hysicl, a startup working on the sim-to-real gap for robotics, raised $10M to build digital worlds real
enough to train robots, reusing a pipeline originally made to market goods online. Simulation is quietly becoming shared infrastructure. #AI #Robotics #SimToReal

Researchers propose a real-to-sim-to-real framework enabling rapid test-time fine-tuning of humanoid parkour policies on unfamiliar terrains, using mesh reconstruction from real-world captures. The two-stage…

#Robotics #HumanoidRobots #SimToReal #TestTimeTraining
https://arxiv.org/abs/2602.02331

A curriculum-driven framework pairs Evolution Strategy policy learning with Statistical Model Checking to co-optimize performance and enlarge the safe operating region for sim-to-real robotic controllers. The approach…

#Robotics #SimToReal #NeuralControl #SafeRL
https://arxiv.org/abs/2608.06481

Researchers propose a Real-to-Sim-to-Real framework with an information-theoretic cost function to reduce the sim-to-real gap, integrating with RL algorithms like PPO and SAC and optionally using differentiable simulators.

#Robotics #SimToReal #ReinforcementLearning
https://arxiv.org/abs/2503.10118

New arXiv work shows vision-language-action models can be reinforcement-learning fine-tuned across diverse simulated scenes using generative 3D worlds, avoiding costly sim-to-real fixes and scene-specific…

#robotics #AI #simtoreal #reinforcementlearning
https://arxiv.org/abs/2603.18532

ExpertGen automates expert policy learning in simulation by refining diffusion-based behavior priors from imperfect demos, including LLM-synthesized ones, using reinforcement learning while keeping the…

#Robotics #SimToReal #DiffusionPolicy #BehaviorCloning
https://arxiv.org/abs/2603.15956

Hugging Face just dropped its $399 Microduck biped robot—packed with seven pre‑trained RL moves straight from a physics simulator. Ready to see sim‑to‑real magic in action? Dive in! #Microduck #NeuralPolicy #SimToReal

🔗 aidailypost.com/news/hugging...

ディズニーリサーチ
CoCo-InEKF: State Estimation with Learned Contact Covariances in Dynamic, Contact-Rich Scenarios
動画
youtu.be/8mfLNICByvE?...
論文
arxiv.org/abs/2605.15122

候補接触点-IMUの速度相関の学習＆接触状態推定による SimToReal

実機がシミュレーション通り動いてすごい
複数動作で統計的一貫性がとれてる
Limaって名前だったのね

元のシミュレーションのノミナルな動作は、モーキャプ＆動作リターゲティングで作成したのかしら

Researchers introduce a hybrid real‑to‑sim‑to‑real learning approach that helps robots adapt quickly to new tasks with minimal trials. The result? Faster learning and stronger performance with fewer attempts.
ieeexplore.ieee.org/abstract/doc...

#RobotLearning #BayesMethod #SimToReal

@NVIDIA ISAAC Sim at @DSUBangalore:
-10k training eps/hr on DGX B200
-Domain randomization
-Photorealistic sensors
-Swarm coordination

Supercharging India’s AI future with DSU. India’s AI-First University.

#ISAACSim #NVIDIA #DSU #SimToReal #IndiaAI

Unitree R1 İnsansı Robot ile Robot kontrol algoritmaları, Görü + hareket AI modelleri ve Simülasyon → gerçek dünya geçişi algoritmalarını test edin. #UnitreeR1 #HumanoidRobot #RoboticsResearch #AI #SimToReal #RobotControl #DeepLearning #UniversityLab #MeseBilisim #MucitRobot