RoboMaster Hero Robot

Real-time electrical control system on an STM32H723, with quaternion-based IMU orientation estimation.

Nov 2025 – Apr 2026. Our team won the Third Prize in the RoboMaster University League (Shanghai, 3v3).

I built the electrical control system for my team’s swerve-drive hero robot in the RoboMaster competition.

  • Embedded control. A real-time control system on an STM32H723 that runs high-frequency control loops and integrates the robot’s peripheral hardware.
  • Orientation estimation. I derived the quaternion kinematics and implemented a Mahony filter from scratch to fuse IMU measurements, which gives stable orientation tracking. The orientation \(q\) evolves as
\[\dot{q} = \frac{1}{2}\, q \otimes \begin{bmatrix} 0 \\ \boldsymbol{\omega} \end{bmatrix},\]

where \(\boldsymbol{\omega}\) is the gyroscope’s angular velocity. The Mahony filter adds a feedback correction to \(\boldsymbol{\omega}\) computed from the accelerometer’s estimate of the gravity direction.

Code: github.com/hanbinzheng/hero