Few-shot Scooping Under Domain Shift via Simulated Maximal Deployment Gaps

TL;DR

This paper introduces a novel vision-based adaptive scooping strategy using deep kernel Gaussian processes trained with simulated deployment gaps, enabling autonomous landers to effectively sample extraterrestrial terrains despite domain shifts.

Contribution

It proposes a deep kernel calibration method with maximal deployment gaps for improved few-shot adaptation to out-of-distribution terrains in planetary sampling missions.

Findings

Significantly outperforms non-adaptive methods in out-of-distribution terrains.

Demonstrates zero-shot transfer to NASA OWLAT platform.

Enables high-quality scooping after few attempts in simulated extraterrestrial environments.

Abstract

Autonomous lander missions on extraterrestrial bodies need to sample granular materials while coping with domain shifts, even when sampling strategies are extensively tuned on Earth. To tackle this challenge, this paper studies the few-shot scooping problem and proposes a vision-based adaptive scooping strategy that uses the deep kernel Gaussian process method trained with a novel meta-training strategy to learn online from very limited experience on out-of-distribution target terrains. Our Deep Kernel Calibration with Maximal Deployment Gaps (kCMD) strategy explicitly trains a deep kernel model to adapt to large domain shifts by creating simulated maximal deployment gaps from an offline training dataset and training models to overcome these deployment gaps during training. Employed in a Bayesian Optimization sequential decision-making framework, the proposed method allows the robot to perform high-quality scooping actions on out-of-distribution terrains after a few attempts, significantly outperforming non-adaptive methods proposed in the excavation literature as well as other state-of-the-art meta-learning methods. The proposed method also demonstrates zero-shot transfer capability, successfully adapting to the NASA OWLAT platform, which serves as a state-of-the-art simulator for potential future planetary missions. These results demonstrate the potential of training deep models with simulated deployment gaps for more generalizable meta-learning in high-capacity models. Furthermore, they highlight the promise of our method in autonomous lander sampling missions by enabling landers to overcome the deployment gap between Earth and extraterrestrial bodies.