PORTLAND, Ore.—Electronic compass applications combine magnetometer-provided headings with corrections from inertial sensors that compensate for stray magnetic fields. As a result, sensor fusion algorithms provide fast, reliable electronic compass—or e-compass—readings for context-awareness applications like mobile location-based services, 3-D games and e-health. For users of Freescale Semiconductor's magnetometers and accelerometers, e-compass sensor fusion algorithms are now available as a free download.
"Our sensor fusion algorithms use the strengths of one sensor to overcome the weakness of the other, resulting in fast, accurate and reliable readings," said Michelle Kelsey, Freescale's product-line manager for sensors in mobile markets. "Freescale's Xtrinsic e-compass software enables applications in augmented reality, e-health, gaming and navigation."
Freescale is offering its sensor fusion algorithms as a free download for its MEMS sensor users to enable them to tap the skyrocketing market for mobile motion sensing technologies, the global revenue for which in smartphones and tablets alone will double from $1 billion today to nearly $2 billion by 2015, according IHS iSuppli (El Segundo, Calif.).
To get accurate readings from MEMS magnetometers and accelerometers, sensor-fusion algorithms need to compensate for stray magnetic fields both inside the mobile device itself as well as outside in the environment. By supplying the algorithms, along with application notes and an embedded simulation framework, Freescale is aiming to simplify sensor fusion for OEMS using its MEMS sensors.
Included in the algorithms is tilt-compensation—for when a user is not holding the mobile device parallel to the ground—as well as realtime calibration that compensates for both hard-iron and soft-iron sources of errors. The processor agnostic C-code algorithms are supplied with annotated documentation. The run-time code measures just 20 kilobytes and requires only 6.5 kbytes of RAM.
Sensor fusion algorithms subtract the magnetic interference from soft- and hard-iron in the local geomagnetic fields.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.