Driving · 2026-10-06
PreAct-Nav: Agentic Reasoning Before Action for Urban Navigation
Jing Xie, Shouwei Ruan, Yubin Wang, Yuxiang Zhang, Junwei Yang, Songchang Jin, Dianxi Shi
AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.
TL;DR
PreAct-Nav is a training-free urban navigation harness that equips frozen navigation policies with anticipatory reasoning. It maintains medium-horizon navigation subgoals in Navigation Memory, previews actions via an action-conditioned world model in a Predictive Sandbox, and uses a vision-language model to retain or revise actions before execution, then updates memory from real observations.
Why it matters
Urban navigation requires consistent long-horizon progress from egocentric, transient observations. The authors propose a reasoning loop that evaluates candidate actions using predicted visual outcomes before execution and then corrects assumptions with execution feedback, aiming to improve action selection on longer and more complex routes without updating model weights.
Method
- Training-free harness with a shared vision-language reasoner coordinating a VLM review/revise loop over frozen navigation policies.
- Navigation Memory maintains medium-horizon subgoals (objective, visual cues, open question, status) and updates subgoal progress using execution feedback comparing expected cues to real observations.
- Predictive Sandbox exposes an action-conditioned world model to forecast future visual cues for candidate movements; predicted rollouts are interpreted under the current subgoal to retain or revise actions.
Abstract (from arXiv)
Urban navigation requires embodied agents to pursue long-horizon goals through local decisions based on egocentric observations. However, existing agentic navigation methods often struggle to translate distant goals into coherent local decisions in large-scale physical environments. Their reliance on linguistic reasoning over transient observations or limited history constrains anticipation of the consequences of actions and future conditions, despite the importance of such foresight for navigating long and complex urban routes. To bridge this gap, we propose PreAct-Nav, an agentic navigation framework that equips frozen policies with anticipatory reasoning for robust urban navigation. Our central idea is to anchor local decisions in persistent medium-horizon subgoals, assess the consequences of predicted actions before execution, and continually update the reasoning context using actual outcomes. At its core, a navigation memory module maintains the active subgoal and relevant experience across decisions, translating distant goals into actionable intermediate objectives. We further introduce a predictive world sandbox that uses an action-conditioned world model (AC-WM) to forecast world dynamics conditioned on candidate movements. A vision-language model (VLM) reasoner interprets these predictions under the current subgoal to retain or revise actions. After execution, real observations are used to assess outcomes, correct inconsistent assumptions, and update memory to continue or reformulate the subgoal. Extensive evaluations demonstrate that the proposed PreAct-Nav improves action selection through memory updates and visual prediction, with more pronounced gains on longer routes and routes with more turns.
Related papers
- InJePA: Intention-First Latent Planning with Coupled JEPA World Models for Visual Navigation
- SearchWorld: Spatial Value-Grounded Imagination for UAV Object Search via World Models
- PlanWAM: Planning-Shaped Future Representations for End-to-End Autonomous Driving
- G$^2$WAM: World Action Models for Efficient and Geometry Grounded Image-Goal Navigation
- RIWANav: Recursive World-Action Models with Self-Improvement for Urban Navigation
Publication record
Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.
First public release: 2026-10-04 · Indexed by Pub-AI: 2026-10-06 UTC
PreAct-Nav: Agentic Reasoning Before Action for Urban Navigation
Preprint posted: 2026-10-04 · Original record ↗
- arxiv
- 2610.04916
Explicit source versions (1)
- v1 · 2026-10-04
- Preprint available
Source date: 2026-10-04 · First detected: 2026-10-25 UTC · Historical baseline
Source evidence (1)
- Official evidence ↗ · atom:published (official metadata)
Abstract · arxiv
Urban navigation requires embodied agents to pursue long-horizon goals through local decisions based on egocentric observations. However, existing agentic navigation methods often struggle to translate distant goals into coherent local decisions in large-scale physical environments. Their reliance on linguistic reasoning over transient observations or limited history constrains anticipation of the consequences of actions and future conditions, despite the importance of such foresight for navigating long and complex urban routes. To bridge this gap, we propose PreAct-Nav, an agentic navigation framework that equips frozen policies with anticipatory reasoning for robust urban navigation. Our central idea is to anchor local decisions in persistent medium-horizon subgoals, assess the consequences of predicted actions before execution, and continually update the reasoning context using actual outcomes. At its core, a navigation memory module maintains the active subgoal and relevant experience across decisions, translating distant goals into actionable intermediate objectives. We further introduce a predictive world sandbox that uses an action-conditioned world model (AC-WM) to forecast world dynamics conditioned on candidate movements. A vision-language model (VLM) reasoner interprets these predictions under the current subgoal to retain or revise actions. After execution, real observations are used to assess outcomes, correct inconsistent assumptions, and update memory to continue or reformulate the subgoal. Extensive evaluations demonstrate that the proposed PreAct-Nav improves action selection through memory updates and visual prediction, with more pronounced gains on longer routes and routes with more turns.