aves: Acquisition, Visualization and Exploration Software

This python module makes it easy to acquire data from a serial port, visualize it on real time and record it. It also provides a module for visualizing data previously aquired with this tool.

Installation

pip3 install aves

Quick start

• We will use an Arduino to send data through the serial port.
• We will use aves to acquire, represent and record the data.

Arduino code

Create a file with the contents from example/arduino.ino.

This sample code reads the analog ports from the arduino board and prints them on the serial port.

Take note of:

• The SAMPLE_TIME (in ms): 100
• The Serial port speed (in bauds): 9600
• The Analog Read Resolution (in bits): 10
• The input range of the analog ports (in volts): (check your board, usually 5V or 3.3V)

Aves configuration

Aves is configured using a yaml file. Please take the example file from example/config.yaml.

The yaml file has four sections:

• version: Just a value, must be 2.
• input: Defines the aves input sources.
• gui: Controls the real time plotting options
• output: Defines the columns with sensor data that will be saved in a text file.

The input section

Aves uses two sources of information, the arduino and the computer clock.

For the arduino input, we have multiple parameters:

• baudrate: The baudrate specified in the arduino code.
• timeout: The seconds the python code will wait for data until it believes the serial connection has been dropped.
• columns: Aves must know what is the arduino printing on the serial port. columns is a list with as many elements as columns. Each element is defined by name which gives a name to the column and conversion_factor that is used to convert the number printed by the arduino to a magnitude meaninful for us. For instance, the conversion_factor is used in the example to convert the time printed by the arduino from milliseconds to seconds (0.001), and the sensor reads (in the range 0-1023) to Volts (in the range 0-5V): (5V/1023 = 0.004887586). The columns should be given in the order that they are printed by the arduino.

The computer clock does not have an entry, as it has no options. However, we should remember that besides the columns defined in the arduino section, we also have the time_computer column, useful to synchronize our experiment with other information.

The gui section

The gui defines the visualization options, including:

• The name of the column used in the x axis (x_column). It usually is the time given by the Arduino.
• Whether or not the zoom for all the subplots should be shared. It is often convenient to have it shared (zoom_all_together).
• The axes: The subplots available in the window. Imagine the subplots layed out in a grid. The first subplot (top-left) would be in row: 0, col: 0. The subplot below the first would appear in row: 1, col: 0, etc. Subplots may span several rows or columns, to make them larger, with the rowspan and colspan options, by default both set to 1. Each subplot should plot at least one column from the input, although more than one column can be plotted. The column names to be plotted for each subplot are given in columns. Additional plotting options (limits, labels) can be given in options.

Besides, there is the name of the window window_title and the refresh_time_ms that controls how often the GUI is refreshed.

The output section

Controls the columns that will be printed to the text file. Note how we have in the example both the computer time and the arduino time printed.

Run it:

python3 -m aves.realtime --port *Serial port where your arduino is connected* --outfile "test.txt"

Check python3 -m aves.realtime --help for all other command line options, for instance:

• --no-save Do not save the captured data to a file
• --outfile test.txt Save the capture data to test.txt
• --tmeas 600 Capture data for 600 seconds maximum (default: unlimited)
• --port COM3 Use the COM3 serial port
• --plot_every_n_samples 10 Wait for at least 10 samples to refresh the GUI
• --plot_win_size 200 Keep up to 200 samples in the plot (use 0 for unlimited)
• --config another.yaml Use another.yaml as config file.

Explore the acquired data:

python3 -m aves.explorer --filename "test.txt"

Known works using aves

• The prototype for fire detection developed at IBEC under the SafeSens project