Spaces in Progress (SIP) was a year-long research project embedded in the Department of Architecture at the Rhode Island School of Design (RISD) in Providence, Rhode Island. The project was made possible through the support of Arkio and their educational partnerships program. Through this project, we were able to gain knowledge and first-hand practical experience in the use of mixed reality technology. The research framework allowed us to become familiar with state of the art uses and case studies of this tool. We were then able to insert our contribution as a critical application of MR to the design and fabrication of architectural assemblies.
This research project is embedded into a larger set of questions involving the evolution and implementation of CAD in architecture. There has always been a tension between the digital and the analogue, between control and participation, and between design complexity and construction labor. The thesis of my current work is that computational design should not be synonymous with any particular aesthetic or project type, and that we should work more critically to avoid alienating these technologies and ourselves as architects from the physical and cultural challenges of the communities we strive to serve. These speculations involve messy, fluctuating workflows of feedback and response that could promote the creation of social and cultural value centered around the process of designing and making. The project team was composed of myself, as faculty and project lead, and Magnolia Moskun, as a research assistant hired through the fund.
The project team was composed of myself, as faculty and project lead, and Magnolia Moskun, as research assistant. An allowance for the purchase of materials was also part of the award.
Process
-
The first step in this year-long project was to research case studies that have used a variety of augmented, virtual, and mixed reality as their main vehicle. These projects vary in aim, and were archived and catalogued as such for internal discussion. The process allowed us to understand the software and headsets that were most commonly used by designers around the world, the most common set-ups and workflows, and the project trends ( or what this technology has been found to be useful for ). We engaged in this process to form our first ideas for the project, but without having used the Arkio platform yet. This intensive search allowed us to trouble-shoot later on in order to obtain the desired project results, instead of letting the platform force the project in any direction.
-
We then moved on to test the technology. The learning curve was somewhat steep for us as non-users. It involved several steps to get the software and headsets up and running. We then had to learn how to interact with the Arkio platform and make it communicate with our 3D modeling software of choice: Rhino. Some of the first tests involved importing existing geometry from previous projects to view them from within Arkio, and creating simple 3D objects. A key feature of Arkio that we continued to exploit for our experiments moving forward is the capacity to “overlay” digital geometry in real physical space. This feature in Arkio is known as “passthrough”.
- We were able to inhabit such geometries, experiencing them at different scales. The tests became increasingly complex, as we tried to test the limits of user control and the translations from orthographic modeling/drawing into Arkio. The latter was often very hard to achieve, prompting a series of workarounds. The next image shows a digital experiment in creating a “generative” tree form out of repeated parts. The parts formed branches, and the branches were copied. Although there was no consideration to “gravity” or joinery in this test, it allowed for a form of intuitive, spatial sketching through the aggregation of modules.
- After understanding the principles of importing geometry, copying objects, and modifying their visibility, we moved to physical space. Blocks (fabricated from foam sheets) were used as the first construction module due to their lightness and resemblance to brick. The experiments involved controlling the projection of geometries in physical space and understanding how to work simultaneously. This proved successful, and led to many iterations. One important discovery was the “center mark” that we drew on the floor, and was used to calibrate the placement of each virtual object in physical space, and our own positions in relation to it.
- Having understood this workflow, we began to introduce formal complexity, and develop a computational script to design a called this series “better” block wall, always produced in our 3D software, then projected into space. We called this series “Wall Monsters”.
- The results were positive, and gave us the confidence to split the blocks in half and aim for a better definition in the forms. We also introduced vertical shifting, as a way to rely constantly on the mixed reality projection, instead of using the bottom row as a template for repetition.
- In parallel with these progressive aggregations, we also explored spatial LiDar scanning, as a way to capture spaces and bring them into virtual reality as “sites” that could be designed on. This emulates recent advances in 3D scanning for the generation of “as built” conditions at the beginning of design projects.
- The last assemblies in the sequence were larger and more complex, needing more planning, additional blocks, and CNC fabrication. We salvaged and reused a couple of sheets of plywood from a previous project. The final iteration involved hosting a larger collaborative experiment to share the found workflow and test some of our previous assumptions. It was largely a success.
