Neuromorphic quadratic programming for efficient and scalable model predictive control

TL;DR

This paper introduces a neuromorphic approach to solve convex quadratic programming problems efficiently on the Loihi 2 chip, enabling real-time model predictive control for robotics with significantly reduced energy consumption.

Contribution

It presents a novel neuromorphic quadratic programming method tailored for Loihi 2, demonstrating superior energy efficiency and speed in robotic control applications compared to traditional solvers.

Findings

Over two orders of magnitude energy-delay reduction compared to OSQP.

Solution times under ten milliseconds for various problem sizes.

Effective application to quadruped robot MPC.

Abstract

Applications in robotics or other size-, weight- and power-constrained autonomous systems at the edge often require real-time and low-energy solutions to large optimization problems. Event-based and memory-integrated neuromorphic architectures promise to solve such optimization problems with superior energy efficiency and performance compared to conventional von Neumann architectures. Here, we present a method to solve convex continuous optimization problems with quadratic cost functions and linear constraints on Intel's scalable neuromorphic research chip Loihi 2. When applied to model predictive control (MPC) problems for the quadruped robotic platform ANYmal, this method achieves over two orders of magnitude reduction in combined energy-delay product compared to the state-of-the-art solver, OSQP, on (edge) CPUs and GPUs with solution times under ten milliseconds for various problem sizes. These results demonstrate the benefit of non-von-Neumann architectures for robotic control applications.