Following Footsteps

In the magical world of Harry Potter, there exists a map that tracks the movements of each person (and ghost) at the Hogwarts School of Witchcraft and Wizardry. The series’ protagonist, Harry, and his friends, can watch their teachers’ tiny footprints move through the school’s corridors as they wait for the coast to clear.

Sean Connell and Samuel House

Connell and House developed a device that could track people's movements indoors.

The idea of a “Marauder’s Map,” as it’s named in the books, might seem whimsical, but three Oregon State University seniors have developed something similar — a device that tracks individuals in an indoor space. But the real-life version isn’t used for sleuthing and avoiding teachers. Rather, it’s used to aid research on healthy aging and independent living for the elderly.

The creators — electrical engineering and computer science students Samuel House, Sean Connell and Ian Milligan — did not use enchanted paper. They used a computer display that traces a person’s path as a line moving across a map, and a circuit board that attaches to the top of a person’s shoe. The result is similar to Harry Potter’s map — it helps track people’s movements indoors.

The creation helped solve a problem for Tamara Hayes, Jeffrey Kaye, and Misha Pavel, researchers at the Oregon Center for Aging and Technology at Oregon Health & Science University (OHSU) who study motor and cognitive decline and how to prevent it or remediate it.

Their goal was to study elderly people in their homes and to measure indicators like walking speed, total activity, and patterns of socialization (for example how often they spent talking to others versus being alone in their room).

Initially, the three outfitted study participants’ homes with sensors on the walls and ceilings to detect movement. But individuals could not be uniquely identified, which limited their ability to draw conclusions from the data.

Using GPS wasn’t a possibility, since such devices do not work indoors. And other devices designed for indoor tracking are complicated, expensive (some upwards of $3,000) and could only work for a couple hours. Their challenge was to make something affordable that could be implemented in a large research study, and that had a long battery life so activity level could be measured for days rather than hours.

So they turned to colleagues at Oregon State University for help.

A not-so-crazy idea

Enter Patrick Chiang, assistant professor of OSU’s School of Electrical Engineering and Computer Science. Chiang had some ideas of how such a device could be built, and sought some undergraduates to help with the project.

The biomedical application piqued the interest of friends House and Connell, who share an interest in human improvement through medical electronics.

“Sean and I have worked together on projects before and what usually happens is I have a lot of crazy ideas and he shoots them down,” House says with a smile. “This happened to be one that worked.”

The beauty of this particular idea, though, is its practicality, which involved combining some common technologies in a way that no one has ever done before.

The two, who received initial funding for the project through the LIFE Scholars Summer Research Program, started with a system that used dead reckoning — the process of estimating location based on one’s starting position and movement data such as time traveled, speed and direction. Although the sensors are accurate enough for short-term tracking, small errors in the measurements would accumulate over time, making long-term tracking inaccurate.

House’s inspiration was to add radio frequency identification (RFID) tags — used in credit cards and at stores like Wal-Mart to track inventory —  placed around the indoor space, and an RFID reader on the device so when the user passes by a tag their exact location can be determined.

“That way you keep your error from growing to insane bounds,” Connell says. “Without updates the errors will send you to the moon.”

Some complicated math was required to write the software that reconciles all the data from the sensors on the device — a three-axis accelerometer, a three-axis gyroscope, a three-axis magnetometer and an RFID reader —  to determine the user’s path. And for that task they requested help from the final member of their senior project, Milligan, a student of electrical and computer engineering who also studied computer science and mathematics.

“It takes that much breadth of knowledge to be able to implement something like that,” says Connell.

House, Connell and Milligan's device, with labels

House, Connell and Milligan's device, with labels

Milligan developed a filter to correct some of the error in the data by comparing what was expected — based on the laws of physics and body motion — to what was actually measured, to produce a path that is closer to reality.

Their final product is a 5×9 cm lightweight device that has a price tag of under $100 which makes it ideal for a variety of applications including any research that aims to look at activity level, for example studies examining the effects of a particular drug or supplement. It could also be useful in elderly care facilities for detecting falls, or keeping track of patients who might wander off.

The Future of Tracking

They designed the device to be modular, so it can be equipped with add-ons like Bluetooth in order to send data to a smartphone, or a GPS unit that turns on when a user steps out of a defined boundary and sends an alert with the location information.

Hayes, of OHSU, says the device has “strong application in dementia care” and they have begun discussing possible applications for it with a community care organization that is searching for a solution to help with the problem of patients who wander off.

Although it is a fully functional unit on its own it is also part of a larger objective by their advisor, Chiang, and members of his research group at OSU to build tiny, low-power, wireless devices to collect a variety of biometric information. Chiang envisions a non-invasive Band-Aid sized device that would include measurements such as heart rate and respiration to provide continuous monitoring of vital signs for researchers studying the elderly or other populations at health risk. He says the devices could help provide clues to find the most effective interventions such as pharmaceuticals, vitamins, and exercise.

Chiang, who regularly mentors undergraduates, says the students’ contribution to this larger goal has been impressive, “I don’t think I could have done as good a job, so I left them to their own devices. What they were able to demonstrate in a short period of time was remarkable.”

He is not alone in his assessment. For their project the three students received the OSU’s CUE (Celebrating Undergraduate Excellence) award for the College of Engineering, and they were selected by industry representatives for the Industry Award at the Engineering Expo for electrical and computer engineering senior design projects. Their paper on the project was accepted for presentation at the 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society in August. Additionally, Texas Instruments awarded them first place at OSU in the Analog Design Contest.
But House found even greater reward in the process.

“I really liked the experience of building something new that nobody’s ever done before. Sure, you run into a lot of problems that nobody can help you with, but that’s really the whole point …that’s what being an engineer is about.”

-By Rachel Robertson

Leave a Reply