An architecture which utilizes two artificial neural systems for planning and control of a robotic arm is presented. The first neural network system participates in the trajectory planning and the motion decision-making process. The second neural network system provides the correct sequence of control actions with a high accuracy due to the utilization of an unsupervised/supervised neural network scheme. The utilization of a hybrid hierarchical/distributed organization, supervised/unsupervised learning models, and forward modeling yielded an architecture with capabilities of high level functionality.
X. J. Avula et al., "Planning and Control of a Robotic Manipulator Using Neural Networks," Proceedings of the First IEEE Conference on Control Applications, 1992, Institute of Electrical and Electronics Engineers (IEEE), Jan 1992.
The definitive version is available at https://doi.org/10.1109/CCA.1992.269858
First IEEE Conference on Control Applications, 1992
Chemical and Biochemical Engineering
Keywords and Phrases
Distributed Control; Forward Modeling; Functionality; Hybrid Hierarchical/Distributed Organization; Learning (Artificial Intelligence); Motion Decision-Making Process; Neural Nets; Neural Networks; Path Planning; Robot Arm Control; Robotic Manipulator; Robots; Supervised/Unsupervised Learning Models; Trajectory Planning
Article - Conference proceedings
© 1992 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.