EmbodiedData
Start here →
Datasets/Navigation/Berkeley GNM RECON — Visual Navigation Trajectories
✓ Source card reviewedApache 2.0Version: LeRobot v3

Berkeley GNM RECON — Visual Navigation Trajectories

A visual-navigation collection with 11,834 episodes and 610,907 frames, pairing low-resolution RGB observations with robot state and navigation actions at 3 Hz.

UB
Published by UC BerkeleyLeRobot / UC Berkeley · Source updated 18 Jul 2026 · Reviewed 17 Jul 2026
View dataset files ↗1.33 GB · LeRobot v3Downloads and licence terms are handled by the original publisher on Hugging Face.

START HERE

This dataset, explained simply

Intermediate

A visual-navigation collection with 11,834 episodes and 610,907 frames, pairing low-resolution RGB observations with robot state and navigation actions at 3 Hz.

WHAT CAN I TRAIN?✓ Robot-learning experiments
WHAT DO I NEED?• Python basics
SUITABILITY✓ Contains robot actions✓ Real-world recordings• Check source for language labels
New to the terminology?Episode = one complete attemptObservation = what the robot sees and sensesAction = the command sent to the robot

VERIFICATION RECORD

What EmbodiedData checked

Source statusAvailable at publisherLast checked17 July 2026Dataset versionLeRobot v3Training instructionsNot tested by EmbodiedData

Verification confirms the linked metadata at the stated date. It is not a guarantee that upstream files, dependencies or licences will remain unchanged.

Dataset previewIllustrative image—not an original dataset frame
Preview only
PUBLISHER-HOSTED MEDIAOpen genuine dataset samplesUC Berkeley · source repositoryView real samples at source ↗
EXAMPLE OF THE TASKNavigate toward the visual goal using the recorded mobile-robot observations.
View real samples on Hugging Face ↗

Original media, file structure, and dataset viewer are provided by the publisher.

DATASET OVERVIEW

What the publisher reports

A visual-navigation collection with 11,834 episodes and 610,907 frames, pairing low-resolution RGB observations with robot state and navigation actions at 3 Hz.

11,834Reported episodes
3 HzReported frequency
56h 34mReported duration
LeRobot v3Dataset format

KNOWN CONFIGURATION

Robot, observations and reuse

Robot or capture systemMobile robot
Reported observationsRGB video, Robot state, Navigation actions
ActionsRobot actions are included
EnvironmentReal-world · visual navigation
Licence shown on source cardApache 2.0

TRUST & LIMITATIONS

What EmbodiedData has—and has not—checked

Original Hugging Face source linkChecked
Summary against publisher cardReviewed
File integrity and missing valuesNot independently tested
Privacy, safety and legal complianceNot independently tested

Always read the original dataset card, licence, limitations and file documentation before training or redistribution.

SOURCE & CITATION

Use the publisher’s citation

EmbodiedData does not invent a citation or dataset version. Open the original card and copy the citation requested by the publisher.

Open original dataset card ↗