Hybrid Control of a Robotic Manipulator by the Neural Network Model. 5th Report, Hierarchical Hybrid Neuromorphic Control System.
Takanori Shibata, Toshio Fukuda, S. Shiotani, Masatoshi Tokita, Toyokazu Mitsuoka · TRANSACTIONS OF THE JAPAN SOCIETY OF MECHANICAL ENGINEERS Series C · 1991
This paper presents a "Hierarchical Hybrid Neuromorphic Control System". This system comprises two levels: a "learning" level and an "adaptation" level. Neural networks are employed for both the long-term "learning" of the control process and the short-term "adaptation" of the dynamic process. The "learning" level has a hierarchical structure and is used for strategic planning of the robotic manipu1ator in conjunction with the knowledge data base syseem, "Neural Knowledge Data Base (NKDB)", in order to enlarge the range of the "adaptation". The NKDB can infer an unknown fact from a priori knowledge for strategic planning, and is updated by the recent information from the adaptation level through the long term learning process. On the other hand, "adaptation" is used for the adjustment of the control law to the current status of the dynamic process. The initial states of the adaptation level are given by the NKDB. The motion controller of the adaptation level is based on the neural network including time-delay elements in the hidden layer, and is particularly valuable in nonlinear dynamical systems with unknown parameters.