An Efficient Real-time Human Posture Tracking Algorithm using Low-cost Inertial and Magnetic Sensors. Gallagher, A., Matsuoka, Y., & Ang, W. T. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, volume 3, pages 2967 - 2972, 2004.
doi  abstract   bibtex   
Real-time accurate human posture tracking in unconstrained environments provides an enabling technology for physicians and other care providers to monitor the movements of their patients in real-life situations. Constructing a posture tracking system with the form factor suitable for human wear requires the development of miniature units that can be attached to the limb segments of interest in an unobtrusive way. Simultaneously, fast algorithms that can produce real-time posture estimates at sufficient rates are needed. In this paper, we focus on the development of efficient and accurate algorithms that compute the human posture information from low-cost miniature inertial and magnetic sensors. We present a new technique that computes posture estimates from the sensor data 23.8 times faster than the most efficient previously proposed technique, and simultaneously increases the accuracy of the estimates.
@InProceedings{Gallagher2004,
  Title                    = {An Efficient Real-time Human Posture Tracking Algorithm using Low-cost Inertial and Magnetic Sensors},
  Author                   = {Gallagher, A. and Matsuoka, Y. and Ang, W. T.},
  Booktitle                = {Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems},
  Year                     = {2004},
  Pages                    = {2967 - 2972},
  Volume                   = {3},

  Abstract                 = {Real-time accurate human posture tracking in unconstrained environments provides an enabling technology for physicians and other care providers to monitor the movements of their patients in real-life situations. Constructing a posture tracking system with the form factor suitable for human wear requires the development of miniature units that can be attached to the limb segments of interest in an unobtrusive way. Simultaneously, fast algorithms that can produce real-time posture estimates at sufficient rates are needed. In this paper, we focus on the development of efficient and accurate algorithms that compute the human posture information from low-cost miniature inertial and magnetic sensors. We present a new technique that computes posture estimates from the sensor data 23.8 times faster than the most efficient previously proposed technique, and simultaneously increases the accuracy of the estimates.},
  Doi                      = {10.1109/IROS.2004.1389860},
  Keywords                 = {human posture; human wear; inertial sensor; limb segments; magnetic sensor; real-time human posture tracking algorithm; computerised monitoring; magnetic sensors; patient monitoring; postural detection},
  Review                   = {Uses accel, gyro, mag sensors.},
  Timestamp                = {2011.03.01}
}
Downloads: 0