Intelligent Task Planning and Execution on Assembly Robotics
The aim of our work is the implementation of an Intelligent Robotic System able to deal with assembly tasks.
While some planning algorithms are too heavy to be used in real on-line robotic assembly tasks, our approach is very efficient taking only a few seconds to give the desired high-level symbolic plan. This efficiency is mainly due to the choice of good heuristics and the early pruning of unecessary paths in the planning search tree.
Once generated, the high level plan must be translated into robot instructions. At this level, some intelligence is needed to garantee the feasibility of the robot actions. This is the case of the object’s grasping, regrasping to change from one stable position to another, collision avoidance and free space management. Sometimes during execution it is felt that a correctly generated plan presents some difficulties in order to be executed. This happens when some specific object locations and configurations are found. In those cases it is said that we are in the presence of geometric constraints.
The modules of our Intelligent Robotic System are implemented in PROLOG and C languages under UNIX environment. The generated plans are executed by a RENAULT-APRA robot controlled by a Personal Computer, using LM language. The Personal Computer and the UNIX system are connected by an Ethernet network.
KeywordsGeometric Constraint Collision Avoidance Assembly Task Grasp Position Robot Hand
Unable to display preview. Download preview PDF.
- [Fikes-72]STRIPS: A New Approach to the application of Theorem Proving to Problem Solving Fikes, R. E and Nilsson, N. J. Artificial Intelligence, Vol. 2, 1972Google Scholar
- [Hutchinson-90]Spar: A Planner that satisfies Operational and Geometric Goals in Uncertain Environments Hutchinson, S. H. and Kak, A. C. AI Magazine, Spring 1990Google Scholar
- [Lozano-Perez-89]Task-Level Planning of Pick-and-Place Robot Motions Lozano-Perez, T., Jones, J.L., Mazer, E., O’Donnell, P.A. Computer IEEE, March 1989, pp 21–29Google Scholar
- [Newell-63]GPS: A Program that simulates Human Thought Newell, A. and Simon, H. in Computer and Thought, Feigenbaum, E. and Feldman, J. N.Y., McGraw-Hill, 1963Google Scholar
- [Oliveira-91]Multi-Agent Cooperation on Assembly Robotics Oliveira, E. and Ramos, C. EURISCON-91 - Corfu, Greece, 1991Google Scholar
- [Ramos-89]Cooperation Between Vision and Planning Agents in a simple Robotic Environment Ramos, C. and Oliveira, E. Intelligent Autonomous Systems 2, Amsterdam, 1989Google Scholar
- [Ramos-9I]The Generation of Efficient High Level Plans and the Robot World Representation in a Cooperative Community of Robotic Agents Ramos, C. and Oliveira, E. ICAR-91–5th International Conference on Advanced Robotics, Pisa, 1991 IEEE pressGoogle Scholar
- [Sacerdoti -77]A Structure for Plans and Behavior Sacerdoti, E. D. Elsevier, New York, 1977Google Scholar
- [Tate-77]Generating Project Networks Tate, A. 5th International Joint Conference on AI, 1977Google Scholar
- [V alavanis-91]