# Deep RL at Scale waste sorting fleet data

From the catalogue. Listed by one researcher from one source page.

- id: `deep-rl-at-scale-waste-sorting-fleet-data`
- kind: real manipulation scripted or autonomous
- robot or device: fleet of 23 mobile manipulators in three office buildings (class: mobile manipulator)
- how collected: autonomous reinforcement learning fleet over about two years
- size as stated: 23 robots, three office buildings; 9,527 hours claimed, not released
- hours: not stated | hours only claimed: 9,527 | episodes: not stated
- year: 2023
- organisation: Google (Robotics at Google and Everyday Robots, per author list)
- licence: not stated (class: not stated)
- access: not released (class: not stated)
- link (paper): https://arxiv.org/abs/2305.03270
- note: One of the largest autonomous manipulation fleet logs with a stated duration; never published.

Confirm the licence at the link before relying on it.