Graphical integration of robot programming and sequence planning for mechanical assembly

Gu, Yunqing (1998) Graphical integration of robot programming and sequence planning for mechanical assembly. Masters thesis, Memorial University of Newfoundland.

[English] PDF (Migrated (PDF/A Conversion) from original format: (application/pdf)) - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (13MB) [English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. (Original Version)

Abstract

A major problem plaguing programmable robot-based assembly systems today is the lack of communication between robot programming and assembly planning. Whereas a robot learns to perform assembly tasks through programming, it assembles mechanical parts into a product by following the assembly sequence determined by planning. Robot programming and assembly planning have been dealt with mostly as two separate research topics. Nevertheless, the internal representation of parts and their topological and geometrical relationship required by planning could be automatically synthesized from the information obtained in programming. This observation becomes a view in advanced assembly automation systems that integrate robot programming and assembly planning for better performance and higher efficiency. -- This thesis addresses the issues of graphical integration of the robot programming process and the sequence planning process for mechanical assembly. It treats the robotic assembly sequencing as a motion-planning problem with special constraints. During the programming phase, the order of effective assembly actions is built once a robot effectively performs an assembly task. As a result, the information of an assembly scene and ordered robotic actions can be produced and stored. In the unified system presented in the thesis, the sequence planning process directly retrieves this information for automatic and fast planning. Since analyzing ordered robotic actions and executing partial geometric checks are usually much fast than performing full geometric checks, this integrated approach offers significant computational advantages in comparison with other 'full-automatic' planning approaches. In addition, the approach feeds all the feasible sequences generated from the planning process to the programming process, and provides the 'automatic re-programming' feature to the programming process. As a link between the programming and planning processes, this integrated approach presents a richer form of communication between them, which is necessary to efficiently solve real-world robot-based automation problem. The validity of this approach is justified with a prototype system, namely INTEG. Experimental results are also given in the thesis that include examples varying from simple to complex assembly tasks.

Actions (login required)

View Item