Neural dynamics emulation using neuromorphic chips

Abstract: Dynamic Adaptive Neural Network Array (DANNA) is a neuromorphic implementation meant for conventional digital hardware that is currently implemented on field programmable gate array (FPGA) with a VLSI
implementation in progress [1, 2]. In this demo, we will demonstrate the FPGA implementation of DANNA running in real-time on at least three applications: pole balancing, one-dimensional navigation with gravity
(a game based on the mobile application Flappy Bird [4]), and two-dimensional navigation with obstacle avoidance. We will also demonstrate a visualization of the DANNA neuromorphic implementation running in real-time alongside the simulation of the hardware on the FPGA.

4819 Emperor Blvd, Suite 300 Durham, NC 27703 Voice: (919) 941-9400 Fax: (919) 941-9450