Systems, methods, and apparatus for 3-D surface mapping, compliance mapping, and spatial registration with an array of cantilevered tactile hair or whisker sensors
Systems, methods, and apparatus are provided using signals from a set of tactile sensors mounted on a surface to determine a surface topography. An example method includes receiving a set of moment and force input data from one or more identified topographies. The example method includes using a neural network to receive input from a training data set based on the first set of moment and force input data from the one or more identified topographies. Network weights to be used by the neural network to produce the training data set are modified via an evolutionary algorithm that tests vectors of candidate network weights. The example method includes receiving a moment and force input from a test object surface and reconstructing the surface topology based on the neural network outputs.
1. A method of determining surface topography of an object, the method comprising:
receiving a first set of moment and force input data from one or more identified topographies;
using a function approximator to receive a training data set based on the first set of moment and force input data from the one or more identified topographies, wherein parameters to be used by the function approximator are modified via an evolutionary algorithm that tests vectors of candidate parameters; and
receiving a second set of moment and force input data from an object surface and using the second set of data to provide a surface mapping of the object, wherein the surface mapping comprises a spatial compliance map of the surface, wherein each contact point in the mapping is to be associated with a compliance value represented as a rate of surface compression with respect to a change in contact force.
2. The method of claim 1 , wherein the evolutionary algorithm employs a fitness function
f
=
1
n
∑
i
=
1
n
(
δ
tol
-
δ
i
)
to modify parameters.
3. The method of claim 1 , further comprising spatially registering local topographic mappings to form a combined mapped surface.
4. The method of claim 1 , wherein the second set of moment and force input data from an object surface is to be obtained using a sensor array of tactile whisker or hair-like sensors.
5. The method of claim 4 , where the sensor array comprises a four by four array.
6. The method of claim 1 , wherein the surface mapping comprises a three-dimensional mapping.
7. The method of claim 1 , wherein the function approximator comprises at least one of a multilayer perception network and a radial basis function network.
8. The method of claim 1 , wherein the moment and force input data comprise two-dimensional moment and one-dimensional force input data.
9. A method of determining material compliance of an object, the method comprising:
receiving a first set of moment and force input data from identified topographies and compliances;
training a function approximator using a training data set based on the first set of moment and force input data from the identified topography and compliances, wherein compliance is defined as a change in compression per change in contact force; and
receiving a second set of moment and force input data from an object surface to form a local topographic and compliance mapping of the object, the forming of the compliance mapping including spatially registering local compliance mappings to form a combined compliance-mapped surface.
10. The method of claim 9 , wherein parameters to be used by the function approximator are modified via an evolutionary algorithm that tests vectors of candidate network weights and wherein the evolutionary algorithm employs a fitness function
f
=
1
n
∑
i
=
1
n
(
δ
tol
-
δ
i
)
to modify parameters in the function approximator.
11. The method of claim 9 , wherein the second set of moment and force input data from the object surface is to be obtained using a sensor array of tactile whisker or hair-like sensors.
12. The method of claim 11 , where the sensor array comprises a four by four array.
13. The method of claim 9 , wherein the local topographic and compliance mapping comprises a spatial map of the surface.
14. The method of claim 13 , wherein the local topographic and compliance mapping comprises a compliance map of the surface, wherein each contact point is to be associated with a compliance value represented as a rate of surface deflection with respect to a change in contact force.
15. The method of claim 9 , wherein the mapping of the surface comprises a three-dimensional mapping.
16. The method of claim 9 , wherein the function approximator comprises at least one of a multilayer perception network and a radial basis function network.
17. The method of claim 9 , wherein the moment and force input data comprise two-dimensional moment and one-dimensional force input data.
18. A tangible computer readable storage medium comprising computer program code which, when executed by a processor, implements a method of determining surface topography of an object, the method comprising:
receiving a first set of moment and force input data from one or more identified topographies;
using a function approximator to receive a training data set based on the first set of moment and force input data from the one or more identified topographies, wherein parameters to be used by the functional approximator are modified via an evolutionary algorithm that tests vectors of candidate parameters;
receiving a second set of moment and force input data from an object surface and using the second set of data to provide a surface mapping of the object, wherein the evolutionary algorithm employs a fitness function
f
=
1
n
∑
i
=
1
n
(
δ
tol
-
δ
i
)
to modify parameters.
19. The computer readable storage medium of claim 18 , wherein the moment and force input data comprise two-dimensional moment and one-dimensional force input data.