A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Hi-index | 0.00 |
This paper describes a new detection method of walking activity with a lower limbs rehabilitative training robot. The robot is designed to improve walking ability of patients who suffer from impairment in locomotion after neurology injuries or aged persons who lack walking capability. Kinematic sensors are applied to monitoring postural transitions (PTs), and the sensor signals are sent to host computer to control lower limbs rehabilitative training robot. If the trained person does not follow the robot, sensors detect abnormal body posture. The host computer sends stop instruction to prevent the trained person from falling down. This paper describes composition of whole wireless data monitoring and control system in details. Original data and processed data from experiment show the status of this system operation. In the study, the wavelet transform is employed to process collected data. At last, a detection scheme of abnormal walking activity is discussed for protecting the person in rehabilitative training.