Make finding a velo spot easy.
Live counts for every bike parking area on the Endress+Hauser Reinach campus — before you even leave your desk.
From a quick browser check to a bike you track
Everything works without an account. Signing in only adds the things an account enables.
Just the info
Open the page, see every stand’s live occupancy. No account, no install, nothing to configure.
Remembers your bike
Install the app. It remembers where you parked and reminds you when you need it. Still no account required.
Rank up
Sign in and the app starts counting. Days by bike. Kilometres on your bike and e-bike. Earn badges for every three days in a row, every ride under −10°C — more to come.
ESP32, solar-powered, LoRaWAN
A small sealed node per stand. No wiring, no WiFi, nothing to service.
- 01ESP32 microcontroller counts bikes locally
- 02Weatherproof casing, mounted on each stand
- 03Battery + solar panel — zero wiring, zero maintenance
- 04Reports via LoRaWAN — no WiFi needed, long range, tiny power footprint
Six sensors → the cloud → every open client
A classic edge-to-cloud fan-out. Nothing exotic, nothing proprietary.
Devices publish counts over LoRaWAN. The backend fans updates out over WebSockets in under a second. See architecture.md for details.
No faces, no plates, no tracking — not even yours
Plan A Counting without cameras
Non-image sensors (ToF arrays, LiDAR, mmWave radar, ultrasonic) or on-device CV. Only one integer — how many bikes the stand sees — ever leaves the device. There is nothing to lose or leak.
Plan B Fallback with camera
Only where no sensor configuration clears the bar. The image is counted by Azure AI Vision and dropped from memory in the same call. The code path exists as a safety net, not as the target.
The Booligans
Seven people, 48 hours, one working stack.