RF-based micro-motion tracking for gesture tracking and recognition
This document describes techniques for radio frequency (RF) based micro-motion tracking. These techniques enable even millimeter-scale hand motions to be tracked. To do so, radar signals are used from radar systems that, with conventional techniques, would only permit resolutions of a centimeter or more.
1. A computer-implemented method for gesture recognition, the computer-implemented method comprising:
receiving first reflections of a radar field off first and second points of a single non-rigid target at a first time, the first and second points of the single non-rigid target moving at first and second Doppler-based velocities, respectively;
generating, based on the received first reflections, Doppler-based measurements representing a velocity profile of the single non-rigid target, the Doppler-based measurements including first and second measurements respectively associated with the first and second Doppler-based velocities, the first and second measurements having higher energies than another measurement corresponding to another point of the single non-rigid target that is moving at another Doppler-based velocity, and the first and second points moving relative to one another within the radar field;
determining, based on the first and second measurements, a first relative velocity between the first and second points at the first time;
receiving second reflections of the radar field off the first and second points at a second time;
determining, based on the second reflections, a second relative velocity between the first and second points at the second time; and
determining, based on the first and second relative velocities, a gesture performed by the single non-rigid target.
2. The computer-implemented method of claim 1 , further comprising determining, based on the first and second relative velocities, a change in displacement between the first and second points between the first and second times.
3. The computer-implemented method of claim 2 , wherein determining the gesture is based further on the change in displacement.
4. The computer-implemented method of claim 1 , wherein the energies correspond to respective ranges to the first and second points and the other point of the single non-rigid target.
5. The computer-implemented method of claim 1 , further comprising identifying, based on a moving target indicator filter, the first and second points from the other point of the single non-rigid target within the first and second reflections.
6. The computer-implemented method of claim 1 , further comprising passing the gesture to an application or device effective to control or alter a display, function, or capability associated with the application or the device.
7. The computer-implemented method of claim 1 , wherein the Doppler-based measurements for the first and second points at the first and second times are determined to have one or more of high accuracy or low noise compared to other determined Doppler frequencies.
8. The computer-implemented method of claim 1 , further comprising creating a range-Doppler-time data cube for the single non-rigid target, wherein the determining the gesture is based further on the range-Doppler-time data cube.
9. The computer-implemented method of claim 1 , wherein a resolution of the gesture recognition is finer than a wavelength of the radar field.
10. The computer-implemented method of claim 1 , wherein the single non-rigid target is a hand.
11. The computer-implemented method of claim 1 , wherein determining the first relative velocity and the second relative velocity does not include determining an absolute velocity of the first and second points.
12. The computer-implemented method of claim 1 , wherein the radar field is a broad beam, fully contiguous radar field.
13. An apparatus comprising:
at least one computer processor;
a radar system comprising:
at least one radar-emitting element configured to provide a radar field; and
one or more antenna elements configured to receive radar signals representing reflections of the radar field off points of a single non-rigid object within the radar field; and
one or more computer-readable storage media having instructions stored thereon that, responsive to execution by the processor, cause the at least one computer processor to:
receive first reflections of a radar field off first and second points of a single non-rigid target at a first time, the first and second points of the single non-rigid target moving at first and second Doppler-based velocities, respectively;
generate, based on the received first reflections, Doppler-based measurements representing a velocity profile of the single non-rigid target, the Doppler-based measurements including first and second measurements respectively associated with the first and second Doppler-based velocities, the first and second measurements having higher energies than another measurement corresponding to another point of the single non-rigid target that is moving at another Doppler-based velocity, and the first and second points moving relative to one another within the radar field;
determine, based on the first and second measurements, a first relative velocity between the first and second points at the first time;
receive second reflections of the radar field off the first and second points at a second time;
determine, based on the second reflections, a second relative velocity between the first and second points at the second time; and
determine, based on the first and second relative velocities, a gesture performed by the single non-rigid target.
14. The apparatus of claim 13 , wherein:
the instructions further cause the at least one computer processor to determine, based on the first and second relative velocities, a change in displacement between the first and second points between the first and second times; and
determining the gesture is based further on the change in the displacement.
15. The apparatus of claim 13 , wherein the instructions further cause the at least one computer processor to identify, based on a moving target indicator filter, the first and second points from the other point of the single non-rigid target within the first and second reflections.
16. The apparatus of claim 13 , wherein the instructions further cause the at least one computer processor to pass the gesture to an application executing on the apparatus or to a device effective to control or alter a display, function, or capability associated with the application or the device.
17. The apparatus of claim 13 , wherein
the Doppler-based measurements for the first and second points at the first and second times are determined to have one or more of high accuracy or low noise compared to other determined Doppler frequencies.
18. The apparatus of claim 13 , wherein a resolution of the gesture recognition is finer than a wavelength of the radar field.
19. The apparatus of claim 13 , wherein the radar field is a broad beam, fully contiguous radar field.
20. The apparatus of claim 13 , wherein the instructions further cause the at least one computer processor to determine the first relative velocity and the second relative velocity without determining an absolute velocity of the first and second points.
21. The apparatus of claim 13 , wherein the energies correspond to respective ranges to the first and second points and the other point of the single non-rigid target.
22. The apparatus of claim 13 , wherein the instructions further cause the at least one computer processor to create a range-Doppler-time data cube for the single non-rigid target, wherein the determination of the gesture is based further on the range-Doppler-time data cube.
23. The apparatus of claim 13 , wherein the single non-rigid target is a hand.