Model-based vision and intelligent task scheduling for autonomous human-type robot arm

This paper describes the concepts and methodologies for model-based vision and intelligent task scheduling for an autonomous human-type robot arm system. The vision and task scheduling method is based on Cognitive Psychology, i.e., a human's way of vision and scheduling is applied. The robot arm which characterizes our approach is similar to a human arm in shape, mechanism, and functions, and has local intelligence for manipulations with 178 tactile sensors. The vision system receives stereo images through two CCD cameras and creates a world model as a mental image of the scene. The intelligent task scheduling system sets a task goal according to a user's request and generates a sequence of tasks to achieve the goal based on the world model.