Robots, Farms & Co-Design: The growBot Garden Project

Interaction Conference

Founded in 2008, the Interaction Design Association Conference brings together practitioners interested in all things around interaction design. Interaction 12 took place in Dublin, Ireland on 1–4 2012. Interactions 13 is set to take place in February 2013 in Toronto.

See all posts

  • ixda

Related posts:

small farm
One of the valuable things academic design research can do is to take on projects that would be impossible within the world of practice. That’s one of our guiding principles in the Public Design Workshop at the Georgia Institute of Technology—to explore the issues and opportunities of design that might go unexplored in the world of practice, and then find ways of communicating the insights we discover back a larger, more general, audience. There are, of course, lots of unexplored areas of design.

Our recent work focuses broadly on how to engage groups with little past experience in design. Specifically, we are interested in how we can use design to bring together diverse stakeholders in an issue to engage in speculation about the future, to collaboratively imagine and design possible futures. A large part of this involves education—how can we foster a level of design and technology literacy in our participants, and just as important, how can we learn from them about their practices, desires, and values?

Over the past year we have been exploring how robotics and sensing technologies might be used to support small-scale local agriculture. There are already many examples of sophisticated technological products and systems for agriculture, but most of these focus on large-scale farming. Moreover, most of these products and systems don’t translate well to small-scale agriculture—you can’t simply take a robotic harvester or a sensor network designed for a 10,000-acre farm and put it to work on a 1-acre farm. But it’s not only an issue of size; there are also different practices of farming and values at play that influence (or should influence) the design of technologies for small-scale agriculture.

As design researchers, for us the process is as important as the product that might be developed.

As design researchers, for us the process is as important as the product that might be developed. So, how do we develop a process to discover and invent these future agricultural technologies? Our answer to that question is through co-design and cooperative inquiry: working together in a collaborative fashion not just to create future products, but also to collectively explore together the issues and opportunities of local-small scale agriculture.This requires a long-term engagement and commitment to both a subject matter and a community, and a willingness to shed the authority that often comes with the title of designer or researcher.

Okay, so enough with the background, what have we been doing?

To address these questions we developed the growBot Garden project. The core part of the growBot Garden project is a series of workshops that bring together designers, researchers, artists, farmers, gardeners, and other food makers (e.g. cheese-makers) to collaboratively explore how robotics and sensing technologies might be used to support small-scale local agriculture. To date, we’ve held three workshops locally in Atlanta, GA and nine days of daily workshops as part of 2010 01SJ biennale in San Jose, CA—the largest new media festival in the United States, attended by over 25,000 people . Each of these workshops is a substantial event, taking weeks of planning and preparation.

The May 2010 growBot Garden symposium, our premiere event, provides an overview of our process. The day-long workshop was held at a local artists’ community called the Goat Farm, which is a complex of a half-dozen or so antebellum factory buildings housing studios, an independent café, a community garden, and of course, some goats, all tucked into a few acres of land in an otherwise industrial part of the city. About 18 participants joined us that day, together with the designers and researchers, making a group of about 25.

The day began with a short presentation surveying contemporary robotics technologies and then quickly segued to an activity in which the participants constructed physical models of farms and gardens. This storytelling relied on crafting as a means to prompt participants to share information about their work, while the act of making physical representations served as a conduit for imagining technological interventions in said spaces. Participants were provided cardboard squares for their models, along with a selection of craft materials such as paper, hot glue, markers, clay, string, and sticky notes. In addition, we had placed blank prompt cards with the question, “What Would Your growBot do?” on each table for participants to use throughout the day. We were pleasantly surprised when so many of our participants shared their ideas on the prompt cards as they introduced their farm/garden model to the rest of the group. Some participants shared as many as eight ideas – a welcome jumpstart to the day’s imaginations. After learning about their farms and farming practices, we shared a series of science-fair-like demonstrations that we had developed to introduce them to the capabilities of robotics and sensing technologies. Members of the design research team staffed these demonstrations of machine vision, proximity sensors, tweeting plants, and a roomba, while answering questions from participants.

Growbot researchAfter a leisurely lunch of local organic foods, participants began prototyping ideas for robots on their farms and in their gardens. Guided by the research team, participants divided into groups, creating sketches of their ‘growbot’ and moving into cardboard prototyping. A variety of materials were available for this exercise, including recycled children’s toys, thrifted wires, and craft materials. Later in the afternoon, participants presented their concepts and prototypes back to the group. By the end of the day, over two dozen idea had been proposed and six of them had been developed into physical models. Each concept and prototype uniquely reflects the specifics of small-scale agriculture, providing insight into what distinguishes these practices and too, begins to hint at the values and motivating desires of our participants.

Each concept and prototype uniquely reflects the specifics of small-scale agriculture, providing insight into what distinguishes these practices and too, begins to hint at the values and motivating desires of our participants.

For example, many of our participants did not use pesticides, as they are organic farmers. Of course, not using pesticides makes dealing with insects more difficult, so it should not be surprising that several of the concepts focused on pest control. In addition, as several participants noted, some insects are beneficial to the crops, so indiscriminate elimination of insects isn’t desirable. The Bio-Rover prototype was one idea developed to balance the issues of insects on the farm. This robot was designed to use a series of vision sensors to collect information about pests throughout the farm environment. Instead of having the robot take action on the pests, the robot would return the information gathered to the farmer, who would then make decisions regarding the pests detected—for example, which should or should not be eliminated and how.

This workshop, and the subsequent ones, have taught us a great deal about the needs of small-scale farmers, about the ways in which they do their work and the structures of their farms, and what they think is important. As we continue on with our endeavor, we are beginning to work on ways to share their ideas through videos, print media, and websites. We also continue to work with some of the participants, to further develop design concepts and our methods of co-design. What effect this sharing will have is unknown, and in this way, the project exemplifies the kinds of work academic design research can contribute to design practice. What we will produce is not a product or service, perhaps not even a set of product or service guidelines, but rather a series of novel insights into the challenges and possibilities of designing to support local small-scale agriculture, and too, a methodological framework for engaging in co-design and cooperative inquiry.
Growbot collaboration
Now, you may be asking,  what does this have to do with interaction design or user experience? The answer is everything. As interaction design continues to expand its boundaries, we need examples of and models for what interaction design and user experience might be like beyond interfaces. For us, these workshops are themselves a kind of interaction design, through which we discover and invent how to work together with various publics to create possible futures. The workshop in effect becomes the product, becomes the interface, as we develop and refine its structure and flow, prototyping and testing it over and again. User experience here is vital, but it’s not with a product or service, it’s with an event; it’s the user experience of participating in design. So, in addition to exploring how robotics and sensing technologies might be used to support small-scale local agriculture and processes for co-design, we are also working to explore the very practice of design itself.

Interaction 11

If you want to meet Carl DiSalvo in real life: he is one of the speakers at Interaction 11. It is the fourth annual conference hosted by the Interaction Design Association (IxDA). Each year, IxDA aims to gather the interaction design community to connect, educate, and inspire each other. This year it is held in Boulder, Colorado (USA).

Additional photos courtesy of Carl Di Salvo

Carl Disalvo

One comment on this article

  1. Not a comment but an acknowledgment that the growBot Garden project is in every way collaborative project, team members include Laura Fries, Thomas Lodato, Beth Schechter, and Thomas Barnwell – gotta be true to the collective spirit