Gaussian Splatting to Real World Flight Navigation Transfer with Liquid Networks

TL;DR

This paper introduces a novel sim-to-real transfer method for quadrotor navigation using Gaussian Splatting and Liquid neural networks, enabling robust real-world flight performance from simulation data.

Contribution

It combines Gaussian Splatting with Liquid neural networks to improve generalization and robustness in sim-to-real quadrotor navigation tasks.

Findings

Successful transfer of navigation skills from simulation to real-world quadrotor flights.

Robust performance maintained under drastic environmental and distribution changes.

Generalization to outdoor multi-step maneuvers from indoor training data.

Abstract

Simulators are powerful tools for autonomous robot learning as they offer scalable data generation, flexible design, and optimization of trajectories. However, transferring behavior learned from simulation data into the real world proves to be difficult, usually mitigated with compute-heavy domain randomization methods or further model fine-tuning. We present a method to improve generalization and robustness to distribution shifts in sim-to-real visual quadrotor navigation tasks. To this end, we first build a simulator by integrating Gaussian Splatting with quadrotor flight dynamics, and then, train robust navigation policies using Liquid neural networks. In this way, we obtain a full-stack imitation learning protocol that combines advances in 3D Gaussian splatting radiance field rendering, crafty programming of expert demonstration training data, and the task understanding capabilities of Liquid networks. Through a series of quantitative flight tests, we demonstrate the robust transfer of navigation skills learned in a single simulation scene directly to the real world. We further show the ability to maintain performance beyond the training environment under drastic distribution and physical environment changes. Our learned Liquid policies, trained on single target manoeuvres curated from a photorealistic simulated indoor flight only, generalize to multi-step hikes onboard a real hardware platform outdoors.