DARPA Develops Non-GPS Navigation Chip 84
Zothecula writes "The Global Positioning System (GPS) has proved a boon for those with a bad sense of direction, but the satellite-based system isn't without its shortcomings. Something as simple as going indoors or entering a tunnel can render the system useless. That might be inconvenient for civilians, but it's potentially disastrous to military users, for whom the system was originally built. DARPA is addressing such concerns with the development of a self-sufficient navigation system that can aid navigation when GPS is temporarily unavailable."
Similar to some existing systems (Score:5, Informative)
For this limited scenario, it appears to the user as if the GPS was active all along.
Article misses the point (Score:5, Informative)
What the article is describing (an IMU) have been around forever (since before GPS), and pretty much any system that uses GPS for navigation has one to supplement the GPS. What is new here is the size; a full IMU on a single chip the size of your pinky finger nail. Pretty cool considering that not too long ago these used to comprise of multiple separate physical devices (gyrometer x3, accellerometer x3, magnetometer), but have been getting progressively smaller over the years. MEMs has come a long way.
it IS existing systems in smaller package (Score:5, Informative)
nothing new about gyroscopes and accelerometers in a package, even an integrated circuit one.....this might be smaller or perhaps more accurate than some I've seen over the DECADES. but definitely no new tech or ideas here.
Horrible summary (Score:4, Informative)
Inertial Nav + Kalman Filter (Score:5, Informative)
The break throughs are not only in getting the platform sensors, the gyros, accelerometers, and magnetometers, onto a single chip, but also in being able to provide the computer horsepower to do the Kalman filtering to integrate all these sensors to come out with a nav/position solution, in a few cubic centimeters of processed sand, and for a few Watts.
It's not just the sensors, it's the processing as well. The sensors just throw data at you (data with all sorts of errors); the Kalman filter lets you bring everything together for your nav/position solution. As a prof long ago said it, "Kalman filtering -- how to stop worrying and learn to love matrix inversion."
Re:Similar to some existing systems (Score:5, Informative)
Dead reckoning, thousands of years old, now with computer. Patent!