DigiKey is reinforcing on applied AI education with a free, 90-minute workshop led by Shawn Hymel: “Train a balance bot with reinforcement learning.” The session, scheduled for Thursday, Aug. 13, 2026, at 10 a.m. CDT, targets engineers, makers, and students looking to move from theory to practice in reinforcement learning (RL) and robotics.
Workshop focus: sim-to-real RL for balance bots
Reinforcement learning enables agents to learn complex behaviors, like balancing, walking, or recovering from falls, by interacting with a simulated environment and optimizing a reward signal before deployment on real hardware.
Hymel’s agenda centers on a complete sim-to-real pipeline using the M5Stack Bala-C platform. Attendees will:
- Import a 3D model of the M5Stack Bala-C into the MuJoCo physics simulator
- Train a neural network policy using a multi-phase PPO curriculum
- Deploy the trained actor network to an ESP32 microcontroller via Arduino
- Walk through the sim-to-real workflow that underpins modern AI-driven robotics
Hardware and access
Participants who want to follow along live are encouraged to use either the M5Stack Bala-C Balance Bot or the M5Stack Bala2 Fire Self-Balancing Robot. Those unable to attend can still register to receive the on-demand recording; the event is free and open to all: register here.
Broader push: six-part RL education series
The webinar complements DigiKey’s newly launched six-part YouTube series with Hymel, which begins with this first episode and guides viewers from simulation through hardware deployment. A companion written tutorial is available on DigiKey’s Maker.io site.
By the end of the series, participants will have a working remote-controlled balance bot that learned its control policy entirely through trial and error in simulation—an outcome DigiKey positions as a practical on-ramp for engineers and students entering RL-enabled robotics.
