[WIP] Building an offline, portable mmWave STEM Demo for High Schoolers — Seeking engagement ideas!

Share your Projects!
1

Hi everyone,

I’m putting together a highly portable, completely offline Home Assistant setup for a STEM outreach Expo at a local university next week. The audience is high school students, and I have a quick 3-to-6 minute window per group to get them interested in IoT, sensor data, and local control.

The Offline Hardware Rig:

  • Server: Dell OptiPlex Micro (running HAOS)
  • Network: Old consumer Wi-Fi router (No internet connection, isolated WCUDemo SSID)
  • Sensors: Everything Presence Lite (EPL) mounted on a desktop stand
  • Lights: 3x LIFX BR30 bulbs in pivoting tabletop spotlights
  • Audio: HA Voice Preview Edition (running Piper for local offline TTS)
  • Display: A university-issued touchscreen laptop to display the dashboard

The Current Plan (The "Zone Defense"):

I’m mapping out three distance zones on the floor with painter's tape. As students step into different zones, the EPL tracks them and changes the lights (e.g., Zone A = Red, Zone B = Blue, Zone C = Green). It can track multiple targets at once to show off multi-target tracking. When they walk away and the sensor clears, the Voice Preview triggers a local TTS sign-off like "Target lost. Initiating power down." while the lights run a rapid rainbow sweep.

Where I Need Your Help:

I am worried that just having "lights change color when you move" isn't going to impress today’s high schoolers who grew up with smartphones and advanced video games.

How can I make this more engaging, gamified, or interactive using this specific hardware?

  • Gamification: Has anyone used the EPL to create a "stealth/ninja" mini-game where they try to beat the sensor?
  • Humor/TTS: What are some funny, unexpected, or sarcastic TTS phrases I could program Piper to say when it detects multiple people or when people approach too fast?
  • Dashboard Visuals: Any tips for showing off the raw data on the dashboard so it looks more like "hacking" or data science and less like a standard smart home app?

Any ideas for automations, dashboard layouts, or offline quirks to watch out for would be hugely appreciated!