Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

GRaD-Nav++: Vision-Language Model Enabled Visual Drone Navigation with Gaussian Radiance Fields and Differentiable Dynamics

About

Autonomous drones capable of interpreting and executing high-level language instructions in unstructured environments remain a long-standing goal. Yet existing approaches are constrained by their dependence on hand-crafted skills, extensive parameter tuning, or computationally intensive models unsuitable for onboard use. We introduce GRaD-Nav++, a lightweight Vision-Language-Action (VLA) framework that runs fully onboard and follows natural-language commands in real time. Our policy is trained in a photorealistic 3D Gaussian Splatting (3DGS) simulator via Differentiable Reinforcement Learning (DiffRL), enabling efficient learning of low-level control from visual and linguistic inputs. At its core is a Mixture-of-Experts (MoE) action head, which adaptively routes computation to improve generalization while mitigating forgetting. In multi-task generalization experiments, GRaD-Nav++ achieves a success rate of 83% on trained tasks and 75% on unseen tasks in simulation. When deployed on real hardware, it attains 67% success on trained tasks and 50% on unseen ones. In multi-environment adaptation experiments, GRaD-Nav++ achieves an average success rate of 81% across diverse simulated environments and 67% across varied real-world settings. These results establish a new benchmark for fully onboard Vision-Language-Action (VLA) flight and demonstrate that compact, efficient models can enable reliable, language-guided navigation without relying on external infrastructure.

Qianzhong Chen, Naixiang Gao, Suning Huang, JunEn Low, Timothy Chen, Jiankai Sun, Mac Schwager• 2025

Related benchmarks

TaskDatasetResultRank
Multi-task Drone NavigationMulti-task Drone Simulation 8 tasks (train)
Reward5.07e+3
4
Multi-task Drone NavigationMulti-task Drone Simulation 4 tasks (untrained)
Reward4.45e+3
4
Multi-task GeneralizationReal Hardware 8 tasks (train)
Stage 1 Success Rate0.875
4
Multi-task GeneralizationReal Hardware 4 tasks (untrained)
Stage 1 Success Rate9
4
Showing 4 of 4 rows

Other info

Follow for update