Blog

Phần 2/11

SimpleVLA-RL (2): Architecture & Algorithm

14 phút đọc

Phần 3/11

SimpleVLA-RL (3): Setup & Training

Phần 4/11

SimpleVLA-RL (4): Results & Key Takeaways

Phần 5/11

SimpleVLA-RL (5): Comparison with LeRobot

Phần 6/11

SimpleVLA-RL (6): OpenArm — Training Roadmap

Phần 7/11

SimpleVLA-RL (7): Collecting Data for OpenArm

14 phút đọc

Phần 8/11

SimpleVLA-RL (8): Training & Deploying on OpenArm

13 phút đọc

Phần 9/11

SimpleVLA-RL (9): OpenArm Simulation & Data Collection

17 phút đọc

Phần 10/11

SimpleVLA-RL (10): SFT & RL Training for OpenArm

Phần 11/11

SimpleVLA-RL (11): Sim-to-Real Transfer for OpenArm

16 phút đọc

Psi0 Hands On

· 6 phần

Phần 1/6

Ψ₀ Hands-On (1): Overview & Key Ideas

Phần 2/6

Ψ₀ Hands-On (2): The Three-Tier Architecture

Phần 3/6

Psi0 Hands-On (3): Data Recipe & Pipeline

Phần 4/6

Psi0 Hands-On (4): Setup & Training Pipeline

14 phút đọc

Phần 5/6

Psi0 Hands-On (5): Inference & Evaluation

13 phút đọc

Phần 6/6

Psi0 Hands-On (6): Ablation Studies & Key Takeaways

RL Locomotion cho Humanoid Robot

· 10 phần

Phần 1/10

RL Locomotion for Humanoids: Fundamentals & Environment Setup

Phần 2/10

Reward Engineering for Bipedal Walking: The Art of Reward Design

Phần 3/10

Unitree G1: Training a Walking Policy from Scratch in Isaac Lab

Phần 4/10

Unitree G1: Terrain Adaptation & Robust Walking

Phần 5/10

Unitree H1: Full-size Humanoid Locomotion Training

Phần 6/10

Unitree H1: Running, Turning & Dynamic Motions

Phần 7/10

Unitree H1-2: Enhanced Locomotion with New Hardware

Phần 8/10

Loco-Manipulation: Walking While Carrying & Manipulating Objects

Phần 9/10

Humanoid Parkour: Jumping, Climbing & Obstacle Courses

Phần 10/10

Sim-to-Real for Humanoids: Deployment Best Practices

RL cho Robot Manipulation

· 8 phần

Phần 1/8

RL for Manipulation: MDP, Reward Design & Environment Setup

Phần 2/8

Grasping with RL: Stable Grasp & Object Variety

Phần 3/8

RL Force Control: Balancing a Cup of Water

Phần 4/8

Precision Pick-and-Place: Position & Orientation Control

Phần 5/8

Carrying & Transporting Objects: Stability During Motion

Phần 6/8

Contact-Rich Manipulation: Assembly, Insertion & Peg-in-Hole

Phần 7/8

Tool Use: Teaching Robots to Use Tools with RL

Phần 8/8

Multi-Step Manipulation: Curriculum Learning for Long-Horizon Tasks

Điều Khiển Truyền Thống cho Robot Arm

· 8 phần

Phần 1/8

Robot Arm Fundamentals: Kinematics & Dynamics for Control

Phần 2/8

Trajectory Types: Joint Space vs Cartesian Space Planning

Phần 3/8

MoveJ & MoveL: Joint Motion and Linear Motion Explained

Phần 4/8

MoveC & Circular Motion: Arc, Helix & Orbital Paths

Phần 5/8

Velocity Blending: Smooth Transitions Between Motion Segments

Phần 6/8

Motion Profiles: Trapezoidal, S-Curve & Polynomial Trajectories

Phần 7/8

Advanced Trajectories: Spline, B-Spline & Time-Optimal Planning

Phần 8/8

Integration: ROS 2 MoveIt2, URScript & Real Robot Deployment

