AI Agent Skills

[RSS 2019] End-to-End Robotic Reinforcement Learning without Reward Engineering

Efficient optimal control solvers for robotic systems.

An open-source SDK for developing and testing algorithms on commonly used robotic hardware.

Overview about state-of-art imitation learning techniques for robotic manipulation, enabling generalization across diverse tasks and environments.

A Benchmark for Evaluating Generalization for Robotic Manipulation

Official implementation of Chain-of-Action: Trajectory Autoregressive Modeling for Robotic Manipulation. Accepted in NeurIPS 2025.

This repo is designed for General Robotic Operation System

Several controllers to move Ridgeback mobile-robot and UR5 robotic-arm.

Open source 3D printed Robotic Actuator

Code for PerAct², a language-conditioned imitation learning agent designed for bimanual robotic manipulation using the RLBench environment. It includes dataset generatio…

Real-time robotic object pose estimation with deep learning

Trajectory optimization algorithms for robotic control.

Robotic grasp dataset for multi-object multi-grasp evaluation with RGB-D data. This dataset is annotated using the same protocal as Cornell Dataset, and can be used as m…

Learning multi-agent robotic mobile manipulation with deep reinforcement learning

RoboDual: Dual-System for Robotic Manipulation

[IROS24 Oral]ManipVQA: Injecting Robotic Affordance and Physically Grounded Information into Multi-Modal Large Language Models