# DISCO transfers robot planning from simulated search into a deployable policy

_Published Monday, October 19, 2026 at 11:07 PM EDT · Science, Robotics · Latest · Tier 2 — Notable_

Researchers report that DISCO, a robot manipulation method described in an arXiv preprint, transfers plans generated in physics simulation into a learned controller that can operate without simulator-based search at deployment. They report better performance than behavior cloning, reinforcement learning and search-based baselines across simulated and real-world object rearrangement tasks.

The method searches possible sequences of continuous actions, using a learned policy to propose movements and a value function to assess progress toward completion. GPU-parallelized simulation periodically feeds successful plans back into that model. The reported results concern object rearrangement, rather than general-purpose robot operation.

## Sources

- [arXiv Query: search_query=cat:cs.RO&id_list=&start=0&max_results=30](https://arxiv.org/abs/2610.25785v1)

---
Canonical: https://techandbusiness.org/newswire/aCBQeDY-8PFa4OFdNfWC1d
Published: 2026-10-20T03:07:15.776Z
Story chronology: 2026-10-19T04:50:13.000Z
Retrieved: 2026-10-20T06:10:58.519Z
Publisher: Tech & Business (techandbusiness.org)