Simulation cho Robotics

· 6 phần

Phần 1/6

Simulation for Robotics: MuJoCo vs Isaac Sim vs Gazebo

Phần 2/6

Getting Started with MuJoCo: From Installation to First Robot Simulation

Phần 3/6

NVIDIA Isaac Lab: GPU-Accelerated RL Training from Zero

Phần 4/6

Domain Randomization: The Key to Sim-to-Real Transfer

Phần 5/6

Sim-to-Real Pipeline: From Training to Real Robot

Phần 6/6

Digital Twins and ROS 2: Simulation in Manufacturing

ROS 2 từ A đến Z

· 4 phần

Phần 1/4

ROS 2 A to Z Part 1: Setup and First Node

Phần 2/4

ROS 2 A to Z (Part 2): Topics, Services, and Actions

Phần 3/4

ROS 2 A to Z (Part 3): Nav2 — Your First Autonomous Robot

Phần 4/4

ROS 2 A to Z (Part 4): ros2_control and Hardware

AI cho Robot

· 7 phần

Phần 1/7

RL for Robotics: PPO, SAC and How to Choose Your Algorithm

Phần 2/7

Imitation Learning: BC, DAgger and DAPG for Robots

Phần 3/7

Action Chunking Transformers (ACT): Architecture in Detail

Phần 4/7

Diffusion Policy: A Revolution in Robot Manipulation

Phần 5/7

VLA Models: RT-2 → Octo → OpenVLA → π0

Phần 6/7

SpatialVLA: 3D Understanding for Robot Manipulation

Phần 7/7

Hands-on: Fine-tune OpenVLA with LeRobot

13 phút đọc

Robot Manipulation Masterclass

· 7 phần

Phần 1/7

Robot Grasping 101: Analytical to Learning-Based

Phần 2/7

Imitation Learning for Manipulation: BC, DAgger, ACT

Phần 3/7

Diffusion Policy in Practice: From Theory to Code

Phần 4/7

VLA for Manipulation: RT-2, Octo, pi0 in Practice

Phần 5/7

Dexterous Manipulation: Teaching Robot Hands

Phần 6/7

Bimanual Manipulation: Teaching Robots to Use Both Arms

Phần 7/7

Building Manipulation Systems with LeRobot: From Zero to Deployment

Locomotion từ Zero đến Hero

· 7 phần

Phần 1/7

Locomotion Basics: From ZMP to CPG

Phần 2/7

RL for Locomotion: PPO, Reward Shaping and Curriculum

Phần 3/7

Quadruped Locomotion: legged_gym to Unitree Go2

Phần 4/7

Walk These Ways: Adaptive Locomotion One Policy

Phần 5/7

Robot Parkour: Jumping and Climbing with RL

Phần 6/7

Bipedal Walking: Controlling 2-Legged Robots with RL

Phần 7/7

Sim-to-Real for Locomotion: Reality and Experience

Humanoid Robot Engineering

· 6 phần

Phần 1/6

Humanoid Robots 2026: Complete Platform Overview

Phần 2/6

Humanoid Control Basics: IK to Task-Space Control

Phần 3/6

Whole-Body MPC: Real-Time Full-Body Control

6 phút đọc

Phần 4/6

RL for Humanoid: From Humanoid-Gym to Sim2Real

Phần 5/6

Loco-Manipulation: Walking and Manipulating Simultaneously

Phần 6/6

The Future of Humanoids: Opportunities for Robotics Engineers

Navigation hiện đại

· 5 phần

Phần 1/5

SLAM A to Z: LiDAR, Visual and How Robots Localize

Phần 2/5

ROS 2 Nav2: Complete Navigation for AMR

Phần 3/5

Learning-based Navigation: GNM, ViNT and NoMaD

Phần 4/5

Vision-Language Navigation: Robot Following Instructions

Phần 5/5

Outdoor Navigation and Multi-Robot Coordination

Support VnRobo

If these tutorials helped you, consider buying me a coffee!

