Design of hardware circuit based on a neural network model for rapid detection of center of gravity position

This paper proposes a rapid detection method for the center of gravity based on a neural network model. It is suitable for the rapid response requirement such as attitude control of a gait robot or real time torque control of a running car. The proposed method detects the center of gravity position on a straight line by using only the hardware circuit composing of common electronic devices instead of software, microprocessor and AD converter. The circuit employs some neural based comparators without the learning function to simplify the circuit structure. The detection circuit using some parallel processing neural comparators rapidly detects the center of gravity position on a straight line. In this paper, the circuit is designed and fabricated with electronic devices, and the circuit experiment shows the performance of the position detection.

[1]  Leon O. Chua,et al.  Memristor Bridge Synapse-Based Neural Network and Its Learning , 2012, IEEE Transactions on Neural Networks and Learning Systems.

[2]  Tetsuya Asai,et al.  A subthreshold MOS neuron circuit based on the Volterra system , 2003, IEEE Trans. Neural Networks.

[3]  Reinhard Eckhorn,et al.  Neural mechanisms of scene segmentation: recordings from the visual cortex suggest basic circuits for linking field models , 1999, IEEE Trans. Neural Networks.

[4]  Noritaka Shigei,et al.  Hardware Implementation and Estimation Property of Neural Network for Discharge Capacity Prediction of Secondary Battery , 2012 .

[5]  Giacomo Indiveri,et al.  Real-Time Classification of Complex Patterns Using Spike-Based Learning in Neuromorphic VLSI , 2009, IEEE Transactions on Biomedical Circuits and Systems.

[6]  Eduardo Ros,et al.  Real-time computing platform for spiking neurons (RT-spike) , 2006, IEEE Trans. Neural Networks.

[7]  Gert Cauwenberghs,et al.  Dynamically Reconfigurable Silicon Array of Spiking Neurons With Conductance-Based Synapses , 2007, IEEE Transactions on Neural Networks.

[8]  Antonette M. Logar,et al.  An iterative method for training multilayer networks with threshold functions , 1994, IEEE Trans. Neural Networks.

[9]  Leon O. Chua,et al.  The simplicial neural cell and its mixed-signal circuit implementation: an efficient neural-network architecture for intelligent signal processing in portable multimedia applications , 2002, IEEE Trans. Neural Networks.

[10]  Tomoyuki Miyazaki,et al.  Logic circuits of variable function using neural networks and their method of calculating synaptic weights , 2000 .

[11]  L. Gatet,et al.  Analog Neural Network Implementation for a Real-Time Surface Classification Application , 2008, IEEE Sensors Journal.

[12]  Tadashi Shibata,et al.  Analog Soft-Pattern-Matching Classifier using Floating-Gate MOS Technology , 2001, NIPS.

[13]  Steven R. Young,et al.  On the Impact of Approximate Computation in an Analog DeSTIN Architecture , 2014, IEEE Transactions on Neural Networks and Learning Systems.