Overview
LeHome is a robot-learning competition entry that teaches a bimanual robot to fold laundry in simulation, training imitation and VLA policies to handle a range of garments.
Back to projectsRobot Learning
Partnered with Kenta Terasaki
LeHome Laundry Folding Simulation
Isaac Sim and Isaac Lab training pipelines for bimanual simulated manipulation in the LeHome Robotics Simulation Competition.
Project Summary
What I built
A short look at the build, the main technical choices, and the pieces I iterated along the way.
What I Built
I built Isaac Sim / Isaac Lab training pipelines, trained ACT and diffusion policies with PyTorch on A100 GPUs, and fine-tuned VLA models including SmolVLA and Pi0-FAST with Hugging Face Accelerate.
Technical Highlights
The core technical work is scalable robot-learning infrastructure for simulated manipulation across garment categories: short sleeves, long sleeves, shorts, and pants, with robot-view videos for each.
Demos
Demos and design process
Videos and images from the current build, earlier iterations, and the small details that shaped the project.
video
Short-Sleeve Folding
video
Long-Sleeve Folding
video
Pants Folding
video