# AI agents learn 3D spatial awareness by choosing where to look next

> Original title: AI进屋后，下一眼该看哪？观察智能体如何建立三维空间认知

- Source: AI Era
- Published: 2026-11-02T00:01:55.000Z
- Bad Influence: https://badinfluence.news/items/d45c4htcv7n9qxqtbudfrkw33
- Original: https://aiera.com.cn/2026/11/02/other/admin/121791/ai%e8%bf%9b%e5%b1%8b%e5%90%8e%ef%bc%8c%e4%b8%8b%e4%b8%80%e7%9c%bc%e8%af%a5%e7%9c%8b%e5%93%aa%ef%bc%9f%e8%a7%82%e5%af%9f%e6%99%ba%e8%83%bd%e4%bd%93%e5%a6%82%e4%bd%95%e5%bb%ba%e7%ab%8b%e4%b8%89%e7%bb%b4/

## Summary

A research agent explores an unseen virtual room with a camera and decides where to turn, move and take the next image, without preset object coordinates or routes. The source article is an excerpt that does not state the model, benchmark scores, or results, so the summary does not extend further.