SUPR
Gaussian processes for energy-efficient navigation
Dnr:

NAISS 2024/22-966

Type:

NAISS Small Compute

Principal Investigator:

Jack Sandberg

Affiliation:

Chalmers tekniska högskola

Start Date:

2024-08-08

End Date:

2025-09-01

Primary Classification:

10201: Computer Sciences

Webpage:

Allocation

Abstract

In this project, we investigate the application of Gaussian process models for decision-making under uncertainty. Specifically, we consider a navigation problem where the energy costs are stochastic with unknown distributions and our goal is to find the most energy-efficient route. The project will develop models that can efficiently explore the road network with minimal prior data.