NGRC on SoC for online learning of dynamical systems

About

Predicting the evolution and control of dynamic systems in real time remains a demanding task for conventional methods that depend on accurate models and intensive continuous optimization, which often breaks under uncertainty, noise or delays. Reservoir Computing (RC) provides a lightweight alternative, needing only small training sets and simple inference. We present a low‑cost, low‑power platform that couples next‑generation RC (NGRC) with continuous online learning to forecast long horizons of the Lorenz attractor. The system runs on a Zynq‑7000 SoC: the FPGA accelerates reservoir inference loop, while the ARM CPU manages data acquisition, weight updates and control. With naïve implementation a latency of 148.5 ms on CPU side + 1.5 ms on FPGA is achieved. Results demonstrate that NGRC on modest SoC hardware can deliver energy‑efficient, real‑time predictions of chaotic systems, enabling edge deployment in autonomous vehicles, robotics, satellites and high‑energy physics.

00:00:00 Slide 1

00:00:57 Slide 2

00:01:28 Slide 3

00:02:11 Slide 4

00:02:33 Slide 5

00:04:02 Slide 6

00:06:07 Slide 7

00:07:06 Slide 8

00:08:13 Slide 9

00:08:32 Slide 10

00:09:17 Slide 11