DeepMind and HHMI open-source flybody: a fruit fly simulated joint by joint in MuJoCo
nikola_mr64990 · x · 2026-09-27
flybody, built by Google DeepMind and HHMI Janelia and published in Nature, is now on GitHub under Apache 2.0: a fruit fly reconstructed joint by joint from real microscopy data that walks, grips, flies, and lands inside a MuJoCo physics engine running on a laptop.
Key points:
- Full body modeling: legs, wings, head, and abdomen, all articulated and physically simulated
- Walking alone uses a 59-dimensional action space a network must coordinate
- Leg adhesion lets the fly grip surfaces instead of sliding off
- Ready-made RL environments for walking imitation, free flight, and vision-guided flight driven by the fly's actual visual input
- Python stack: clone the repo, open MuJoCo, and the fly appears on screen
More from Embodied
- IROS 2026 workshop bets on proprioception: 131 teams, 2,800+ IMU challenge submissions — GhaffariMaani · 2026-09-27
- Copy BOM Is Fine, Copying Product Definition Is the Real Worry in Humanoid Robotics — yongqianme · 2026-09-27
- Dutch nonprofit unveils X17 autonomous litter-picking robot with gripper arm, arriving fall 2026 — MarvinTBaumann · 2026-09-27
- Europe is automating street trash cleanup with robots — MarvinTBaumann · 2026-09-27
- PEARL team heads to IROS 2026 with orals on robot RL and a generalist-vs-specialist debate — GeorgiaChal · 2026-09-27
- Gen-HumanEgo: a 1TB+ egocentric robotics dataset with hand tracking trends on Hugging Face — genrobot2025 · 2026-09-27