Mobile device control leveraging user kinematics
View Patent ↗Some embodiments of the invention provide methods and apparatus for controlling an aspect of the presentation of objects in a mobile or wearable device, where the user is performing a gait activity such as walking, jogging or running, and the controlling is performed leveraging the gait characteristics of the user. In some embodiments, the gait characteristics include velocity and stride length. In some embodiments, the only sensors utilized to obtain any contextual information are accelerometers.
1. A method for monitoring a user of a mobile device in real time, the method comprising:
reading data from an accelerometer within the mobile device;
a dynamic selection of a sampling frequency of the accelerometer, and a dynamic selection of wavelet transformation parameters, comprising levels of decomposition and a mother wavelet, following at least two criteria: an accuracy optimization and a minimization of computational costs;
obtaining an energy of wavelet transformation coefficients of the accelerometer data, and leverage said energy to estimate a velocity of the user;
presenting on the mobile device a moving object whose movement is controlled by the estimated velocity;
leveraging energies of wavelet transformation detail coefficients to choose detail coefficients from which to obtain a reconstructed wave from where each step of the user is identified;
combining a step time information with the velocity estimation to estimate a step length.
2. The method of claim 1 , wherein the energy of wavelet transformation coefficients of the accelerometer data leveraged to estimate the velocity of the user, is the energy of wavelet transformation approximation coefficients.
3. The method of claim 2 , wherein wavelet transformation parameters in the dynamic selection of wavelet transformation parameters, comprise: sampling frequencies ranging from 5 Hz to 200 Hz, and mother wavelets including several of the Biorthogonal, Daubechies, Coiflets, Haar, Meyer, Mexican Hat, Morlet, Symlet and other families.
4. An apparatus comprising:
a processor;
a non-transitory processor-readable medium including one or more instructions which, when executed by the processor, causes the processor to monitor a user of a mobile device in real time by:
reading data from an accelerometer within the mobile device;
a dynamic selection of a sampling frequency of the accelerometer, and a dynamic selection of wavelet transformation parameters, comprising levels of decomposition and a mother wavelet, following at least two criteria: an accuracy optimization and a minimization of computational costs;
obtaining an energy of wavelet transformation coefficients of the accelerometer data, and leverage said energy to estimate a velocity of the user;
presenting on the mobile device a moving object whose movement is controlled by the estimated velocity;
leveraging energies of wavelet transformation detail coefficients to choose detail coefficients from which to obtain a reconstructed wave from where each step of the user is identified;
combining a step time information with the velocity estimation to estimate a step length.
5. The apparatus of claim 4 , wherein the energy of wavelet transformation coefficients of the accelerometer data leveraged to estimate the velocity of the user, is the energy of wavelet transformation approximation coefficients.
6. The apparatus of claim 5 , wherein wavelet transformation parameters in the dynamic selection of wavelet transformation parameters, comprise: sampling frequencies ranging from 5 Hz to 200 Hz, and mother wavelets including several of the Biorthogonal, Daubechies, Coiflets, Haar, Meyer, Mexican Hat, Morlet, Symlet and other families.
7. A non-transitory processor-readable medium including instructions which, when executed by a processor, causes the processor to monitor a user of a mobile device in real time by:
reading data from an accelerometer within the mobile device;
a dynamic selection of a sampling frequency of the accelerometer, and a dynamic selection of wavelet transformation parameters, comprising levels of decomposition and a mother wavelet, following at least two criteria: an accuracy optimization and a minimization of computational costs;
obtaining an energy of wavelet transformation coefficients of the accelerometer data, and leverage said energy to estimate a velocity of the user;
presenting on the mobile device a moving object whose movement is controlled by the estimated velocity;
leveraging energies of wavelet transformation detail coefficients to choose detail coefficients from which to obtain a reconstructed wave from where each step of the user is identified;
combining a step time information with the velocity estimation to estimate a step length.
8. The non-transitory processor-readable medium of claim 7 , wherein the energy of wavelet transformation coefficients of the accelerometer data leveraged to estimate the velocity of the user, is the energy of wavelet transformation approximation coefficients.
9. The non-transitory processor-readable medium of claim 8 , wherein wavelet transformation parameters in the dynamic selection of wavelet transformation parameters, comprise: sampling frequencies ranging from 5 Hz to 200 Hz, and mother wavelets including several of the Biorthogonal, Daubechies, Coiflets, Haar, Meyer, Mexican Hat, Morlet, Symlet and other families.