Developing a User Interface for Live 3D Mapping of Wildfires
Total Page:16
File Type:pdf, Size:1020Kb
Developing a User Interface for Live 3D Mapping of Wildfires An Interactive Qualifying Project submitted to the Faculty of WORCESTER POLYTECHNIC INSTITUTE in partial fulfillment of the requirements for the Degree of Bachelor of Science by Ben Newmark, Chris Kirven, Jordan Hollands and Victoria Crowley October 12, 2017 Report submitted to: Professors Melissa Belz and Lauren Mathews Worcester Polytechnic Institute Stephen Guerin President and Founder Simtable, LLC Santa Fe Project Center This report represents the work of four WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review. For more information about the projects program at WPI, please see: http://www.wpi.edu/Academics/Projects. Executive Summary In June of 2013, nineteen firefighters time. LiveTexture will aggregate imagery of died trying to contain a wildfire in Arizona an area from several sources and generate a (Coe & Merrill, 2013). An after action three-dimensional model of that area, with report of the tragedy revealed a local relevant information overlaid onto the resident was streaming live video of the fire model. The goal of LiveTexture is not only as it burned on the mountainside (S. Guerin, to allow firefighters to communicate faster personal communication, April 5, 2017). but also to enhance the firefighters’ The video displayed a key shift in wind understanding of the situation by giving direction that ultimately caused the fire to them a visual and intuitive view. turn back onto the firefighters on the ground Currently, Simtable has not yet and smother them. If the fire crew had the developed a user interface (UI) for most up-to-date information, they could LiveTexture. The UI, or web-based screens, have communicated the wind shift to the represents the physical means through which ground troops and evacuated them to a safe users will interact with LiveTexture. The area. Unfortunately, the crew did not adoption and utility of LiveTexture is discover the video until after the fire had largely limited by its UI design; For end already taken all nineteen lives. users, a useful software has a UI that enables Because emergency situations cannot users to accomplish their goals by offering be predicted, emergency personnel need to the functionalities they need. Thus, Simtable be vigilant and prepared to respond must develop a UI that meets the needs of its appropriately. Executing an emergency users, including emergency personnel response that minimizes danger to civilians involved in wildfire response. and first responders depends upon effective communication and coordination. Optimal Background communication during emergency During an active wildfire response, management enables involved personnel to “coordination and communication support be fully aware of the situation. are of the utmost importance” (Scholz, 2012, Consequently, crews can work quickly and p. 113). Communication between distribute resources appropriately (Chen, firefighters is necessary for relaying relevant Sharman, Rao, Upadhyaya, 2008). However, information such as the locations of current wildfire management resources, personnel, and potentially communications systems do not always hazardous areas. Currently, most firefighters meet the needs of those involved in wildfire in the field employ two-way radios to response, contributing to communicate with one another (Scholz, miscommunications and delays in 2012). However, radio communication responding to emergencies (Scholz, 2012). currently presents a number of obstacles for Simtable LLC, a company based in firefighters, such as difficulty hearing and Santa Fe, NM, develops technologies that lack of visual information (US Fire advance three-dimensional modeling, data Administration, 1999). visualization and human-computer Firefighters often enter situations interaction. Simtable is currently developing with little to no information (Litzenberg, web-based software called LiveTexture that personal communication, 2017). This is due will allow firefighters to view and annotate to a lack of ability to gather information three dimensional maps of a wildfire in real quickly about the incident (Hand, ii Wibbenmeyer, Calkin, & Thompson, 2015). will be able to add and share layers of The longer it takes for firefighters to prepare information onto the point cloud, as well as a response, the more time the fire has to view the point cloud of the area from any expand and become unmanageable. angle in real time. The capabilities of Firefighters should thus harness information LiveTexture will one day allow firefighters as quickly as possible so that a response plan to communicate much faster and more can be formulated during the early stages of effectively. However, LiveTexture does not a wildfire, keeping the initial spread to a currently have a UI to allow firefighters to minimum. do so. Social media holds one of the largest existing reservoirs of data (Yin, 2012) and can deliver data nearly instantly (Sachdeva, 2012). The real-time nature of social media as well as the geographic data that is becoming inherent in social media might provide invaluable opportunities for firefighting professionals. During wildfires, firefighters create and employ maps to identify potentially hazardous areas and possible routes to take when sending personnel to the scene of a fire (Simon, 2015). Thus, firefighters could use social media, combined with attached geographic Figure A: 3D Point Cloud of a City information, to help them determine locations of fires and construct more Project Goal and Objectives accurate maps of the surrounding area. The goal of our project was to create Firefighters could also use imagery from and prototype a user interface for social media to more easily visualize a LiveTexture intended for the real time wildfire in near real time and better aggregation and presentation of wildfire understand the surrounding environment, information. To accomplish our goal, we enabling them to make more appropriate completed the following objectives: decisions (Crowley, 2011). 1. Understand the limitations of When visualizing a situation, three information flow within the current dimensional models can give emergency wildfire management system. personnel who are not on scene a much 2. Understand the domain, context, and more comprehensive and detailed view of a constraints of what LiveTexture situation. Currently, three dimensional should accomplish. “point cloud” models (see Figure A) of an 3. Develop mock up user interfaces. area can be constructed by stitching together 4. Assess the effectiveness of the user multiple two dimensional images of the interface to determine where same area from different angles. Simtable improvements can be made. LLC is developing LiveTexture as a way to take in imagery of an area both directly and Methods from social media, then use the imagery to We employed a user-centered design construct and update three dimensional point process in order to create and prototype clouds in real time. Users of LiveTexture interfaces for LiveTexture. User-centered design processes focus first and foremost on iii understanding the needs and goals of each type of user who will interact with the interface. This design method allows developers to make sure they implement all of the necessary functionalities the users need in order to be successful. The scope of our project was limited to three specific user groups: members of the general public, firefighters on the ground reporting information about a wildfire, and virtual operation support team (VOST) members who collect and analyze Figure B: The Iterative Design Process information from social media during a wildfire. To empathize with the needs of our Results users and learn the functionalities they Current Wildfire Management Systems Lack would require from LiveTexture, we Real-Time Visual Communication conducted semi-structured interviews with From our interviews with professional representatives from each user group. We firefighting personnel, we learned that used insights from the interviews to develop firefighters currently operate on a twenty- user personas, descriptions of the target four hour information cycle, meaning that audience that will utilize the interface. The the fire managers are operating off of data personas were used to identify the features that is at least a day old at all times. our interface would need to improve the Furthermore, firefighters cannot quality of life for our users. Our user communicate information visually during an interface design was designed to address the emergency response; they only have access needs and goals of our user-personas. To to two way radios when in the field. This help us design a user-friendly interface, we prevents firefighters from gaining a full also investigated existing successful UIs, situational awareness prior to arriving on such as Periscope and Google Drive, to learn scene and even while on scene due to the what types of features and designs are large, unpredictable nature of wildfires. intuitive and easy to navigate. Finally, we evaluated the effectiveness of our UI by User Interfaces Design Should Be Centered conducting evaluations during the interface On the User design process. The purpose of the Through researching interface design evaluations was to gain feedback on how to and development frameworks, we learned design a more usable interface and what that understanding the users