Hi! We’re really happy about this! We just installed a sensor+totem for counting bikes, and we are proud to say we did all of the software in our office, in Buenos Aires, Argentina.
This is all part of our strategy of transforming the city to an “open by default” mode, and specifically, the first stage of our sensor based project for measuring the bike lanes’ traffic of the city.
This stage comprehends the installment of a sensor and a visualization totem on the bike lane of del Libertador Ave., right next to the Law Faculty of Buenos Aires University.
The project, lead by Mariano Vassallo and Sebastián Mandaradoni (New Technologies department, eGov Office), has the goal of collecting real-time information of bike use on different bike lanes throughout the city.
The measuring device consists of a pressure sensor connected to a hose that crosses the bike lane, buried on the ground. This hose can detect bicycle wheels according to the pressure difference, and registers that as an event with a time and date.
The device then sends the information to the totem that is a block away, and saves it in a centralized server so it can be put to other uses. It has a metallic structure more than 2m tall (6.5 ft) and shows, through LEDs, the amount of bicycles that pass through that point every day and the yearly agreggate. Each day, at 12am the daily counter is reset.
What can we do with the data
- Publish it on the web
- Add it to the other measurments that are already being produced since 2010 on different points of the bike lane network
- Calculate CO2 emissions savings
- Evaluate distribution of traffic, by day, time, season.
- Get the weather conditions of the day
Building the sensor and the totem / Nerd stuff
- We used a Freescale MPX5050 pressure sensor connected to Arduino UNO
- We used Rasberry Pi + socat + Bash scripts and AWK for data transfer
- We deployed a Python Server to collect data and send notification using ZeroMQ
- We made a controller with an Arduino UNO for the sign
How does it work:
The device is a microprocessor with a pressure sensor connected to a hose that crosses the bike lane. Measuring the hose’s pressure difference it can detect bicycle wheels stepping on it. We had a long calibration process. Although technology simplifies many variables, the weather (temperature, clouds, rain, humidity) affected the sensor, which took us days of testing of adjusting from November to today.
Open by Default
You can access and fork our sensor source code at GitHub/GCBA
Questions? Comments? @eGovBA