experiment is performed in a grid environment, the method is successfully applied to path planning of the mobile robot in a multi-obstacleenvironment, and the result proves feasibility of the algorithm; the improved Q-learning algorithm can converge faster, with the learning frequency being obviously reduced and the efficiency being improved by 20% to the maximum; and the algorithm framework exhibits high universality for solving similar problems.

Description

technical field [0001] The invention relates to a path planning method, in particular to a path planning method based on a Q-learning algorithm. Background technique [0002] At present, an important milestone in reinforcement learning is the Q-learning algorithm. Q-learning is the most representative reinforcement learning method similar to the dynamic programming algorithm proposed by Watkins [1] in 1989. It provides intelligent systems A learning capability that uses experienced action sequences to select optimal actions in a Markov environment, and does not require a model of the environment. The Q-learning algorithm is actually a variation of the Markov decision process. It is currently the most understandable and widely used reinforcement learning method, and it learns in an incremental manner. Since Watkins proposed the Q-learning algorithm and proved its convergence, the algorithm has received widespread attention in the field of artificial intelligence and machine ...

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): G05D1/02
CPCG05D1/0221G05D1/0223G05D1/0246G05D1/0276
Inventor 千承辉马天录刘凯张宇轩
Owner JILIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap