English

Neuromorphic quadratic programming for efficient and scalable model predictive control

Neural and Evolutionary Computing 2024-06-21 v3 Emerging Technologies Robotics

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.

Keywords

Cite

@article{arxiv.2401.14885,
  title  = {Neuromorphic quadratic programming for efficient and scalable model predictive control},
  author = {Ashish Rao Mangalore and Gabriel Andres Fonseca Guerra and Sumedh R. Risbud and Philipp Stratmann and Andreas Wild},
  journal= {arXiv preprint arXiv:2401.14885},
  year   = {2024}
}

Comments

Accepted by Robotics and Automation Magazine on 25-05-2024

R2 v1 2026-06-28T14:28:10.404Z