Code for simulation to reality (Sim2Real) transfer research for autonomous driving

Code to run and train a model able to autonomously drive a RC car. Using Reinforcement Learning the model is trainable on simulator. The code to drive the real car is also available

Development of an image-based autonomous driving system for an e-FSAE.

CabiNet: Scaling Object Rearrangement in Clutter

Source code for "Evolution of Adaptive Force Chains in Reconfigurable Granular Metamaterials"

Deep reinforcement learning for simultaneous robotic manipulation and locomotion

Sim2Real for joint robotic locomotion and manipulation with RCAN

Code for learning a road segmentation network. Duckietown Simulator for data generation with domain randomization: https://github.com/niksaz/randomized-duckietown

Automatic Domain Randomization (ADR) proposed in "Solving Rubik's Cube with a Robot Hand"

SkyScenes: A Synthetic Dataset for Aerial Scene Understanding

Sim2Real transfer of trained deep neural networks for OpenCat robots.

A panoptic segmentation deep learning architecture for sim2real autonomous driving scene understanding

Supervisor for controlling data flow in the BenchBot software stack: https://github.com/qcr/benchbot

Manager for add-ons in the BenchBot software stack: https://github.com/qcr/benchbot

Kuka Reacher Reinforcement Learning Sim2Real Environment for Omniverse Isaac Gym/Sim

RPMArt: Towards Robust Perception and Manipulation for Articulated Objects

Evaluation tools for Semantic Scene Understanding with the BenchBot software stack: https://github.com/qcr/benchbot

State Representations as Incentives for Reinforcement Learning Agents: A Sim2Real Analysis on Robotic Grasping

Code for the Continual Domain Randomization paper