Computer scientists at Rice University in Houston say they’ve come up with a new mapping solution for interior navigational location detection by linking it to existing sensors in mobile devices: the gyroscope and accelerometer.
They note that in recent years, low-cost inertial sensors based on accelerators and gyroscopes have emerged as a well-known solution for navigation in indoor or hybrid indoor-outdoor environments. However, their usefulness has been limited due to noise and errors.
What the Rice researchers did was combine the accelerometer and gyroscope measurements with cheap coarse-grained machine learning along with a map of the environment for accurate navigation. Their results were presented in a paper at last month's 2017 Design, Automation and Test in Europe (DATE) Conference in Lausanne, Switzerland.
The same researchers published a paper six months ago on an indoor mobile positioning system called CaPSuLe. The navigational location detection system began as a solution for mobile device users inside large indoor spaces like shopping malls and campuses where GPS navigation falters under poor signals that quickly deplete battery life, according to Rice News.
Both CaPSuLe and the DATE paper technology use machine learning for location detection, increasing the speed of calculations and decreasing energy expenditure compared with existing location technologies, Rice reported. But where CaPSuLe depends on image-matching techniques and uploaded data, the new technology taps into sensors that already exist in most mobile devices.
The gyroscope and accelerometer sensors also require little energy to run. “We're building a solution that uses cheap existing sensors, such as gyroscope and accelerometer," said assistant professor of computer science Anshumali Shrivastava in the release. "These sensors track acceleration and rotation, but the location signals are 'noisy' because of irrelevant movements. For example, we can use information from these sensors to track walking movements, but the sensors also pick up swinging arms and waving hands. So when we try to apply physical laws of motion to compute the final location, the result is an accumulation of errors."
The researchers assume the walking paths in the environments in which they're mapping are straight lines, with turns, not necessarily right angles, including U-turns. They say this is not an unrealistic assumption as humans usually walk along paths that can be accurately approximated with a union of reasonable small straight line segments. Using this scenario, their method allows them to eliminate costly and inaccurate numerical integration.
Juan Jose Gonzalez Espana, a graduate student in the Department of Electrical and Computer Engineering, said that by aggregating all the information, the team "demonstrated a solution that is twice as accurate as GPS services, while being around 27 times cheaper in terms of energy, which directly translates into battery life."
International collaborators with the Rice team include Moon Yongshik, Soonhyun Noh, Daedong Park and Seongsoo Hong, all of Seoul National University in South Korea. The research was supported in part by the U.S. Defense Advanced Research Projects Agency.
The researchers say they plan to compare energy usage and accuracy with other competing approaches, including Wi-Fi and Bluetooth, beyond GPS.