NeuroShield
Trainable pattern recognition board
The NeuroShield is an evaluation board featuring one NM500 chip ready to interface with a FPGA via a 28-pin input/output bus. It is designed to help academic and research teams get a quick perception of the NM500 chip’s simplicity of deployment, speed, and low-power consumption.
The NM500 chip is the sole active on-board component. It can learn and recognize digital patterns in real-time. The output pins of the NM500 continually report the category recognized by the neurons.
Block Diagram

SPECIFICATION
Components
NM500 Chip | 576 neurons |
Size | 80 x 53.3 x 19.5 [mm] |
Features
Features | Open Arduino connector for NM500 access |
USB1.1 for NM500 access | |
Arduino/MBed compatibility | |
6-axis Gyro/Acce sensor on board | |
Extendable up to 7 NM500s with NeuroBrick | |
Capable to save/restore using uSD card | |
Support Knowledge Studio |
Hardware Specification
Power | 5V (Shield or USB) |
AI | default 1 NM500 |
FPGA | LCMXO3LF (Lattice) |
BackUp | uSD Card |
Sensor | 6-axis Gyro/Acce sensor |
SDK and Tools
The nepes NM500 Software Development Kit allows you to manage the nepes NM500 neuromorphic chip and implement custom solutions by using libraries, utilities, and programming interfaces provided with the kit.
Related Videos
Gesture Recognition RC Car
Smart LED
door detection