LEARNER Dataset: SLAM and Path Planning synthetic environments for robots in challenging conditions

LEARNER Dataset: SLAM and Path Planning simulation environments for robots in challenging conditions

This repository contains simulated environments for training and evaluating Deep Learning approaches, mainly targeting SLAM and Path Planning architectures for robots in dynamic and highly varying conditions.The dataset contains two different worlds:

Warehouse, produced in Nvidia Omniverse

Office, produced in Unreal Engine

Both cases were developed with realistic lighting and atmospheric conditions and contain dynamic events, such as:

Switching lights off

Fire outbursts

Heavy smoke

Moving objects and humans

We offer both the source files for the simulated worlds (USD_Files) and sequences of images recorded under different conditions (Frames) for each world.

Acknowledgment

This research is implemented in the framework of H.F.R.I call ``Basic research Financing (Horizontal support of all Sciences)'' under the National Recovery and Resilience Plan ``Greece 2.0'' funded by the European Union – NextGenerationEU (H.F.R.I. Project Number: 15339).

We encourage the utilization of our dataset under the restriction of following the above acknowledgment.

For more information, please visit the project's official website or contact us at: lbampis [at] ee.duth.gr