A Wearable Pedestrian Localization and Gait Identification System Using Kalman Filtered Inertial Data
Nasim Hajati, Amin Rezaeizadeh
Abstract
In this article, we introduce a pedestrian dead reckoning (PDR)-based navigation device that does not require global navigation satellite system (GNSS) signals or beacons and works with an inertial measurement unit (IMU) mounted on its waist belt. The system identifies the individual by their walking pattern to use the proper gains in the computations, estimates the attitude by applying an unscented Kalman filter, and finally derives the position in three dimensions with the help of a step detection algorithm. The experimental results show an outdoor 4.7-km walk resulting in an error of 0.96%.
Topics & Concepts
Dead reckoningKalman filterInertial measurement unitComputer scienceBeaconGNSS applicationsComputer visionExtended Kalman filterInertial navigation systemPedestrianAccelerometerSimultaneous localization and mappingWearable computerUnits of measurementSatellite systemArtificial intelligenceGlobal Positioning SystemInertial frame of referenceEngineeringReal-time computingMobile robotTelecommunicationsTransport engineeringEmbedded systemPhysicsQuantum mechanicsOperating systemRobotIndoor and Outdoor Localization TechnologiesGait Recognition and AnalysisTarget Tracking and Data Fusion in Sensor Networks