Developmental Robots - A New Paradigm
Abstract
It has been extremely challenging for humans to program a robot to such a degree that it acts properly in a typical unknown human environment. This is especially true for a humanoid robot due to the very large number of redundant degrees of freedom and large number of sensors that are required for a humanoid to work safely and effectively in a human environment. How can researchers address this fundamental problem? Motivated by human mental development from infancy to adulthood, the author presents a theory, an architecture, and some experimental results showing how one can enable a robot to develop its mind autonomously through online, real-time interactions with its environment. Humans mentally "raise" the robot through "robot sitting" and "robot schools" instead of task-specific robot programming. The complex and changing nature of the human environment has made the issue of autonomous mental development of robots -- the way the human mind develops -- more important than ever. Why are robots that are programmed using human designed, task-specific representation unable to do well in complex, changing, or partially unknown or totally unknown environments? What are the self-organization schemes that robots can use to autonomously develop their mental skills through interactions with the environment? Is it more advantageous to enable robots to autonomously develop their mental skills than to program robots using human-specified, task-specific representation? Although robot mental development is very much a new concept, a lot of well-known self-organization tools can be used in designing a developmental robot. This paper summarizes recent investigations on this issue and provides some answers to the above questions. The author first outlines previous and current projects related to robot mental development conducted by his group, and then presents a theory of autonomous mental development of robots. Experimental results on the SAIL robot are included.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2005
- Accession Number
- ADA437286
Entities
People
- Juyang Weng
- Yilu Zhang
Organizations
- Michigan State University