In this month’s installment of the Innovation of the Month series, we explore Carnegie Mellon University’s research study on 3-D visualization software for city planning and urban design applications.
MetroLab’s Ben Levine spoke with three faculty members from Carnegie Mellon University: Stephen Quick, research associate at the Remaking Cities Institute and adjunct professor in the School of Architecture; Kristen Kurland, professor in the Heinz College and the School of Architecture; and Ray Gastil, director of the Remaking Cities Institute, professor in the School of Architecture and Director of the Pittsburgh Department of City Planning from 2014 to 2019.
Ben Levine: Can you please describe the objective of his project, and who you have involved in it?
Stephen Quick: The objective of the 3-D visualization project was three-fold: 1) to research state-of-the-art 3-D urban planning software; 2) to test and evaluate the software; and 3) to develop a protocol for implementing 3-D software for city planning departments, planning consultant, and university researchers. The final product was a report describing and analyzing the most appropriate 3-D programs for everyday use in city planning, urban design and urban development.
An interdisciplinary research team was created that included Carnegie Mellon University (CMU), Pittsburgh Department of City Planning, SimCoach Games (a private gaming consultant) and Esri. CMU was the team leader, including faculty researchers from the Remaking Cities Institute (RCI), the Heinz College for Information Systems, the School of Design, and graduate students from the Master of Urban Design (MUD) program, the School of Design and the Entertainment Technology Center (ETC). Funding was provided by grants from the Heinz Endowments, Deloitte Foundation and the CMU Metro21 Smart Cities Institute.
The 3-D visualization research project built upon an earlier collaboration in 2015 between CMU, Esri, SimCoach and the city of Pittsburgh to produce a pilot simulation using existing 3-D tools to test development scenarios on a commercial corridor in downtown Pittsburgh.
Simulation of Downtown Pittsburgh, a portion of the geo-locational citywide 3-D model.
Levine: Who is the intended user in your deliverable? How did they shape your approach to the project?
Ray Gastil: Our users include anyone who could use and benefit from better visualization and communication of planning and design information. That included city planners, urban designers, architects, developers and others in the real-estate and building industries. It also included the general public, who would have an interest in changes to where they live, work and play. We were looking for visualization and communications tools that would make it easier to understand development plans and proposals, and perhaps also have the capability for back-and-forth dialog.
Pittsburgh’s Department of City Planning actively participated as a team partner throughout all phases of the project. City Planning was bringing GIS staff with the capability of integrating visualization tools into the department’s planning and zoning operations. The research team, likewise, valued the department’s participation for understanding day-to-day needs, software capabilities and concerns that ranged from internal ease-of-use, communication with professionals outside the department and concerns for maintaining confidentiality when needed.
The Department of City Planning assisted the team with honing the project’s goals and objectives for their practical application. The department also partnered to help the team develop and test the idea of a prototype VR app to be used as a working tool for urban designers, selecting recommended softwares for testing in an urban design studio setting at CMU while simultaneously testing the same software within the department’s activities. Finally, the department worked with the team to determine an external communications and civic engagement approach.
Experimental VR app capable of environmental, use, and design alternatives using gaming software (Entertainment Technology Studio, Phase II).
Levine: There must have been a lot of data you could have included in your project. How did you choose which data to use and how to present information to a user?
Kristen Kurland: Yes, there was a lot of data used in many software applications and it came in various types: survey information, CAD drawings, BIM models, GIS layers, lidar points, renderings, photographs, teleportation points and text. Data was visualized in different ways for each phase of the project and data management became a task in itself.
The team developed a wiki website for internal documentation and communication during the project. As each phase of the work was completed, a report was added to the website describing the process, outcomes and persons involved.
During Phase I (benchmarking and documentation of existing 3-D software) Heinz College and School of Architecture students and faculty reviewed more than 30 off-the-shelf 3-D software programs for modeling and representation. Interviews were conducted with city planning departments, universities, urban design and planning firms, and software companies. Results were posted via a software matrix and website links to project examples.
Phase II (use of selected software for developing a virtual reality design tool) was work completed by students and faculty in the Entertainment Technology Center. VR technologies were used as planning tools to demonstrate how VR could create a sense of place for an urban development corridor selected by the Department of City Planning. Additionally, various user interface/user experience (UI/UX) techniques were developed to test on-the-fly design changes (by guests in the VR experience) and how that might impact sense of place.
