Adaptive Robotic Arm Control with a Spiking Recurrent Neural Network on a Digital Accelerator

TL;DR

This paper presents an adaptation of a recurrent spiking neural network chip for robotic arm control, demonstrating high-speed event processing and preserved accuracy in an embedded system setup.

Contribution

It introduces a new implementation of the ReckOn spiking neural network chip on a Xilinx MPSoC for robotic control applications.

Findings

Peak performance of 3.8 million events per second

Preserved accuracy in robotic arm control scenario

Enhanced deployment flexibility in embedded systems

Abstract

With the rise of artificial intelligence, neural network simulations of biological neuron models are being explored to reduce the footprint of learning and inference in resource-constrained task scenarios. A mainstream type of such networks are spiking neural networks (SNNs) based on simplified Integrate and Fire models for which several hardware accelerators have emerged. Among them, the ReckOn chip was introduced as a recurrent SNN allowing for both online training and execution of tasks based on arbitrary sensory modalities, demonstrated for vision, audition, and navigation. As a fully digital and open-source chip, we adapted ReckOn to be implemented on a Xilinx Multiprocessor System on Chip system (MPSoC), facilitating its deployment in embedded systems and increasing the setup flexibility. We present an overview of the system, and a Python framework to use it on a Pynq ZU platform. We validate the architecture and implementation in the new scenario of robotic arm control, and show how the simulated accuracy is preserved with a peak performance of 3.8M events processed per second.