Liquid.ai

🥳🥳🥳🥳🥳🥳 ANNOUNCING ROS 2: FOXY FIZTROY!!! It runs on Ubuntu 20.04, OSX, and Windows 10 with three years of support. Learn about all of the features and where to download the binaries in this blog post: openrobotics.org/blog/2020/6/5/…

I am presenting our paper on continuous-wave LIDAR sensor simulation @icra2020 where we use automatic differentiation to estimate simulation parameters from real-world measurements. Paper: arxiv.org/abs/1912.01652 Video: youtu.be/hGphZbMDbYA join.slack.com/share/zt-ersnv…

Excited to share my latest work on "Model-based Generalization under Parameter Uncertainty using Path Integral Control" out at #ICRA2020 . w/ @ankurhandos , Nathan, Kendall, @todd_murphey, Dieter Fox checkout videos+paper sites.google.com/view/emppi on #ICRA2020 slack: #mod08_3

The upcoming workshop on Learning in Artificial Open Worlds at @icmlconf has announced a call for papers. They're looking for work that focuses on, or is enabled by, machine learning in open worlds. Submissions are due by May 19, 2020: aka.ms/AA7xar3 #ICML2020

Traversing the Reality Gap via Simulator Tuning arxiv.org/abs/2003.01369

Bayesian Domain Randomization for Sim-to-Real Transfer arxiv.org/abs/2003.02471

Fast Online Adaptation in Robotics through Meta-Learning Embeddings of Simulated Priors arxiv.org/abs/2003.04663

We'd like to draw your attention to the 2nd Workshop on Closing the Reality Gap in Sim2Real Transfer for Robotics. It is a full day workshop at R:SS 2020 (Corvallis, Oregon, USA), 12th July, led by @hoeferse website: sim2real.github.io

Facebook's RIDE encourages AI agents to explore their environments wp.me/p8wLEc-aQFK by @Kyle_L_Wiggers

Benchmarking Simulated Robotic Manipulation through a Real World Dataset arxiv.org/abs/1911.01557

Gym-Ignition: Reproducible Robotic Simulations For Reinforcement Learning arxiv.org/abs/1911.01715

Closing the Reality Gap with Unsupervised Sim-to-Real Image Translation for Semantic Segmentation in Robot Soccer arxiv.org/abs/1911.01529

DeepRacer: Educational Autonomous Racing Platform for Experimentation with Sim2Real Reinforcement Learning arxiv.org/abs/1911.01562

I wrote a guide on transferring neural networks from simulation to the real world: josh-tobin.com/assets/pdf/ran…

Sim2real transfer learning for 3D pose estimation: motion to the rescue arxiv.org/abs/1907.02499

We released our physics engine and a few RL examples. Check out our github repos below! raisimLib (github.com/leggedrobotics…): simulator raisimOgre (github.com/leggedrobotics…): visualizer raisimGym (github.com/leggedrobotics…): gym examples

Synthetic Data for Deep Learning arxiv.org/abs/1909.11512

RLBench: The Robot Learning Benchmark & Learning Environment arxiv.org/abs/1909.12271 website: sites.google.com/view/rlbench

Assistive Gym: A Physics Simulation Framework for Assistive Robotics arxiv.org/abs/1910.04700

Modelling Generalized Forces with Reinforcement Learning for Sim-to-Real Transfer arxiv.org/abs/1910.09471