GaussFly: Contrastive Reinforcement Learning for Visuomotor Policies in 3D Gaussian Fields

• URL: http://arxiv.org/abs/2604.05062v1
• Date: Mon, 06 Apr 2026 18:10:52 GMT
• Title: GaussFly: Contrastive Reinforcement Learning for Visuomotor Policies in 3D Gaussian Fields
• Abstract summary: GaussFly is a novel framework that explicitly decouples representation learning from policy optimization.<n>It enables robust and zero-shot policy transfer to unseen real-world environments with complex textures.
• Score: 10.10713417247828
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Learning visuomotor policies for Autonomous Aerial Vehicles (AAVs) relying solely on monocular vision is an attractive yet highly challenging paradigm. Existing end-to-end learning approaches directly map high-dimensional RGB observations to action commands, which frequently suffer from low sample efficiency and severe sim-to-real gaps due to the visual discrepancy between simulation and physical domains. To address these long-standing challenges, we propose GaussFly, a novel framework that explicitly decouples representation learning from policy optimization through a cohesive real-to-sim-to-real paradigm. First, to achieve a high-fidelity real-to-sim transition, we reconstruct training scenes using 3D Gaussian Splatting (3DGS) augmented with explicit geometric constraints. Second, to ensure robust sim-to-real transfer, we leverage these photorealistic simulated environments and employ contrastive representation learning to extract compact, noise-resilient latent features from the rendered RGB images. By utilizing this pre-trained encoder to provide low-dimensional feature inputs, the computational burden on the visuomotor policy is significantly reduced while its resistance against visual noise is inherently enhanced. Extensive experiments in simulated and real-world environments demonstrate that GaussFly achieves superior sample efficiency and asymptotic performance compared to baselines. Crucially, it enables robust and zero-shot policy transfer to unseen real-world environments with complex textures, effectively bridging the sim-to-real gap.

