This repository contains tutorial implementations for the paper "Volumetric Ergodic Control", presented at the IEEE International Conference on Robotics and Automation (ICRA) 2026.

Please visit the project website for more information.

The tutorials/ directory contains self-contained Jupyter notebooks with guided examples. For faster performance in Colab, enable GPU: Runtime → Change runtime type → T4 GPU.

• [Notebook | Google Colab] Erasing task with double-integrator dynamics, where the target is a point-cloud distribution and the volumetric state is a 2-D tool controlled from a pivot point.
• [Notebook | Google Colab] Erasing task with double-integrator dynamics, where the target is a Gaussian mixture distribution and the volumetric state is a 2-D tool controlled from a pivot point.
• [Notebook | Google Colab] Search task with a differential-drive robot, where the volumetric state is modeled as a forward-facing solid-state LiDAR.

Point-Cloud Erasing

Gaussian Erasing

Ground Search

All the tutorials are implemented using JAX and LQRax.

The implementations of VEC contained herein are copyright (C) 2025-2026 by Jueun Kwon, Max M. Sun, and Todd Murphey and are distributed under the terms of the GNU General Public License (GPL) version 3 or later. Please see the LICENSE for more information.

Contact: Jueun (jueun@u.northwestern.edu)