Yam Propagation, Monitoring And Visualizing

Yam yields have decreased as a result of disease and bad seed. We designed and prototyped this for rural farmers in Nigeria. The objective of this project is to address some of those issues by using hydroponics for propagation of seedlings by tissue culture clones and plant cuttings. A further objective is to use as much local materials and a monitoring system for research as well as assisting in production.

The first version of these use sand based hydroponic beds as shown below:

The grow beds are raised ergonomically and are filled with a hydroponic nutrient solution. Nutrient solution can be exchanged and aerated in two ways: emptying of bed into the collection bucket and manually recirculated or via sponge filters which will aerate and circulate the water within the bed..

It is desired that an integrated smart monitoring system keep track of variables in the garden:

• water/root zone temperature
• ambient temperature
• ambient humidity
• light level in photosynthetic photon flux density ppfd
• video timelapse

Due to unreliable power supplies as well as poor Internet networks, it is required that the system run on solar energy and use a cellular network for data transmission. The schematics of the monitoring system are shown below

A deep cycle battery is charged using the solar panels for use during low sun conditions. An LCD and LED displays are also included for direct monitoring for local users without smartphones. The functional system is shown below

Equipment

There are two potential styles for the controller. The controller and solar panel integrated as one unit or the controller and solar panel as separate units for say roof mounting. These are shown below with the LCD screens shown as well.

The sensors connect using audio 3.5mm connectors or they will be soldered onto the board. The batter will be inside the enclosure. A slot for the cellular sim card is located on the other side of the enclosure.

Data Collection

Sensor readings are sent from the garden to the server once every thirty minutes. An mjpeg image is also transmitted to the server once every thirty minutes. The images will be converted to a time-lapse video stream at the end of day. Transmission of data will be done using the cellular network. Data will be sent to API application at http://api.kijanigrows.com for logging, visualization and alerting. Data is stored indefinitely and is retrieved from the API as json objects for further custom presentation.

Sensor and video data is also stored locally on the controller in a micro SD Card for off-line analysis.

Data Visualizations

The system sensor data will be visualized in different ways. The API based visualizations are graphical and interactively change color and shapes based on the sensor conditions. The icon dashboard is shown below

A typical garden functional animation is shown below

Typical trending visualization is shown below

LCD Visualizations

As WIFI and cellular networks are not guaranteed, it is desirable that a local LCD display is added. This is typically shown below:

LED Status Indicator

A multicolor LED will be used to indicate the overall status of the controller. Typical indicator status conditions are shown below

Alerting

Remote alerts messages are sent over the Internet. Alerts are easy to configure using the sensor details interface screen shown below.

Exceeding the high or low set-points for each sensor will trigger alerts. Currently only twitter alerts are supported. Email, voice and text messages will soon be included. Typical twitter account are configured using the device admin screen shown below:

Typical twitter alerts are shown below:

Project Task List