# FlashNav: Training Deployable Robot Navigation Policies in Seconds

arXiv: https://arxiv.org/abs/2606.15846 · HTML: https://mlsanity.com/paper/2606.15846/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot navigation and UAV control, one of 15 directions in Deep reinforcement learning methods.
- Builds on 22 of its 33 references that are in the map.
- First posted on arXiv 14 Jun 2026.

## Placed in

- Deep reinforcement learning methods / [RL for robot navigation and UAV control](https://mlsanity.com/deeprl/rl-for-robot-navigation-and-uav-control/)

## Builds on (most cited first, 22 of 22)

- [Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning](https://mlsanity.com/paper/2108.10470/) (2021)
- [VMAS: A Vectorized Multi-Agent Simulator for Collective Robot Learning](https://mlsanity.com/paper/2207.03530/) (2022)
- [EnvPool: A Highly Parallel Reinforcement Learning Environment Execution Engine](https://mlsanity.com/paper/2206.10558/) (2022)
- [FastTD3: Simple, Fast, and Capable Reinforcement Learning for Humanoid Control](https://mlsanity.com/paper/2505.22642/) (2025)
- [Reinforced Imitation: Sample Efficient Deep Reinforcement Learning for Map-less Navigation by Leveraging Prior Demonstrations](https://mlsanity.com/paper/1805.07095/) (2018)
- [DRL-VO: Learning to Navigate Through Crowded Dynamic Scenes Using Velocity Obstacles](https://mlsanity.com/paper/2301.06512/) (2023)
- [Learning to Fly in Seconds](https://mlsanity.com/paper/2311.13081/) (2023)
- [No Regrets: Investigating and Improving Regret Approximations for Curriculum Discovery](https://mlsanity.com/paper/2408.15099/) (2024)
- [Rethinking Sim2Real: Lower Fidelity Simulation Leads to Higher Sim2Real Transfer in Navigation](https://mlsanity.com/paper/2207.10821/) (2022)
- [NavRL: Learning Safe Flight in Dynamic Environments](https://mlsanity.com/paper/2409.15634/) (2024)
- [Benchmarking Metric Ground Navigation](https://mlsanity.com/paper/2008.13315/) (2020)
- [Large Batch Simulation for Deep Reinforcement Learning](https://mlsanity.com/paper/2103.07013/) (2021)
- [Learning Sim-to-Real Humanoid Locomotion in 15 Minutes](https://mlsanity.com/paper/2512.01996/) (2025)
- [NeuPAN: Direct Point Robot Navigation with End-to-End Model-based Learning](https://mlsanity.com/paper/2403.06828/) (2024)
- [IPAPRec: A promising tool for learning high-performance mapless navigation skills with deep reinforcement learning](https://mlsanity.com/paper/2103.11686/) (2021)
- [HEIGHT: Heterogeneous Interaction Graph Transformer for Robot Navigation in Crowded and Constrained Environments](https://mlsanity.com/paper/2411.12150/) (2024)
- [Deep Reinforcement Learning based Robot Navigation in Dynamic Environments using Occupancy Values of Motion Primitives](https://mlsanity.com/paper/2208.08034/) (2022)
- [Train a Real-world Local Path Planner in One Hour via Partially Decoupled Reinforcement Learning and Vectorized Diversity](https://mlsanity.com/paper/2305.04180/) (2023)
- [Robo-centric ESDF: A Fast and Accurate Whole-body Collision Evaluation Tool for Any-shape Robotic Planning](https://mlsanity.com/paper/2306.16046/) (2023)
- [UniLab: A Heterogeneous Architecture for Robot RL Beyond GPU-Dominant Paradigms](https://mlsanity.com/paper/2605.30313/) (2026)
- [IR-SIM: A Lightweight Declarative Simulator for Navigation Learning and Benchmarking](https://mlsanity.com/paper/2606.08729/) (2026)
- [FastDSAC: Unlocking the Potential of Maximum Entropy RL in High-Dimensional Humanoid Control](https://mlsanity.com/paper/2603.12612/) (2026)

## Abstract (arXiv)

Training Deep Reinforcement Learning (DRL) navigation policies for different robot configurations remains time-consuming. We present FlashNav, a GPU-based framework that trains robot-specific navigation policies within tens of seconds. A unified robot specification configures a lightweight simulator for batched motion updates, range sensing, and footprint collision checking over a shared occupancy map. The framework supports nonconvex footprints, different sensor configurations and drive types. Blockwise ray queries and selective observation recomputation after resets reduce simulation overhead, while GPU-resident replay and overlapping experience collection and learner updates support efficient off-policy training. Experiments covered five robot configurations and three computing platforms. With FastDSAC on a single RTX 5090 GPU, FlashNav can train a deployable navigation policy in under 30 seconds. FlashNav achieved the highest success rate and score in the benchmark comparison. The selected policies were deployed on wheeled, quadrupedal, humanoid, and irregularly shaped robots without additional policy training.