Author:
Nagai Isaku, ,Tanaka Yutaka
Abstract
We developed a visual device that tracks floor images and calculates the movement of a camera on a mobile robot. The mobile robot has caterpillar-tread wheels and uses our visual tracking device for localization. The robot is localized and controlled in real time based on the information on the estimated position and direction using FPGA, SRAM, and a small CPU board. Location and direction error over a closed path is eliminated by searching for an original floor image memorized initially at the point from which the robot started the run. Experimental results demonstrate the advantages of the proposal using the visual tracking device localizing a mobile robot with wheel slippage and under changing light conditions. We also show that the robot runs along a closed path repeatedly without a straying from the track by using the original image to correct accumulated error.
Publisher
Fuji Technology Press Ltd.
Subject
Electrical and Electronic Engineering,General Computer Science
Reference10 articles.
1. G. Antonelli, S. Chiaverini, and G. Fusco, “An odometry calibration method for mobile robots based on the least-squares technique,” Proc. 2003 American Control Conference, pp. 3429-3434, 2003.
2. A. T. Le, D. C. Rye, and H. F. Durrant-Whyte, “Estimation of tracksoil interactions for autonomous tracked vehicles,” Proc. IEEE Int. Conf. on Robotics and Automation, pp. 1388-1393, 1997.
3. A. Kelly, “Mobile Robot Localization from Large Scale Appearance Mosaics,” The International Journal of Robotics Research, Vol.19, No.11, pp. 1104-1125, 2000.
4. N. Gracias, S. van der Zwaan, A. Bernardino, and J. Santos-Victor, “Mosaic Based Navigation for Autonomous Underwater Vehicles,” IEEE Journal of Oceanic Engineering, Vol.28, No.4, pp. 609-624, 2003.
5. I. Nagai and Y. Tanaka, “Tracking of Floor Image Based on Correlation Processing,” Proc. of the 4th IFAC Symposium on Intelligent Autonomous Vehicles, pp. 151-156, 2001.
Cited by
6 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献