all research

Construction Management · 7 min read

Agents still get lost: a real-scale city says the hard part is not seeing

A simulation of Hong Kong built from territory-wide 3D geospatial data tested whether AI agents can act in a real built environment. They recognise things well and lose the plot over distance.

Territory-wide

3D geospatial data, Hong Kong

Strong

Atomic visual recognition

Weak

Sustained goal-directed action

Analysis of published research

UrbanGround: From Local Perception to Spatial Agency in a Real-Scale City

Tianjie Ju, Zheng Wu, Yueqing Sun, Yuhan Cui, Bobo Li, Shengqiong Wu, Pengzhou Cheng, Haodong Zhao and others · arXiv · 27 August 2026

Most agent benchmarks are toys. A grid world, a simplified room, a game. This one is Hong Kong, replicated from territory-wide 3D geospatial data, with first-person navigation and spatial reasoning tasks.

The question it asks is the one that matters for anything autonomous on a site or in a building: can an agent turn what it sees locally into dependable action across a real, large, messy space?

The answer, and it is a useful no

Contemporary multimodal agents showed useful atomic abilities in visual recognition. They can look at a thing and say what it is. But they struggled with sustained goal-directed urban exploration, and the paper is precise about the failure mode: errors accumulate without effective correction.

It is not that the agent cannot see. It is that it cannot notice it has been wrong for the last twenty minutes.
Where the capability runs out, based on the reported findings
  1. 1

    Perceive

    Recognise objects and features in the immediate view. Reported as a genuine strength of current models.

  2. 2

    Decide

    Choose the next action from what is visible now. Also broadly works.

  3. 3

    Sustain

    Hold the goal across many steps while the scene changes. This is where performance falls away.

  4. 4

    Correct

    Notice that the accumulated path has drifted and recover. The reported failure point.

Why a negative result is the useful one this fortnight

Four of the five papers I read this fortnight report improvements. This one reports a boundary, and boundaries are what you need when someone is selling you autonomy.

The pattern generalises well beyond city navigation. Any agent operating over a long horizon in a physical environment faces the same structure: perception is solved enough, single decisions are fine, and the difficulty is accumulated error nobody catches. That is exactly the failure mode a site robot, an autonomous survey drone or a long-running inspection agent will exhibit.

The design implication

Work with me

Run this model against your own project

I am Kanishk Kapoor, Technical Accounts Manager at AI Institute in Dublin. I build agentic AI systems with built-environment teams across Ireland and the UK. If any figure here looks wrong for your business, that is the useful conversation. Send me your assumptions and I will re-run it.

Continue reading