2016 · 10 citations · 18 references
ComputingComputer ScienceRobotics EngineeringMobile RoboticsRoboticsSensor FusionKalman FilteringMotion AnalysisMotion EstimationPose EstimationRobot LocalizationTracking SystemsAccuracyExtended Kalman FilterInertial Measurement Unit FusionMotion TrackingMotion Tracking SystemsReal-time EstimationUbiquitous Computing
In order to track the localizations of mobile devices in an unknown environment, this paper presents an architecture combining a monocular camera and an inertial measurement unit (IMU) in ubiquitous mobile devices. The IMU module provides acceleration and angular velocity with high-frequency, but the IMU-based motion tracking is more inclined to collapse due to the drift integration. While the vision-based motion tracking can provide higher accuracy, but it cannot work in the environment with weak texture or dynamic scenes. Based on the fusion of the IMU and monocular camera, this paper proposed a loosely couple method in the error-state Extended Kalman Filter framework. With the combination of the advantages the monocular camera and IMU, the proposed method can achieve real-time ego-motion estimation in resource-constrained mobile devices. Finally, the validity of the proposed motion tracking method is evaluated by experiments.
18
ORB: An efficient alternative to SIFT or SURF
Ethan Rublee, Vincent Rabaud, Kurt Konolige et al. · 2011 · 10.4K citations
Introduction to autonomous mobile robots
Choice Reviews Online · 2011 · 2.2K citations
Davide Scaramuzza, Friedrich Fraundorfer · IEEE Robotics & Automation Magazine · 2011 · 1.5K citations