Hands-on Training
Humanoid Robot Reinforcement Learning Training
Master Programming Reinforcement Learning for Humanoid Robots in 3 Days
Barcelona, Spain
April 8 – 10, 2026
25 seats per cohort only
Seamless Simulation-to-Real Humanoid Training
Train RL policies in high-fidelity simulation and deploy them on real humanoid robots.
Reinforcement Learning for Humanoid Control
Learn how RL enables walking, whole-body control, imitation, and actions.
Discover what humanoids can actually do!
Designed to help institutions evaluate what humanoid robots can realistically do today.
This training is organized by
in cooperation with
Training Details
This training focuses on reinforcement learning for humanoid robots, combining advanced simulation with real-world deployment. You will learn how to train and validate RL-based control policies and apply them to real humanoid robots.
You will work through a full humanoid robotics pipeline, starting from low-level robot access and ending with learned control policies running on physical humanoid robots.
During the training, you will:
- Use a Unitree G1 EDU development PC and Unitree’s SDK to command, monitor, and control the robot at a low level
- Build and train reinforcement learning policies for humanoid locomotion and motion control in simulation
- Deploy learned policies on real humanoid robots
The training combines high-fidelity simulation, real robot execution, and hands-on debugging of simulation-to-real gaps that commonly appear in humanoid systems.
Who Should Attend
✔ Research institutes working on humanoid robotics
✔ Robotics labs & R&D teams
✔ Organizations evaluating or deploying humanoid robots
✔ Engineers and researchers moving beyond simple demos
Humanoid Platforms Used in the Training
Unitree G1 Edu
Developed by Unitree
Physical Robot
Kangaroo
Developed by PAL Robotics
Physical Robot
Training Highlights
Practical Experience
Hands-on sessions with real humanoid robots, guided by instructors.
Expert Guidance
Learn from experienced robotics developers actively working on humanoid systems.
Certificate of Completion
Earn a shareable certificate recognizing your proficiency in humanoid reinforcement learning.
Training Schedule
🧭 Day 0 — Required Online Preparation
Access opens 2 days before the on-site training
Early access to the full course content and training environment. You will:
-
Complete the first hands-on simulation lesson
-
Get familiar with the humanoid simulation workflow
Goal: Arrive on Day 1 ready to work.
🗓 Day 1 — RL for Humanoid Locomotion and Teleoperation
WEDNESDAY, APRIL 8
Focus: Learning-based motion at the locomotion level
You will work on:
-
Reinforcement learning based walking control for the Unitree G1 humanoid robot using Isaac Lab
-
Teleoperation and imitation learning using Isaac Lab Mimic
➡️ Outcome: You will train and evaluate your first stable humanoid locomotion policies in simulation and understand how they translate to real hardware.
🗓 Day 2 — RL for Humanoid Whole-Body Control
THURSDAY, APRIL 9
Focus: From locomotion to full-body coordination
You will work on:
-
Full-body control from motion tracking using reinforcement learning
-
Deployment of learned policies to the robots
➡️ Outcome: You will run whole-body reinforcement learning policies on real humanoid robots and understand the system-level integration required for stability and safety.
🗓 Day 3 — Vision-Language-Action Models and MuJoCo Lab
FRIDAY, APRIL 10
Focus: Generalization and research-level experimentation
You will work on:
-
Vision-Language-Action models for generalized humanoid skills using Gr00t
-
Training and validation of humanoid behaviors in MuJoCo Lab
➡️ Outcome: You will understand how learned humanoid skills can generalize beyond scripted demos and how these methods are used in current research and development.
The Construct Robotics Institute Certification
Upon successful completion of this training, participants receive a shareable Certificate of Completion issued by The Construct Robotics Institute.
This certificate recognizes demonstrated proficiency in:
- Reinforcement learning for humanoid robot control
- Simulation-to-real deployment workflows
- Hands-on operation of real humanoid robots
venue
On-site Training in Barcelona
Master humanoid robot reinforcement learning—hands-on training with real robots,
real intelligence, in vibrant Barcelona.
What Past Participants Say
Testimonials
Join Over 10,000+ Engineers & Learn Robotics Here
Trusted by:
Enroll now
Humanoid Robots RL Training
Location:
Barcelona, Spain
Dates:
April 8 – 10, 2026
Time
9:00 – 17:00 Central European Time (CET)
Cohort Structure
Two groups, limited to 25 participants per group
Hands-on robot sessions organized per group
Prerequisite Knowledge
-
Linux, Python, C++, ROS 2
-
Familiarity with DDS
⚠️ This training is not suitable for ROS 1–only users.
Pricing
-
Early Bird: €1,999 (until Feb 28, 2026)
-
Standard: €2,499 (until Mar 30, 2026)
Fee Includes:
3 days of live training guided by robotics experts
Two real humanoid robots: Unitree's G1 and PAL Robotics’ Kangaroo
Many Hands-on Exercises with Simulated Robots
15 days of access to the course content after the course
Lifetime Access to the Code Developed During the Training
Fast Support & Guidance from Instructors
Training Completion Certificate
Lunches & coffee breaks
Questions? Email us at info@theconstruct.ai
FAQ
Frequently Asked
+ What is the language used in training?
English
+ What are the prerequisites for this training?
You require a basic knowledge of Linux, Python, C++ and ROS 2, and be familiar with DDS. If you do not have this knowledge, you can use our online courses: (*Please note that completing these courses is optional, and you may use any other learning resources. Paid courses are not included in the training fee and cannot be accessed for free.)
- Linux for Robotics (Free)
- Python3 for Robotics (Free)
- Advanced Modern C++ for Robotics
- ROS 2 Basics
- DDS for Robotics
IMPORTANT! Do not enroll in this training if you are not comfortable with ROS 2. This training is not appropriate for individuals with ROS 1 knowledge only.
Equipment You Will Need:
- Computer: Linux, macOS, or Windows – all are fine
- Browser: Google Chrome or Firefox
- Software: No installation required in advance
- Optional Accessories: Mouse, keyboard, or other peripherals as preferred
+ Do I need to install ROS in advance?
No. You only need to bring your laptop and don’t need to have ROS installed. You can use Windows, Linux, or macOS.
+ Refund Policy
Participant Cancellation
- Cancellations made more than 10 business days before the start of the training are eligible for a full refund.
- Cancellations made less than 10 business days of the training start date will incur a 50% cancellation fee.
- No-shows or cancellations not communicated in advance will be charged the full training fee.
- Participants who cancel their registration for a scheduled training session will not be eligible to participate in subsequent training sessions.
Training Provider Cancellation
- In the event that the training is canceled by the provider, a full refund of any fees paid will be issued.
Minimum Enrollment Requirement
- This training requires a minimum of 20 participants to proceed. If this number is not met, The Construct Robotics Institute reserves the right to cancel.
- In such cases, at least 10 days’ notice will be given to all registered participants, and full refunds will be provided.