An Implementation of Opportunistic Scheduling for Robotic Assembly
AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH
Pagination or Media Count:
The goal of this thesis is to combine computerized vision and artificial intelligence programming in an application of robotic assembly that will use opportunistic scheduling. Opportunistic scheduling is making a schedule based on current opportunities. A robot provided with a vision system has the capability of recognizing random opportunistic events. However, vision systems have many limitations. A heuristic method of taking pictures is developed to improve object recognition reliability. The robot is given basic assembly knowledge using the production rule methodology, and assembly precedence information using a database of partial order sets. Dynamic state information is also maintained by the program. Parts are delivered randomly on conveyor belts. The robot is given the capability of assembling a mix of products and assembling multiple products concurrently. Thus, the robot can assemble a product in any feasible way and schedule an optimal plan according to the random arrival of parts. A user friendly interface with the robot is developed.