A New Realm of Artificial Intelligence Control for Real-World Autonomous Mobile Robotics: Behavior Based Autonomous Systems

John B. Watson

Uncertainty. Dealing with it is perhaps the largest problem facing autonomous mobile robotics. While it is commonly held that engineering problems will be solved when enough effort is applied, control systems, though constantly evolving, cannot adequately address many situations autonomous robots encounter or the complexities of defining and achieving goals. If it were possible, it would be optimal to abstract many of the qualities of the human mind, program them into a compact module, and place them in a mobile robot. Current exploration into a more human-like autonomous system may have a positive impact on autonomous mobile robotics. An experimental artificial intelligence model which is well suited for robotics control is a behaviorbased autonomous system, developed over the past few years (Watson 1991). The goal-oriented system specifically designed for real-time interaction with its environment and can adapt to constantly changing situations. This is achieved by modeling the algorithm after human behavioral qualities. The first implementation of the model, IAM-1, was written to demonstrate some of the qualities of the behavior-based autonomous system theory. IAM-2, the latest version of the application is being designed to have the capability to simulate an autonomous robotics control system.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.