Oopsie-data: Building the first open dataset of robot failure trajectories
chris_j_paxton · x · 2026-10-30
Robotics has a data problem, but not the one you'd expect: nearly all robotics data consists of successful policy rollouts, while failed manipulation trajectories are routinely thrown out and deleted. Yet world models and reward function learning—among the most promising paths to reliable real-world robotics—need exactly those failure examples.
The oopsie-data project by cvoelcker, Arpit Bahety, and Zhiyuan Zhou aims to build the first large community-driven open-source dataset of failure trajectories, and is organizing a workshop around the effort. Chris Paxton highlighted the initiative on RoboPapers episode 114.
More from Embodied
- SCAMPER demo: virtual rat trots and gallops with gaits it never saw in training — tyrell_turing · 2026-10-30
- DIY humanoid build, day 7: creator gives up fixing physics, adds 1-DOF ankles — _Stocko_ · 2026-10-30
- Lambda Surgical launches first-in-human study of LiDAR-based LambdaGuide at PINE — ditzikow · 2026-10-30
- Gaussian splats double as shared GPS: one city map localizes drones, robots and phones — bilawalsidhu · 2026-10-30
- SimReady dataset offers 2,677 simulation-ready 3D assets to build virtual test worlds — victormustar · 2026-10-30
- MS-Emulator: one policy controls 700 muscles for whole-body human motion tracking — tweetsatpreet · 2026-10-30