Some content of this application is unavailable at the moment.
If this situation persist, please contact us atFeedback&Contact
1. (WO2019025016) METHOD AND SYSTEM FOR DETERMINING THE INTENTION OF A USER OF A VEHICLE TO BRAKE OR ACCELERATE
Note: Text based on automatic Optical Character Recognition processes. Please use the PDF version for legal matters

CLAIMS

1. A method for determining the intention of a user of a vehicle to brake or accelerate, comprising:

acquiring (100) a plurality of EEG signals on the user,

applying (101) a predetermined spatial filter on the plurality of EEG signals so as to obtain a target EEG component,

detecting (102) a spectral pattern in the EEG component corresponding to an intention to brake or detecting a spectral pattern in the EEG component corresponding to an intention to accelerate.

2. The method according to claim 1, comprising determining the instantaneous power of the EEG component according to the frequency and time, and wherein detecting said spectral patterns comprises detecting a spectral pattern in the instantaneous power of the EEG component within a predetermined spectral saliency mask corresponding to an intention to brake or detecting a spectral pattern in the instantaneous power of the EEG component within the predetermined spectral saliency mask corresponding to an intention to accelerate.

3. The method according to claim 1 or 2, wherein the predetermined spatial filter is centered in the vicinity of the Fz and Cz locations in the 10-20 EEG system.

4. The method according to any one of claims 1 to 3, wherein the EEG signals are acquired at the Cz, P3, 02, Fz, F4, and C5 locations in the 10-20 EEG system.

5. The method of any one of claims 1 to 4, wherein detecting said spectral patterns is performed in a frequency band comprising at least the a and θ frequency bands.

6. The method of any one of claims 1 to 5, further comprising an additional detection (103) of a phase pattern in the EEG component or in the EEG signals corresponding to an intention to brake or an additional detection of a phase pattern in the EEG component or in the EEG signals corresponding to an intention to accelerate.

7. The method of claim 6, wherein detecting said spectral pattern is performed at a first instant in the EEG component and detecting said phase patterns is performed at a second instant in the EEG signals or in the EEG component, the first instant and the second instant being separated by a first predetermined duration.

8. The method of claim 7, wherein the predetermined first duration is comprised between 0.8 and 1.2 seconds.

9. The method according to any one of claims 1 to 8, comprising storing a time window of the EEG component having at least a predetermined second duration.

10. The method according to claim 9, comprising:

detecting an intention to move the right foot of the user using the EEG signals at a third instant,

if an intention to move the right foot of the user is detected, performing said detection of the spectral patterns at the first instant in the stored time window of the EEG component, the first instant and the third instant being separated by said predetermined second duration.

11. The method according to claim 9 or 10, wherein the predetermined second duration is comprised between 0.2 and 1.2 seconds

12. The method according to claim 9 taken in its combination with any one of claims 6 to 8, wherein detecting said phase patterns is performed at the second instant in the stored time window of the EEG component.

13. The method according any one of claims 6 to 8 or 12, wherein said phase patterns comprise a phase realignment.

14. The method according to claims 13 and 10, wherein detecting said phase patterns comprises:

generating a plurality of sinusoidal signals having predetermined frequencies and a phase set at said third instant,

calculating a cross correlation between the signals of the plurality of sinusoidal signals and stored time window of the EEG component at said second instant in the stored time window of the EEG component.

15. The method according to claim 13, wherein detecting a phase realignment comprises:

extracting the Hilbert transform from the EEG signals or the EEG component so as to obtain instantaneous phase shifts,

extracting the first derivative from the instantaneous phase shifts, so as to obtain the phase realignment.

16. The method according to claim 13, wherein detecting a phase realignment comprises:

extracting the wavelet decomposition of the EEG signals or of the EEG component,

assessing, within a predetermined phase saliency map, whether the instantaneous phase approaches predetermined or learned phase values.

17. The method of claim 1 or 6, wherein detecting a spectral pattern is performed using a machine-learning classifier and/or wherein detecting a phase pattern is performed using a machine-learning classifier.

18. A system for determining the intention of a user of a vehicle to brake or accelerate, comprising:

an acquisition module (202) of a plurality of EEG signals on a user in the vehicle,

a processor (201),

a memory (203) comprising instructions (INST) executable by the processor to:

acquire (204) a plurality of EEG signals on the user,

apply (205) a predetermined spatial filter on the plurality of EEG signals so as to obtain a target EEG component,

detect (206) a spectral pattern in the EEG component corresponding to an intention to brake or detecting a spectral pattern in the EEG component corresponding to an intention to accelerate.

19. A computer program including instructions for executing the steps of a method according to any one of claims 1 to 17 when said program is executed by a computer.

20. A recording medium readable by a computer and having recorded thereon a computer program including instructions for executing the steps of a method according to any one of claims 1 to 17.