- Collection:
- Atlanta University and Clark Atlanta University Theses and Dissertations
- Title:
- Knowledge-based image management for cooperative tele-assistance, 1996
- Creator:
- DuPont, Versonya M.
- Date of Original:
- 1996-05-01
- Subject:
- Degrees, Academic
Dissertations, Academic - Location:
- United States, Georgia, Fulton County, Atlanta, 33.749, -84.38798
- Medium:
- theses
- Type:
- Text
- Format:
- application/pdf
- Description:
- The research in this thesis presents an approach for improving human visual perception in a tele-assistant environment. This research, which is part of a collaborative project between Clark Atlanta University and Colorado School of Mines, establishes a framework in which a remote semi-autonomous robot, a human supervisor and an intelligent assistant can cooperate in a tele-assistant system. Colorado School of Mines is responsible for housing the robot and developing teleSFX, the system which supports the semi-autonomous robot. Clark Atlanta University is responsible for developing tele VIA, the system which supports the intelligent assistant. TeleVIA serves as an intermediary between the robot and the supervisor and aids the supervisor with visual perception and problem solving. The focus of this research is to use knowledge-based management of remotely sensed data to support human visual perception, especially in the case of a robot failure. Knowledge-based management of the sensor data from the robot, which is a component of tele VIA, is concerned with automatic retrieval of sensor data and knowledge-based enhancement of the retrieved data. Issues of reducing bandwidth requirements for image transmission, retrieving sensor data relevant to the robots failure, selecting appropriate data enhancements, etc. are addressed in this research. Prior to this research, it was determined that tele VIA would be developed for a blackboard-styled architecture and a prototype for a knowledge base was created. Additional knowledge about the robot configuration and failure condition was needed to perform knowledge-based image management. The approach taken in the research was to adapt the knowledge representation of the original system, modify the knowledge base, and design knowledge sources, or ��experts�, to manage the sensor data.
- External Identifiers:
- Metadata URL:
- http://hdl.handle.net/20.500.12322/cau.td:1996_dupont_versonya_m.pdf
- Rights Holder:
- Clark Atlanta University
- Holding Institution:
- Atlanta University Center Robert W. Woodruff Library
- Rights:
-