Related papers

• GASE: Gaussian Splatting-Based Automated System for Reconstructing Embodied-Simulation Environments [63.222920701596564]
  Training embodied agents in the real world requires skilled operators and expensive hardware.<n> Simulation environments offer a compelling alternative by enabling large-scale, cost-effective data augmentation.<n>Fastly constructing high-fidelity simulation scenes with a minimal sim-to-real gap has become a critical objective in robot learning.
  arXiv  Detail & Related papers  (2026-06-16T05:00:42Z)
• Trans2Occ: Voxel Occupancy Estimation and Grasp for Transparent Objects from Simulation to Reality [125.76063573012853]
  We present a framework for transparent object perception and manipulation based on single-view RGB input.<n>Our approach predicts voxel-space occupancy directly from a single image, providing a geometry-aware representation that supports downstream robotic grasping.
  arXiv  Detail & Related papers  (2026-06-01T06:59:13Z)
• GS-Playground: A High-Throughput Photorealistic Simulator for Vision-Informed Robot Learning [55.272458304179025]
  GS-Playground is a multi-modal simulation framework designed to accelerate end-to-end perceptual learning.<n>Our system achieves a breakthrough throughput of 104 FPS at 640x480 resolution, significantly lowering the barrier for large-scale visual RL.
  arXiv  Detail & Related papers  (2026-04-28T10:05:39Z)
• R-PGA: Robust Physical Adversarial Camouflage Generation via Relightable 3D Gaussian Splatting [59.29134814041703]
  Physical adversarial camouflage poses a severe security threat to autonomous driving systems.<n>Current methods fail to generalize across diverse geometric (e.g., viewing configurations) and radiometric variations.<n>We propose the Relightable Physical 3D Gaussian Splatting framework (R-PGA)
  arXiv  Detail & Related papers  (2026-03-27T04:38:04Z)
• Zero-Shot UAV Navigation in Forests via Relightable 3D Gaussian Splatting [11.31291385822484]
  UAV navigation in unstructured outdoor environments using passive monocular vision is hindered by the substantial visual domain gap between simulation and reality.<n>We propose a novel end-to-end reinforcement learning framework designed for effective zero-shot transfer to unstructured outdoors.<n>We show that a lightweight quadrotor achieves robust, collision-free navigation in complex forest environments at speeds up to 10 m/s.
  arXiv  Detail & Related papers  (2026-02-06T15:51:03Z)
• ExoGS: A 4D Real-to-Sim-to-Real Framework for Scalable Manipulation Data Collection [27.18540416686717]
  ExoGS is a robot-free 4D Real-to-Sim-to-Real framework.<n>It captures both static environments and dynamic interactions in the real world and transfers them seamlessly to a simulated environment.<n>It provides a new solution for scalable manipulation data collection and policy learning.
  arXiv  Detail & Related papers  (2026-01-26T16:04:12Z)
• Opening the Sim-to-Real Door for Humanoid Pixel-to-Action Policy Transfer [59.02729900344616]
  GPU-accelerated, photorealistic simulation has opened a scalable data-generation path for robot learning.<n>We develop a teacher-student-bootstrap learning framework for vision-based humanoid loco-manipulation.<n>This represents the first humanoid sim-to-real policy capable of diverse articulated loco-manipulation using pure RGB perception.
  arXiv  Detail & Related papers  (2025-11-30T20:07:13Z)
• EmbodiedSplat: Personalized Real-to-Sim-to-Real Navigation with Gaussian Splats from a Mobile Device [33.22697339175522]
  Embodied AI predominantly relies on simulation for training and evaluation.<n>Sim-to-real transfer remains a major challenge.<n>EmbodiedSplat is a novel approach that personalizes policy training.
  arXiv  Detail & Related papers  (2025-09-22T07:22:31Z)
• HoliGS: Holistic Gaussian Splatting for Embodied View Synthesis [59.25751939710903]
  We propose a novel deformable Gaussian splatting framework that addresses embodied view synthesis from long monocular RGB videos.<n>Our method leverages invertible Gaussian Splatting deformation networks to reconstruct large-scale, dynamic environments accurately.<n>Results highlight a practical and scalable solution for EVS in real-world scenarios.
  arXiv  Detail & Related papers  (2025-06-24T03:54:40Z)
• VR-Robo: A Real-to-Sim-to-Real Framework for Visual Robot Navigation and Locomotion [25.440573256776133]
  This paper presents a Real-to-Sim-to-Real framework that generates and physically interactive "digital twin" simulation environments for visual navigation and locomotion learning.
  arXiv  Detail & Related papers  (2025-02-03T17:15:05Z)
• EnvGS: Modeling View-Dependent Appearance with Environment Gaussian [78.74634059559891]
  EnvGS is a novel approach that employs a set of Gaussian primitives as an explicit 3D representation for capturing reflections of environments.<n>To efficiently render these environment Gaussian primitives, we developed a ray-tracing-based reflection that leverages the GPU's RT core for fast rendering.<n>Results from multiple real-world and synthetic datasets demonstrate that our method produces significantly more detailed reflections.
  arXiv  Detail & Related papers  (2024-12-19T18:59:57Z)
• MCGS: Multiview Consistency Enhancement for Sparse-View 3D Gaussian Radiance Fields [100.90743697473232]
  Radiance fields represented by 3D Gaussians excel at synthesizing novel views, offering both high training efficiency and fast rendering.<n>Existing methods often incorporate depth priors from dense estimation networks but overlook the inherent multi-view consistency in input images.<n>We propose a view synthesis framework based on 3D Gaussian Splatting, enabling scene reconstruction from sparse views.
  arXiv  Detail & Related papers  (2024-10-15T08:39:05Z)
• Transferable Active Grasping and Real Embodied Dataset [48.887567134129306]
  We show how to search for feasible viewpoints for grasping by the use of hand-mounted RGB-D cameras. A practical 3-stage transferable active grasping pipeline is developed, that is adaptive to unseen clutter scenes. In our pipeline, we propose a novel mask-guided reward to overcome the sparse reward issue in grasping and ensure category-irrelevant behavior.
  arXiv  Detail & Related papers  (2020-04-28T08:15:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.