Skip to content
Log in
#graph neural networks Dataset Open access

NEATWork: A Spatial Network Optimizer

Ethan Schaler
Aug 2026 · JPL Data

Abstract

NASA’s planned sustained presence of humans on the Moon will require significant volumes of icy regolith transport across the lunar surface. Algorithmic planning of efficient spatial transport networks connecting multiple terminals and optimizing across a range of objectives remains an open challenge. To address this problem, we developed NEATWork, a spatial network planner. By modifying the well-known NEAT genetic algorithm—traditionally used for the development of neural networks—we present a graph optimization approach that can take into account various spatial and topological objectives. We illustrate our solution using the case of lunar transport network planning, though it can be easily deployed in other network planning scenarios where spatial cost maps can be made and the flexibility for a black box cost function optimization is of interest. We provide two open-source implementations.

DOI
View source

Similar papers

#graph neural networks Oct 2026

Locality-Preserving Graph Laplacian Manifold Learning Based Model Predictive Control for Three-Phase Inverters

This article presents a model predictive control (MPC) strategy for three-phase inverters based on locality preserving projections (LPPs). Unlike conventional machine learning–based MPC approaches that rely on predefined or high-dimensional input features, the proposed LPP-MPC automatically extracts compact, informative representations by preserving the data’s intrinsic geometric structure. This dimensionality reduction enables fast linear control-law evaluation with computational complexity O(1), making the controller well-suited for real-time implementation. Experimental results demonstrate that the LPP-MPC achieves lower total harmonic distortion (THD) and reduced tracking error compared to quadratic-programming MPC under both linear and nonlinear load conditions, and the proposed controller maintains consistently lower THD throughout load transients than other methods such as two-degree-of-freedom MPC. Compared to existing model-free MPC and deep learning neural network, the LPP-MPC has the lowest THD and root mean square error with the least computational time owing to its efficient linear structure and strong generalization capability.

Jianwu Zeng, Lizheng Cheng, V. Winstead et al. · 1 citation

Related blog posts