SBIR Phase I: Mapping Human Signals with Multimer and MindRider

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project includes high quality data that indicates a population?s sense of well-being and environmental safety--two essential factors in the effort for long-term sustainability. MindRider and Multimer collect and analyze human signals at the neighborhood, town, city, and regional scale, thus providing proactive, standardized, geospatial, real-time information to improve both personal mobility and urban monitoring. MindRider and Multimer can help overcome limitations in two prevalent monitoring methods: 1) A reactive methodology for quantifying danger is to count accidents. While government-collected accident data is often reliable and includes many years' worth of information, it does not provide precise spatial details such as direction of movement. Most importantly, an injury or fatality must occur before a statistic is produced, which therefore misses all the non-reported minor injuries, as well as perceptions of danger. 2) Many researchers use qualitative instruments like surveys and interviews to study perceptions of safety without the occurrence of an accident But while qualitative methods are highly valuable, these methods cannot be deployed continuously, are often not geospatial, and their results may not be standardized. The proposed project addresses the challenges that stakeholders like transportation planners and real estate analysts encounter as they monitor rapid change in cities and towns. For these stakeholders, improved data helps to reduce the complexity, cost, and risk of site selection decisions like choosing the location for a store, transit hub, outdoor advertisement, home, or office. Existing location analysis tools show similar kinds of evaluations for very different neighborhoods, based on limited data that doesn't often change. MindRider, which pairs custom ergonomic biosensors with GPS-enabled smartphones to record EEG and location data, and Multimer, which categorizes and analyzes MindRider data against existing geographic inputs, has the potential to provide key insights for data-driven decision making. We have built the basic system to collect and analyze geolocated biosensor data and have proven the technical feasibility of collecting these data from a pilot sample population for an extended period of time. Our primary objective for Phase I is to improve our data modeling and analysis--with a focus on affective interpretation, biosensor comparison, geospatial and temporal statistics, and reproducibility--so as to offer a fully robust product that stakeholders will confidently integrate into their existing workflows.