01About 02Tournament 03Technology 04Partners 05Contact
EngineAI × URKL // Shenzhen, China

AI. Robotics. Fighting Spirit.

A Polish team of engineers and AI researchers building motion-control systems for humanoid robots in an international tournament of physical humanoid combat. This is Physical AI in action — intelligence that operates in the real world.

Brave Mind Fighters — humanoid robot
SYSTEM: PHYSICAL_AI // STATUS: TOP 32 / 128
Scroll
128
Teams at the start
TOP 32
Advanced to the next round
3–4
Our place in qualifiers
Polish team in the tournament
01 Who we are

Poland
in global
competition

We are an interdisciplinary team of engineers, researchers and AI technology creators from Poland. We build systems that let humanoid robots move in a dynamic, stable and effective way under the conditions of physical competition.

Our goal is a system that lets a robot not just perform individual moves, but react to the changing situation in a fight, keep its balance, make decisions and adapt its strategy in real time.

Humanoid robotics Reinforcement Learning Motion Control Motion Retargeting Physics simulation Sim-to-Real Autonomous systems HW/SW integration
02 Our moves

What we taught the robot

From a single move to a full sequence. Below we show behaviors we trained the robot to perform — the result of training in simulation carried over to real hardware through sim-to-real.

Moves learned by the BMF robot
▶ BMF — robot's learned moves
How it works
03 What the tournament is

A combat gala —
the fighters are robots

The tournament organized by EngineAI / URKL in Shenzhen is an event where humanoid robots compete in physical bouts resembling modern combat-sports galas. The difference? The fighters aren't humans, but robots controlled by systems developed by engineering teams from around the world.

Official trailer of the humanoid robot fighting gala
▶ Official gala trailer

From mocap
to the robot

The motion-retargeting pipeline — from a motion-capture recording, through skeleton tracking and verification of posture, timing and continuity, to a movement ready to transfer onto the robot.

Motion-retargeting pipeline from motion capture
▶ Motion retargeting — from mocap to robot
// 01

Balance

Walking and keeping stability in guard and after physical contact.

// 02

Reaction

Responding to contact, offensive and defensive moves in a dynamic clash.

// 03

Recovery

Regaining stability after strikes in an unpredictable environment.

04 Our technology

Physical AI Stack

Five key areas that let the robot learn movement, keep stability and carry out tasks under physical contact.

// 01

Motion control

Motion policies that let the robot walk, turn, take a guard and react to loss of balance.

// 02

Reinforcement Learning

Reinforcement learning that trains the robot's behaviors in simulation before they reach the physical machine.

// 03

Motion Retargeting

Analyzing human movement and transferring it onto the robot's body while respecting mechanical constraints.

// 04

Sim-to-Real

Validation and testing that minimize the gap between the virtual world and the physical robot.

// 05

Physical AI

AI systems acting in the real world — decisions shaped by physics, contact, dynamics and hardware constraints.

// PROVING GROUND

Tournament =
laboratory

We treat the fights as a proving ground for technology with far broader use than sport.

Our mission

Poland can build world-class technology in one of the most important directions of AI — Physical AI and humanoid robotics.

05 Partners

Supporting
our journey

HumanTech Center, SWPS University
HumanTech Center · SWPS
Warsaw University of Technology
Warsaw University of Technology
Join us

Join us at
the start of the journey

We're looking for partners, sponsors, engineers and researchers who want to build the future of humanoid robotics together. It's a chance to be part of a project combining AI, robotics, sport and Polish innovation — with global media potential.