analiza aiqbench
WayFinder addresses the computational bottleneck of deploying Visual Language Action models by decoupling high-level reasoning from low-level control, enabling zero-shot adaptation without fine-tuning. This hierarchical approach significantly boosts navigation reliability in autonomous robots while minimizing expensive inference costs through on-demand querying of large multimodal models.
WayFinder: Hierarchical Visual-Language-Action for Zero-Shot Waypoint Generation and Low-Level Kinematic Control
arXiv:2609.37922v1 Announce Type: new Abstract: Visual Language Action (VLA) models offer unprecedented generalization for autonomous robots; however, their real-world deployment is frequently bottlenecked by unreliable execution and the prohibitive computational cost of fine-tuning for specific robot embodiments and tasks. To bridge this gap, we propose WayFinder, an end-to-end, closed-loop hierarchical VLA framework that circumvents the need for fine-tuning by decoupling high-level task reasoning from low-level kinematic control. WayFinder utilizes a zero-shot, offboard Multimodal Large Language Model (MLLM) policy to process linguistic context and state maps for strategic waypoint generation. Asynchronously, a lightweight, onboard policy executes real-time kinematic control at high frequency based on continuous sensor feedback. We evaluate WayFinder in Microsoft AirSim, testing on four environments of varying complexity and three MLLM scales to balance prediction efficacy with computational efficiency. Our results demonstrate that WayFinder achieves superior navigation reliability compared to baseline low-level policies. By querying the high-level MLLM only during navigation failures, WayFinder eliminates the need for fine-tuning, minimizes expensive inferences, and significantly increases navigation success rates by up to 27.45%.
Abstract:Visual Language Action (VLA) models offer unprecedented generalization for autonomous robots; however, their real-world deployment is frequently bottlenecked by unreliable execution and the prohibitive computational cost of fine-tuning for specific robot embodiments and tasks. To bridge this gap, we propose WayFinder, an end-to-end, closed-loop hierarchical VLA framework that circumvents the need for fine-tuning by decoupling high-level task reasoning from low-level kinematic control. WayFinder utilizes a zero-shot, offboard Multimodal Large Language Model (MLLM) policy to process linguistic context and state maps for strategic waypoint generation. Asynchronously, a lightweight, onboard policy executes real-time kinematic control at high frequency based on continuous sensor feedback. We evaluate WayFinder in Microsoft AirSim, testing on four environments of varying complexity and three MLLM scales to balance prediction efficacy with computational efficiency. Our results demonstrate that WayFinder achieves superior navigation reliability compared to baseline low-level policies. By querying the high-level MLLM only during navigation failures, WayFinder eliminates the need for fine-tuning, minimizes expensive inferences, and significantly increases navigation success rates by up to 27.45%.
Robotics (cs.RO)
arXiv:2609.37922 [cs.RO]
(or arXiv:2609.37922v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2609.37922 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 260 cuvinte
Subiecte conexe
AgenticNav: Zero-Shot Vision-and-Language Navigation as a Tool-Calling Harness
arXiv · 28 iul. 2026
AgenticNav: Zero-Shot Vision-and-Language Navigation as a Tool-Calling Harness
arXiv · 10 iun. 2026
A Brain-inspired Hierarchical Framework for Zero-Shot Robot Task Reasoning and Execution
arXiv · 9 sept. 2026
SpaceVLN: A Zero-Shot Vision-and-Language Navigation Agent with Online Spatial Cognitive Memory and Reasoning
arXiv · 9 iun. 2026