“A connected dash cam bicycle computer that collects your sensor data like heart rate and guides you to the nearest ice cream shop.
This project aims to build a bicycle computer using the Sony Spresence main board in combination with the LTE extension board and various other peripherals.
The main features are
Available features
- Capture a low resolution video stream and show it on a display with an option to take high resolution image for storage on a SD card
- Track the location via GNSS and combine the location with weather data and points of interest (POI) data received from cloud services via the LTE connection.
- Connect to bicycle sensors (currently heart rate) via Bluetooth Low Energy, show the data on the display and record it.
- Remote access to the camera and various data incl. location via MQTT
Partly available features / still under development
- Theft detection and notification via GNSS geofence, acceleration sensor and smartphone nearby monitoring
- Remote access to an audio stream from the microphone. Also ability to record audio to SD card.
Not yet started features
- Accident detection and notification using the IMU sensor (and AI?)
Motivation
After doing a project with nRF52 / nRF91 based MCUs on Zephyr, I wanted to see how other MCUs with their development environments work to extend my skills and have a comparison.
My attention was put to the Spresense board after reading about where it is in use (in space!) and the book “Starting low power edge AI with SPRESENSE” by Ota Yoshinori O’Reilly Japan - SPRESENSEではじめるローパワーエッジAI
I tried to build a battery based bird house camera before. Unfortunately power consumption and reliability was improvable, I wanted to give a try to the Spresense board as it is advertised as very low power. It also has camera capabilites and Sony is doing cameras very well.
After thinking about what kind of project would use the unique features of Spresense best, I came to the bicycle computer which is something I wanted to build for many years, but did not find fitting hardware yet. There seems to be not really anything similar you can buy out there which made it more valuable to start.
I set myself the goal to try to get most out of what Spresense offers as (unique) features. Also I put myself the guiding constraint to do an Arduino Envrionment based proejct.
A online research on dash cams for cars revealed that the camera quality at low light and lenses for a wide recording are differentiating selling points. The camera is a main part of the bicycle computer, but in this category I cannot win. Therefore I want to have cloud connectivity and connectivity to bicycle sensors as my differentiating points. Smart logic for theft detection and accident detection specific for bicycles would be another point.”