Jeffrey Hightower
Researcher, Intel Research Seattle

My research is in location-enhanced mobile computing. I'm focusing on devices, services, sensors, and user interfaces that enable computing to fade quietly into the background of daily life. This is the concept of "calm computing." The goal is to simplify the experience of mobile computing.

The Platform Location Capability research project that I lead explores adding low-power, low-cost inertial sensors into a mobile platform. Current location technology only provides a location fix periodically-say, every second or every few seconds. The addition of inertial sensors such as accelerometers, gyros, barometers, and magnetometers (compasses) could help to fill those gaps and improve overall estimates between location fixes. An algorithm embedded in the platform could analyze the input from multiple sensors and estimate on the fly which available sensors are providing the most accurate location information at any given time-what we call the "Always Best Location" estimate.

We're particularly interested in the capability to detect altitude and heading, or direction, by incorporating a barometer and magnetometer in the mobile platform. The barometer would sense changes in pressure, indicating the altitude or floor of the building where the device is located, and the magnetometer would indicate the direction in which the device is moving. That combined information could be useful in emergency situations; for instance, it could help firefighters to quickly identify the location and movement of people trapped in a burning building.

We also want to enable mobile devices to do self-mapping indoors. Just as robots learn their environment through Simultaneous Localization and Mapping, or SLAM, a mobile device could "learn" the significant places that the user frequents by recognizing the unique signatures of Wi-Fi radios in the area. Ultimately, you could use this capability to assign labels (My Home, My Office, My Favorite Coffee Shop and so on) to the patterns that the device has learned. For example, you could build an instant messenger client that, when you put it into dynamic mode, could automatically fill in the name of the place where you're located and share that name with your friends. Another example of how you might use self-mapping is in configuring the local computing infrastructure. For instance, if your mobile device detected that you're at home, it could set up the home network, choose the right default printer, change the browser's home page, and make other configuration changes.

In addition to pursuing our own research goals, we're contributing to the activity recognition research project going on in our lab. One goal of that project is to enable people to age in place in their homes. We have built a multi-sensor platform that can sense and infer people's activities. The addition of inertial sensors to the platform will make it even more robust, enabling the activity recognition system to track the movement of elderly people in their homes and proactively offer assistance if needed.

Our research represents a significant step in the direction of building mobile platforms that have location capability as a basic feature, just as today's notebook computers come equipped with wireless capability. Once location capability becomes ubiquitous in mobile devices, this could inspire as host of new location-based applications. For instance, you could envision an application that enables you to navigate in whatever way you choose-the fastest route, the most scenic, the most familiar, and so on. Another application might enable you to automatically reconfigure your device as it moves from one favorite place to the next. Through such applications, location capability promises to simplify and enrich the experience of mobile computing.

Jeffrey Hightower is a member of the research staff of Intel Research Seattle. His research focuses on location-enhanced mobile computing. More generally, his goal is to design "calm technology" that fades into the background of daily life.

Hightower has co-authored numerous articles, including seminal papers on location technology (IEEE Computer Aug. 2001) and Wi-Fi device positioning (Pervasive Computing 2005). He has served as a peer reviewer in many venues, including IEEE Pervasive Computing magazine, the International Conference on Ubiquitous Computing (Ubicomp), Conference on Human Factors in Computing (CHI), and Pervasive Computing. He earned a Ph.D. in Computer Science & Engineering from the University of Washington.