An Fpga Hardware/Software Co-Design Towards Evolvable Spiking Neural Networks for Robotics Application

Abstract

This paper presents an approach that permits the effective hardware realization of a novel Evolvable Spiking Neural Network (ESNN) paradigm on Field Programmable Gate Arrays (FPGAs). The ESNN possesses a hybrid learning algorithm that consists of a Spike Timing Dependent Plasticity (STDP) mechanism fused with a Genetic Algorithm (GA). The design and implementation direction utilizes the latest advancements in FPGA technology to provide a partitioned hardware/software co-design solution. The approach achieves the maximum FPGA flexibility obtainable for the ESNN paradigm. The algorithm was applied as an embedded intelligent system robotic controller to solve an autonomous navigation and obstacle avoidance problem.

Testing of Information Condensation in a Model Reverberating Spiking Neural Network

“…In the context of cognitive physiology, the process of reduction of information aimed at conceptual representation/recognition of external objects is known as information condensation, [21]. Usually pattern recognition phenomenon, which is closely related to information condensation, is considered in parallel with training, see [12,17,24,28,30,31,39], learning, [1,11,13,21], or other plasticity, [16], in the corresponding network. In a biological network, the learning mechanism involves biosynthesis [27] and is therefore very slow process, which requires seconds or minutes, [3].…”

Evolving spiking neural network—a survey

“…This model, if further developed, can facilitate efficient real time applications such as: EEG pattern recognition for BCI; fMRI pattern recognition; neurorehabilitation robotics; neuro-prosthetics; cognitive robots; personalized modeling for the prognosis of fatal events such as stroke and degenerative progression of brain disease, such as AD. A combination of the fast evolving one-pass learning of eSNN with the automatic parameter adaptation of a genetically controlled recurrent SNN could open the path for numerous relevant real-world applications including neuromorphic computation systems [26,27,39].…”

Evolving Spiking Neural Networks and Neurogenetic Systems for Spatio- and Spectro-Temporal Data Modelling and Pattern Recognition

“…-jAER (http://jaer.wiki.sourceforge.net) [23]; -Software simulators, such as Brian [16], Nestor, NeMo [79],etc; -Silicon retina camera [23]; -Silicon cochlea [107]; -SNN hardware realisation of LIFM and SDSP [47][48][49][50]; -The SpiNNaker hardware/software environment [89,116]; -FPGA implementations of SNN [56]; -The IBM LIF SNN chip, recently announced. Fig.14 shows a hypothetical engineering system using some of the above tools (from [47,25]).…”