Buy me a coffee

NEWTutorial

vlanvidianvlabsqwen2.5-vllibero

VLA-0: State-of-the-Art Robot VLA Without Changes

NVIDIA NVlabs proves action as text reaches 94.7% on LIBERO, beating pi_0 and GR00T-N1 with zero architecture modification — just Qwen2.5-VL-3B.

4/5/202612 phút đọc

rdt2foundation-modelbimanualumimanipulation

RDT2: Zero-Shot Cross-Embodiment Foundation Model for Bimanual UR5e/Franka

Detailed guide to RDT2 from THU-ML — a foundation model that zero-shot deploys to bimanual UR5e and Franka, with open-source code.

1/5/20269 phút đọc

so-101isaac-lablerobotsim-to-realmanipulation

SO-101 Sim-to-Real Training: Isaac Lab & LeRobot Guide

Step-by-step guide to train SO-101 robot arm in NVIDIA Isaac Lab, collect teleop data, fine-tune GR00T N1.5, and deploy the policy on real hardware.

29/4/202611 phút đọc

isaac-labsim2realwheeled-robotreinforcement-learningrc-car

Wheeled Lab: Train RC Car Drift Sim-to-Real with Isaac Lab

Beginner-friendly walkthrough of Wheeled Lab — open-source ecosystem to train RC cars to drift, climb, and visually navigate in Isaac Lab and deploy zero-shot t

27/4/20269 phút đọc

humanoidreinforcement-learningsacworld-modelmujoco-playground

LIFT humanoid ICLR 2026: SAC pretrain + world model finetune

Reproduce LIFT (ICLR 2026): pretrain SAC in MuJoCo Playground, then sim2sim fine-tune a physics-informed world model in Brax for Booster T1 and Unitree G1.

24/4/202611 phút đọc

vlaqwendiffusion-transformeropen-sourceinference

Xiaomi-Robotics-0: Run 4.7B VLA on Consumer GPU

Step-by-step guide to install and run Xiaomi-Robotics-0 — a 4.7B VLA combining Qwen3-VL and Diffusion Transformer with 80ms real-time inference on RTX 4090.

22/4/202610 phút đọc

Nghiên cứu

agibotimitation-learningdatasetmanipulationgo-1

AGIBOT WORLD 2026: Real-World Dataset for Robot Imitation Learning

Hands-on guide to AGIBOT WORLD 2026 — the world's largest real-world robot dataset, the GO-1 ViLLA model, installation walkthrough, and the ICRA 2026 challenge.

21/4/20269 phút đọc

dual-armmanipulationdomain-randomizationsynthetic-datasimulation

RoboTwin 2.0: Complete Guide to Dual-Arm Manipulation Data

A comprehensive guide to RoboTwin 2.0 — the scalable synthetic data framework for bimanual robots, featuring MLLM code generation and strong domain randomizatio

21/4/20268 phút đọc

Nghiên cứu

vladual-systemrobot-manipulationdiffusion-policyopen-source

OpenHelix: Dual-System VLA for Robot Manipulation

A technical deep-dive into OpenHelix — the open-source dual-system VLA that pairs LLaVA-7B (System 2) with a 3D diffusion policy (System 1) via a learned ACT to

21/4/202610 phút đọc

gr00tnvidiavlafine-tuninghumanoid

Fine-Tuning GR00T N1.7 with EgoScale: Zero to Deploy

Complete guide to fine-tuning NVIDIA GR00T N1.7 with EgoScale's 20,854-hour egocentric video dataset — from setup to deploying your policy on a humanoid robot.

21/4/202611 phút đọc

humanoidreinforcement-learningunitree-g1isaac-labresidual-rl

SteadyTray: Unitree G1 Tray Carrying with Residual RL

Reproduce SteadyTray on IsaacLab — train a Unitree G1 humanoid to carry a tray without dropping objects using a two-stage Residual RL architecture.

19/4/202610 phút đọc