C-bro

Well I just finished my latest programming project. It's darn cool too. The basic premise is a cellphone tracking system. The program reads a series of angle measurements from a noisy sensor, filters/optimizes the measurements, and then computes the starting location, speed, and direction of travel for the cellphone user. The key point to the program is that only an angle measurement is needed. The need for a distance measurement is eliminated by moving the sensor in a prescribed motion.

I ended up using a multivariable Newton optimization to filter the noise, and the final results have shown accuracy of about +/-18cm at 1km from the sensor. The values are all simulated, so I'm sure the variances would be much different in the real-world, but I thought it was alot of fun anyway.

Anyone else ever do similar projects?

__________________

Automatically backup files - Why computers have clocks