Next: Programming Up: Planning Previous: Multiple Robot Workcell

Robot Action Planning using State Matrix Representation

Authors: [tex2html_wrap4358]S.M. Noorhosseini, A.S. Malowany

Investigator username: malowany

Category: robotics

Subcategory: planning

This research addresses the action planning problem in an intelligent robotic system which is considered as a pre-requisite to the autonomous behavior of the system. A new representation scheme called State Matrix representation which uses inter-relationships of the world elements to represent the world as well as a new planning algorithm based on this representation scheme were developed. The planning algorithm can solve non linear monotone planning problems including most assembly planning problems. A planning system based on this planning methodology has been implemented. This system includes a graphic user interface which allows the user to define a task. The action planner then determines the sequence of actions to be performed in order to accomplish the task.


baron@cim.mcgill.ca