RoboDojo: A Comprehensive Benchmark for Robotics
机器之心 · wechat · 2026-07-17
RoboDojo is an embodied intelligence benchmark covering simulation and real robot manipulation, with 42 simulation tasks and 18 real robot tasks, comparing 30 representative robot policies under the same standard.
Results show current top general-purpose robot policies achieve an average success rate of about 8.80% in simulation and 12.8% in the real world; human experts achieve 76.03% in simulation and 100% in the real world. The author notes that existing robot foundation models still have significant gaps in generalization, memory, fine manipulation, long-horizon execution, and open semantic understanding. RoboDojo also provides heterogeneous parallel simulation and a unified access layer XPolicyLab for easy model submissions, task expansion, and real robot evaluation.
More from Embodied
- Samsung really sells a "Bespoke AI Kimchi Plus" fridge, complete with kimchi cartons — airkatakana · 2026-10-26
- OmniEgoCap Turns Any Egocentric Device into a Full-Body 3D Mocap System — Scobleizer · 2026-10-26
- Tesla FSD approved in Latvia, the 10th European country — mitchdeg · 2026-10-26
- On-device VLM goes production: GenieX runs Liquid AI's LFM-VL live at PyTorch Conference — PyTorch · 2026-10-26
- One long Claude Code session builds a full Android TV launcher with a 1-second-latency live camera — aristofeles · 2026-10-26
- LeCun co-authored Temporal Straightening: straightening latent paths boosts gradient-based planning success rates — ylecun · 2026-10-26