Title :
A programmable array of silicon neurons for the control of legged locomotion
Author :
Tenore, Francesco ; Etienne-Cummings, Ralph ; Lewis, M. Anthony
Author_Institution :
Dept. of Electr. & Comput. Eng., John Hopkins Univ., Baltimore, MD, USA
Abstract :
The biological foundation of most natural locomotory systems is the central pattern generator (CPG). The CPG is a set of neural circuits found in the spinal cord, arranged to produce oscillatory periodic waveforms that activate muscles in a coordinated manner. A 2nd generation VLSI CPG emulator chip - with more and improved neurons, enhanced flexibility, and a higher degree of programmability - has been developed to synchronize oscillators with different frequencies and phases, also produced by the chip, through the coupling of integrate-and-fire (IF) silicon neurons. These oscillators are then used to control the movement of robot´s limbs by using the IF neurons to set a specific phase difference between the oscillators. The chip´s architecture is examined in detail, and the construction and implementation of the artificial neural networks that produce the waveforms required for locomotion is described.
Keywords :
CMOS integrated circuits; VLSI; legged locomotion; motion control; neural chips; programmable circuits; programmable logic arrays; silicon; IF neurons; VLSI CPG emulator chip; artificial neural networks; central pattern generator; chip architecture; legged locomotion; muscles; natural locomotory systems; neural circuits; oscillators synchronization; oscillatory periodic waveforms; programmable silicon neuron array; robot limbs; silicon neurons coupling; spinal cord; Circuits; Frequency synchronization; Legged locomotion; Muscles; Neurons; Oscillators; Robot kinematics; Silicon; Spinal cord; Very large scale integration;
Conference_Titel :
Circuits and Systems, 2004. ISCAS '04. Proceedings of the 2004 International Symposium on
Print_ISBN :
0-7803-8251-X
DOI :
10.1109/ISCAS.2004.1329534