World modelling and multi-sensor integration for a mobile robot
In this paper we present a strategy for the reduction of systematic errors through the fusion of ultrasonic range and visual intensity information. This is done within a framework consisting of several world models, integration procedures, and sensor models which include appropriate sensor-specific signal and information processing algorithms. Among the world models are linked lists of geometric features extracted from the intensity images, and 2D Cartesian maps built from the sonar range data. Spatial reasoning was used in the fusion of ultrasonic data from multiple scans, and in the analysis of visual edge information following exchange of range and angular information across the sensor domains. 11 refs., 9 figs.
- Research Organization:
- Oak Ridge National Lab., TN (USA)
- DOE Contract Number:
- AC05-84OR21400
- OSTI ID:
- 6358132
- Report Number(s):
- CONF-890304-2; ON: DE89005124
- Resource Relation:
- Conference: 3. topical meeting on robotics and remote systems, Charleston, SC, United States, 13 Mar 1989; Other Information: Portions of this document are illegible in microfiche products
- Country of Publication:
- United States
- Language:
- English
Similar Records
An intelligent integrated sensor system for the ORNL mobile robot
Multi-sensor integration for autonomous robots in nuclear power plants