Phase III (testing of modeling, visualization and communications software) involved work by students and faculty in the School of Architecture Master of Urban Design program. Through a progressive series of design assignments, for the same Phase II urban development corridor, the students first tested geospatial software and then advanced to 3-D modeling, visualization and VR software.
Phase IV (public communications and operationalizing 3-D visualization technology) was the development of a beta website utilizing 3-D material by faculty and students in the School of Design for planning department public communications.
Testing spatial impact of urban design scenarios in VR using a variety of modeling softwares (Master of Urban Design studio, Phase III).
Levine: What were some of the technical challenges you overcame in this project?
Kurland: We found several technical challenges within the software applications that presented roadblocks for visual understanding and communication. Because each discipline uses tools specific to its function, it was impossible to identify a one-stop solution. Software that cuts across modeling, virtual reality and communications had protocol problems when working across various platforms and hardware.
We found challenges working between modeling software that used different data references. CAD [computer-aided design] and BIM [building information modeling] software used by architects, engineers, contractors and developers use Cartesian coordinates as a point of reference (0,0,0). GIS software uses locational reference (latitude, longitude, altitude). Because of this, digital models are not easily overlaid with each other, making it very difficult for planners to work with unfamiliar spatial-referenced architectural models and drawings created in CAD or BIM.
The data and software used are complex and require dedicated people with technical expertise. Without in-house expertise there will be functionality limitations with all of these visualization and communications tools.
Levine: How did you choose the best way to visualize this data to a user?
Quick: We found there were two visualization issues to contend with: how to communicate compatibility issues between software produced by many vendors and how to record the project when the content and outcomes would cover a wide variety of data types and content.
Overcoming the protocols and compatibility issues led the team to develop visual workflow diagrams to illustrate user pathways between the four types of software we studied: spatial and locational modeling, virtual reality, and communications apps. The diagrams allow any user to begin with a CAD or BIM modeling software and work through VR and communication tasks with software developed by others or begin with communications and work back to whichever modeling or VR program available to the user. The idea was to create a universal workflow between all the tested apps that prioritized functionality over app families of software.
Visual workflow diagrams to illustrate user pathways between the four types of software studied: spatial and locational modeling, virtual reality, and communications apps. The diagrams allow any user to begin with a CAD or BIM modeling software and work through VR and communication tasks with software developed by others or begin with communications and work back to whichever modeling or VR program is available to the user.
Levine: What were some of the most important resources from CMU and the city of Pittsburgh that you leveraged through this project?
Quick: Our multidisciplinary team and approach were the most important, not only within CMU but also with the Department of City Planning. The team had experts in both spatial and locational software, app developers, and practicing professionals along with the resources of three colleges within the university. City Planning contributed with policy, communications and personnel who worked side-by-side with the CMU team. We collectively worked with graduate student teams to develop and test applications and in the process learned a lot. For example, we learned where instruction and expertise were needed and that the languages of planning and design are not universally known or understood by those in software development. We also learned how much information a person can absorb, and the time it takes to develop a command of visualization material.
The team was encouraged by the response and enthusiasm we received from both the professional and the software communities with respect to the projects and its potential impact. The fact that a major research university and a progressive city were teamed for a purpose useful to the planning and design communities was appreciated and encouraged. The approach also led to the project’s credibility and continued communication with several software developers about future software needs.
Levine: Where do you imagine this project going next? What next steps will the team take?
Quick: We are hopeful that the software development community begins to understand that planning and urban design professionals are in need of user-friendly software, beyond the apps available to architects, engineers and product designers. These professionals would also benefit from economic development and BIM capabilities integrated into planning software for more functionality.
The CMU research team foresees that 3-D software will soon improve its ability to document context and provide higher-quality libraries of realistic objects and textures for importing into design models. VR and real-time graphics are forecast to be the next “new wave” of innovation with the ability of inserting models of buildings and public spaces into real contexts. This is now happening in gaming, the industry leader in these software advances.
For city planning tasks, location-based geospatial platforms currently offer the greatest potential for both geospatial and 3-D modeling compatibility. This platform is now capable of real-time design and modeling (e.g., of buildings and other design objects) and VR positioning. Future geospatial software development likely will focus on the commercialization of 3-D software tools developed specifically for city planning departments, urban design firms and university urban design programs.
We anticipate that our next steps with visualization will focus on bridging the communications gap between professionals and the public. For most people, these technical programs are beyond comprehension and paint an idealistic picture, yet we also know the public understands 3-D visualization significantly better than 2-D maps and perspective renderings. Realistic visualization is understandable.
