NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Brad Morris
Date: 2013 Mar 28, 15:15 -0400
Hi Don
If all it took was a 0.1 hertz low pass filter (which could easily be realized with digital software) then cell phone inertial navigation would be resolved. I think there's much more to it than that.
A 0.1 hertz low pass filter also means a very low frequency response. Hold that device very steady for a long time.
Functional devices of the "automated celestial" variety that I have read about are vessel mounted with a gimbaled mirror for pointing. Excellent accelerometers mounted to a large inertial mass (the vessel). High end servo motion control of the mirror. Not some $10 chip you hold in your hand.
If it were up to me, I would use the Celestron Sky Scout as a starting point, replacing the optics so as to restrict the FOV, providing higher magnification and cross hairs. Additionally, much higher resolution (and corresponding rise in accuracy) tilt sensors would be needed. Your DR would be injected as opposed to the known GPS location. The stars would be restricted to the known 57 or similar. Most nav stars are separated from each other to preclude confusion. It would be up to the operator to follow the Sky Scout 's general guidance to find the star and then switch mode to a recording the altitude feature. After snapping several altitudes, LOPs would be computed from the measured tilt(s) and 'known' star(s) This seems practical and workable, and in near time frame. The limitation would be, as always, price for accuracy.
Brad
Brad wrote:
I fear that the high frequency accelerations associated with a hand held device such as proposed will preclude an accurate vertical. One only has to download one of the free apps *(such as Accelogger on the Android system) to observe the noise present. This noise is the same issue that precludes inertial navigation using your cell phone, many have tried, all have failed. Simply search for "inertial navigation" for your cell phone. I know I want that application.
Sure, the phone understands which way is generally down, so as to orient the screen, but that is far from understanding fine resolution vertical.
So holding the device will most likely not work with chips present in modern smart or cell phones.
___________This brings up bandwidth considerations. The accelerometers in cell phones and game controllers are intended for a broad range of applications, some of which require a quick response to hand motions. This implies a bandwidth of probably more than 10 Hz. For a star sighter application, the bandwidth needs are much, much less. It would be highly advantageous to low pass filter the accelerometer outputs with a cutoff of 1 Hz or less. Perhaps as low as .1 Hz might work well in reducing the errors of external vibration. Additionally, it would significantly reduce an inherent noise source in all MEMS accelerometers that is proportional to the square root of the bandwidth. As an example, the ADXL203 has a noise limitation of about 1 mg when operated with a 60Hz bandwidth, an order of magnitude too great for celestial navigation. But add on a .1Hz low pass filter and this error source drops down to an acceptable level.
My suspicion is that a hand held device will still be unpractical, but I do hold out hope for a well-damped tripod.
Don Seltzer
----------------------------------------------------------------
NavList message boards and member settings: www.fer3.com/NavList
Members may optionally receive posts by email.
To cancel email delivery, send a message to NoMail[at]fer3.com
----------------------------------------------------------------