The Cloacina Project & Sensor Hub
We’re working to build inexpensive, locally-controlled logistics and environmental monitoring that re-connect communities to their environment and each other. Over the past 40 years, manufacturers have leveraged networked and computer logistics to coordinate globalized supply chains.Our goal is to build dramatically smaller, cheaper, and easier to use systems that perform the opposite function, coordinating local residents with the productivity of their environment. We’re starting with composting, because composting closes the loop between waste generation and food production. With careful management, compost is the safest way to dispose of the entire organic portion of household waste: food scraps including meat & cooked food, paper, cardboard, and even human excrement. Compost management can be assisted by environmental monitoring to yield better results in dense environments where space is at a minimum and pest management is crucial.
Data democracy projects such as DataDyne’s MIP, deployed in Chile as DatAgro, have demon- strated how cell phone text message delivery of environmental and weather news improves farm yields and empowers farm cooperatives. Our system, Sensor Hub, will do more than democratize access to environmental data, it will democratize the production and analysis of data. Sensor Hub’s integration of environmental monitoring, end-user feedback collection, and automated alerts will enable neighbor- hoods to aggregate residents’ diffuse time and interest into coordinated land management strategies. Sensor Hub’s data analysis package and server federation will bring quantitative measures and direct comparison to small-scale, user-driven R&D, inexpensively generating the documentation needed to turn backyard experiments into proven, replicable systems.
Our goals around composting are to use data collection to promote trust- neighbors know the treatment regime of a compost pile, and don’t have to worry about pathogens, parasites, or blights that might be added to the pile. Finished compost can be trusted for use on lawns and gardens without adding unwanted weeds or blights. Automated messaging tells maintainers when compost needs to be added, turned, or watered without constant checkups. As a whole, the system lowers the information barriers to composting and the expertise needed to maintain a high-performance, high temperature system at the neighborhood level. The community will see the direct benefits of reduced waste collection bills and increased soil fertility, and get to participate directly in distributed R&D to spread their ideas.
Cheap self-calibrating sensors, microcontrollers, cell phone networks, and server-side management software are advanced enough to hack together a workable almost-real time open- source monitoring system at a fraction of the cost of closed commercial systems. HOBOlogger and other commercial systems are good for environmental monitoring, but we can’t integrate them into our future logistics, or afford to embed their sensors. Open source development allows us to turn mutual interest into mutual aid, attracting developers working in communications, security, and the semantic web to help us build the net- work and analytic tools we need on top of existing platforms. It’s like spreading a frosting of our features onto someone else’s cake, and getting to eat the whole thing.
The biggest hardware problem in monitoring a compost pile is not the electronics, it’s the weath- erproofing. Our prototype sports a rafting dry bag with a project box inside.
HOW IT WORKS
We have a good work- ing prototype system built on the backs of several robust open frameworks: Our prototype consists of Arduino-based hardware and Drupal-based software. Self calibrating DS18B20 temperature sensors are housed in waterproof PEX piping and wired to an Arduino and ATT C168i GoPhone housed in a dry bag. This hardware connects to the internet using the SMS to e-mail bridge cell carriers provide. The e-mailed data is retrieved, parsed, and inserted into a SQL database at regular intervals by system software. There it is accessed by a R.J. Steinert’s Drupal module of Ward Cunningham’s Sensor Server, as well as the graphing package Flot , so researchers can design plots of data through the web. Dennison Williams has joined the software team to aid in security and project management.