Indoor Localization
hello, im working on project require accuracy down two-three inches, , speed of communicating location crucial.
this topic has been discussed on these forums share game plan.
using 4 xbee radio/or wifi beacons localization , 1 on person.
unlike plans, need weakest radios possible, there variable transmission long range, , lag speed.
on external server (my computer), map out beacons, , tag them unique identifiers.
lay out digital coordinates within xbee's area of responsibility, based on identifier.
when individual walks in, picks 4 of beacons, uses rssi, , calculates exact grid coordinates. know there variations of error in xbees, way supposed work (in head) takes distances, , smooth's them out through code. if 1 of beacons waivers more 5% expected value (based on other 3 beacons), assumes momentarily beacon off, , renders output temporarily unavailable (and outputs expected result) until gets within accuracy.
when individual enters new grid, localization swaps on new grid, , process starts on again.
the variables not stored @ each beacon, server, , server performs calculations.
for project working on, need output coordinates @ least 24 times per second.
my question is, can done, or improved. cost no issue. in advance.
this topic has been discussed on these forums share game plan.
using 4 xbee radio/or wifi beacons localization , 1 on person.
unlike plans, need weakest radios possible, there variable transmission long range, , lag speed.
on external server (my computer), map out beacons, , tag them unique identifiers.
lay out digital coordinates within xbee's area of responsibility, based on identifier.
when individual walks in, picks 4 of beacons, uses rssi, , calculates exact grid coordinates. know there variations of error in xbees, way supposed work (in head) takes distances, , smooth's them out through code. if 1 of beacons waivers more 5% expected value (based on other 3 beacons), assumes momentarily beacon off, , renders output temporarily unavailable (and outputs expected result) until gets within accuracy.
when individual enters new grid, localization swaps on new grid, , process starts on again.
the variables not stored @ each beacon, server, , server performs calculations.
for project working on, need output coordinates @ least 24 times per second.
my question is, can done, or improved. cost no issue. in advance.
quote
when individual walks in, picks 4 of beacons, uses rssi, , calculates exact grid coordinates.this not possible. rssi extremely poor indicator of distance, depends more on relative antenna orientation , presence of nearby reflecting or absorbing objects distance.
Arduino Forum > Using Arduino > Project Guidance > Indoor Localization
arduino
Comments
Post a Comment