Share Email Print

Proceedings Paper

The research of autonomous obstacle avoidance of mobile robot based on multi-sensor integration
Author(s): Ming Zhao; Baoling Han
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The object of this study is the bionic quadruped mobile robot. The study has proposed a system design plan for mobile robot obstacle avoidance with the binocular stereo visual sensor and the self-control 3D Lidar integrated with modified ant colony optimization path planning to realize the reconstruction of the environmental map. Because the working condition of a mobile robot is complex, the result of the 3D reconstruction with a single binocular sensor is undesirable when feature points are few and the light condition is poor. Therefore, this system integrates the stereo vision sensor blumblebee2 and the Lidar sensor together to detect the cloud information of 3D points of environmental obstacles. This paper proposes the sensor information fusion technology to rebuild the environment map. Firstly, according to the Lidar data and visual data on obstacle detection respectively, and then consider two methods respectively to detect the distribution of obstacles. Finally fusing the data to get the more complete, more accurate distribution of obstacles in the scene. Then the thesis introduces ant colony algorithm. It has analyzed advantages and disadvantages of the ant colony optimization and its formation cause deeply, and then improved the system with the help of the ant colony optimization to increase the rate of convergence and precision of the algorithm in robot path planning. Such improvements and integrations overcome the shortcomings of the ant colony optimization like involving into the local optimal solution easily, slow search speed and poor search results. This experiment deals with images and programs the motor drive under the compiling environment of Matlab and Visual Studio and establishes the visual 2.5D grid map. Finally it plans a global path for the mobile robot according to the ant colony algorithm. The feasibility and effectiveness of the system are confirmed by ROS and simulation platform of Linux.

Paper Details

Date Published: 9 January 2017
PDF: 9 pages
Proc. SPIE 10025, Advanced Sensor Systems and Applications VII, 1002514 (9 January 2017); doi: 10.1117/12.2242249
Show Author Affiliations
Ming Zhao, Beijing Institute of Technology (China)
Baoling Han, Beijing Institute of Technology (China)

Published in SPIE Proceedings Vol. 10025:
Advanced Sensor Systems and Applications VII
Tiegen Liu; Shibin Jiang; Rene Landgraf, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?