Neural Network Architectures for Robotic Applications

This paper proposes a ring VLSI systolic architecture for implementing artificial neural networks (ANN’s) with applications to robotic processing. Key design issues on algorithms, applications, and architectures are examined. A variety of neural networks are considered, including single-layer feedback neural networks, competitive learning networks, and multilayer feed-forward networks. It is demonstrated that the ANN’s are suitable to all three levels of robotic processing applications, including task planning, path planning, and path control levels. For these applications, a programmable systolic array is developed, which can exploit the strength of VLSI to provide intensive and pipelined computing. Both the retrieving and learning phases are integrated in the design. The proposed architecture is more versatile than other existing ANN’s; therefore, it can accommodate all the useful neural networks for robotic processing.