Method and apparatus for robot collision avoidance by full surface proximity detection
An apparatus for collision avoidance by surface proximity detection includes a plurality of piezoelectric elements disposed adjacent to a surface of an object, a memory storing instructions, and at least one processor configured to execute the instructions to control a first one among the piezoelectric elements to generate an acoustic wave along the surface of the object, and receive, via a second one among the piezoelectric elements, an acoustic wave signal corresponding to the generated acoustic wave. The at least one processor is further configured to execute the instructions to filter the received acoustic wave signal, using a band-pass filter for reducing noise of the received acoustic wave signal, obtain a proximity signal for proximity detection, from the filtered acoustic wave signal, using a linear time-invariant filter, and detect whether an obstacle is proximate to the surface of the object by inputting the obtained proximity signal into a neural network.
1. An apparatus for collision avoidance by surface proximity detection of an obstacle relative to an object, the apparatus comprising:
a plurality of piezoelectric elements attached to a surface of the object;
a memory storing instructions; and
at least one processor configured to execute the instructions to:
control a first one among the piezoelectric elements to generate an acoustic wave along the surface of the object;
receive, via a second one among the piezoelectric elements, an acoustic wave signal corresponding to an acoustic impedance change in an acoustic pressure field perturbed by a standing wave pattern or interference pattern established by the obstacle;
reduce noise of the received acoustic wave signal by filtering, using a band-pass filter, the received acoustic wave signal;
obtain a proximity signal for proximity detection, from the filtered acoustic wave signal, using a linear time-invariant filter; and
detect whether the obstacle is adjacent to the surface of the object by inputting the obtained proximity signal into a neural network.
2. The apparatus of claim 1 , wherein the object is a robot, and
the at least one processor is further configured to execute the instructions to, based on the obstacle being detected to be adjacent to the surface of the object, control the object to avoid collision with the obstacle.
3. The apparatus of claim 2 , wherein the noise comprises electrical noise of a power supply of the robot and mechanical noise of a motor of the robot.
4. The apparatus of claim 1 , wherein the linear time-invariant filter comprises any one or any combination of a Hilbert transform and a Fourier transform of the filtered acoustic wave signal.
5. The apparatus of claim 1 , wherein the at least one processor is further configured to execute the instructions to:
truncate the received acoustic wave signal into a window; and
filter the truncated acoustic wave signal, using the band-pass filter.
6. The apparatus of claim 1 , wherein the at least one processor is further configured to execute the instructions to:
truncate the obtained proximity signal into a plurality of windows; and
detect whether the obstacle is adjacent to the surface of the object by inputting the truncated proximity signal into the neural network.
7. The apparatus of claim 1 , wherein the at least one processor is further configured to execute the instructions to:
truncate the obtained proximity signal into a plurality of windows;
input the truncated proximity signal into the neural network to obtain a plurality of predictions of whether the obstacle is adjacent to the surface of the object; and
detect whether the obstacle is adjacent to the surface of the object, based on a majority vote among the obtained plurality of predictions.
8. A method of collision avoidance by surface proximity detection of an obstacle relative to an object, the method being performed by at least one processor, and the method comprising:
controlling a first one among piezoelectric elements attached to a surface of the object, to generate an acoustic wave along the surface of the object;
receiving, via a second one among the piezoelectric elements, an acoustic wave signal corresponding to an acoustic impedance change in an acoustic pressure field perturbed by a standing wave pattern or interference pattern established by the obstacle;
reducing noise of the received acoustic wave signal by filtering, using a band-pass filter, the received acoustic wave signal;
obtaining a proximity signal for proximity detection, from the filtered acoustic wave signal, using a linear time-invariant filter; and
detecting whether the obstacle is adjacent to the surface of the object by inputting the obtained proximity signal into a neural network.
9. The method of claim 8 , wherein the object is a robot, and
the method further comprises, based on the obstacle being detected to be adjacent to the surface of the object, control the object to avoid collision with the obstacle.
10. The method of claim 9 , wherein the noise comprises electrical noise of a power supply of the robot and mechanical noise of a motor of the robot, from the received acoustic wave signal.
11. The method of claim 8 , wherein the linear time-invariant filter comprises any one or any combination of a Hilbert transform and a Fourier transform of the filtered acoustic wave signal.
12. The method of claim 8 , further comprising truncating the received acoustic wave signal into a window,
wherein the filtering comprises filtering the truncated acoustic wave signal, using the band-pass filter.
13. The method of claim 8 , further comprising truncating the obtained proximity signal into a plurality of windows,
wherein the detecting comprises detecting whether the obstacle is adjacent to the surface of the object by inputting the truncated proximity signal into the neural network.
14. The method of claim 8 , further comprising:
truncating the obtained proximity signal into a plurality of windows; and
inputting the truncated proximity signal into the neural network to obtain a plurality of predictions of whether the obstacle is adjacent to the surface of the object,
wherein the detecting comprises detecting whether the obstacle is adjacent to the surface of the object, based on a majority vote among the obtained plurality of predictions.
15. A non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor, cause the at least one processor to:
control a first one among piezoelectric elements attached to a surface of an object, to generate an acoustic wave along the surface of the object;
receive, via a second one among the piezoelectric elements, an acoustic wave signal corresponding to an acoustic impedance change in an acoustic pressure field perturbed by a standing wave pattern or interference pattern established by an obstacle;
reduce noise of the received acoustic wave signal by filtering, by using a band-pass filter, the received acoustic wave signal;
obtain a proximity signal for proximity detection, from the filtered acoustic wave signal, using a linear time-invariant filter; and
detect whether an obstacle is adjacent to the surface of the object by inputting the obtained proximity signal into a neural network.
16. The non-transitory computer-readable storage medium of claim 15 , wherein the object is a robot,
wherein the instructions, when executed by the at least one processor, further cause the at least one processor to, based on the obstacle being detected to be adjacent to the surface of the object, control the object to avoid collision with the obstacle, and
wherein the noise comprises electrical noise of a power supply of the robot and mechanical noise of a motor of the robot.
17. The non-transitory computer-readable storage medium of claim 15 , wherein the linear time-invariant filter comprises any one or any combination of a Hilbert transform and a Fourier transform of the filtered acoustic wave signal.
18. The non-transitory computer-readable storage medium of claim 15 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to:
truncate the received acoustic wave signal into a window; and
filter the truncated acoustic wave signal, using the band-pass filter.
19. The non-transitory computer-readable storage medium of claim 15 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to:
truncate the obtained proximity signal into a plurality of windows; and
detect whether the obstacle is adjacent to the surface of the object by inputting the truncated proximity signal into the neural network.
20. The non-transitory computer-readable storage medium of claim 15 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to:
truncate the obtained proximity signal into a plurality of windows;
input the truncated proximity signal into the neural network to obtain a plurality of predictions of whether the obstacle is adjacent to the surface of the object; and
detect whether the obstacle is adjacent to the surface of the object, based on a majority vote among the obtained plurality of predictions.