Pedestrian Tracking andNavigation Using Neural Networks andFuzzyLogic
暂无分享,去创建一个
Themaingoaloftheresearch presented hereisto location (indoor oroutdoor) isdetermined basedonthe develop theoretical foundations andimplementation algorithms, received signal strength, angle ofarrival, timeofarrival orthe whichintegrate GPS, micro-electro-mechanical inertialdifference intimeofarrival. Inaddition toGNSS/RF measurement unit(MEMSIMU),digital barometer, electronic navigation, multi-sensor systemstypically incorporate compass, andhumanpedometry toprovide navigation and additional sensorsor sensornetworks suchas inertial tracking ofmilitary andrescue groundpersonnel. Thispaper measurement units (IMUs), magnetometers, laser scanners, discusses thedesign, implementation andtheinitial performance ' ' o analyses ofthepersonal navigator prototype1, withaspecialinfrared sensors, optical imagers, etc. toprovide absolute or emphasis ondead-reckoning (DR)navigation supported bythe relative position information. Recenttechnological humanlocomotion model. Tofacilitate this functionality, the developments inpositioning andtracking sensors, including adaptive knowledge system, basedontheArtificial Neural theGlobal Positioning System (GPS)modernization program Networks (ANN)andFuzzyLogic, istrained during theGPS andadvances inhigh-sensitivity receiver technology, capable signal reception andusedtomaintain navigation underGPS- of supporting navigation indoorsand in confined denied conditions. Thehumanlocomotion parameters, step environments (14), well-established MEMS accelerometer frequency (SF)andsteplength (SL)areestimated during the technology, andsteadily improving MEMS gyrotechnology, systemcalibration period, thenthepredicted SL,together with miniaturized magnetometers anddigital barometer/altimeter theheading information fromthecompass andgyro, support DR technology, aswellastheavailability ofother RF signals navigation. Thecurrent target accuracy ofthesystem is3-5m chnolof suwell ation of oter to develo CEP(circular error probable) 50%. capable ofsupporting navigation offer apotential todevelop multi-sensor, portable systems forpersonal tracking and
[1] Dorota A. Grejner-Brzezinska,et al. Gravity Modeling for High‐Accuracy GPS/INS Integration , 1998 .