UPenn's GRASP lab releases OctoSense: 8.5TB multi-sensor robotics dataset with event cameras
RexDouglass · x · 2026-10-10
UPenn's GRASP lab and Voxel51 released OctoSense, a multimodal robotics dataset now on Hugging Face.
- A hardware-synced 8-sensor rig: stereo RGB at 100Hz, two event cameras (millions of events per second vs lidar's 10Hz), thermal, lidar, 400Hz IMU, and RTK GPS
- Collected on three platforms with the same rig: a car, a boat, and a Unitree Go2, covering lane-less waters and GPS-denied scenarios
- Includes runs with deliberately degraded sensors for robustness testing
- Ships 8 representative episodes as MCAP loadable in FiftyOne with one line; full release is 8.5TB under MIT license
More from Embodied
- Musk: Digital Optimus beats Diablo halfway through with no APIs, just screen pixels — XFreeze · 2026-10-10
- Tongji team builds spiderweb-inspired 6-axis robotic 3D printer with self-supporting structures — lukas_m_ziegler · 2026-10-10
- Robot hand turns book pages: UVTA visual-tactile-action model hits 70% vs 29% baseline — CyberRobooo · 2026-10-10
- Sony surgical robot's autonomous suturing demo sparks debate over medical AI liability — TansuYegen · 2026-10-10
- VioLA trains humanoid policies on 140M mostly-human frames, hitting 88.6% zero-shot manipulation — Michael_J_Black · 2026-10-10
- Second humanoid robot deathmatch shows dramatically improved fall-recovery skills — cixliv · 2026-10-10