Applications
NaviFormer: A Deep Reinforcement Learning Transformer-like Model to Holistically Solve the Navigation Problem
arXiv:2604.16967v1 Announce Type: new Abstract: Path planning is usually solved by addressing either the (high-level) route planning problem (waypoint sequencing to achieve the final goal) or the (low
arXiv:2604.16967v1 Announce Type: new Abstract: Path planning is usually solved by addressing either the (high-level) route planning problem (waypoint sequencing to achieve the final goal) or the (low-level) path planning problem (trajectory prediction between two waypoints avoiding collisions). However, real-world problems usually require simultaneous solutions to the route and path planning subproblems with a holistic and efficient approach. In this paper, we introduce NaviFormer, a deep reinforcement learning model based on a Transformer architecture that solves the global navigation problem by predicting both high-level routes and low-level trajectories. To evaluate NaviFormer, several experiments have been conducted, including comparisons with other algorithms. Results show competitive accuracy from NaviFormer since it can understand the constraints and difficulties of each subproblem and act consequently to improve performance. Moreover, its superior computation speed proves its suitability for real-time missions.
Related
- EgoWalk: A Multimodal Dataset for Robot Navigation in the Wild
- Weakly-supervised Learning for Physics-informed Neural Motion Planning via Sparse Roadmap
- Foundation Models in Robotics: A Comprehensive Review of Methods, Models, Datasets, Challenges and Future Research Directions
- Model-Based Reinforcement Learning under Random Observation Delays
Source: arXiv cs.RO | 2026-04-21