Tactile sensor including a memory with a classification scheme and method for operating a tactile sensor

Patent metadata

Field	Value
Publication number	US-12092534-B2
Application number	US-202017622043-A
Country	US
Kind code	B2
Filing date	Jun 22, 2020
Priority date	Jun 24, 2019
Publication date	Sep 17, 2024
Grant date	Sep 17, 2024

Abstract

In an embodiment a tactile sensor includes a plurality of stress sensors and at least one contact body, wherein the stress sensors are configured to detect a load pattern applied on a detection surface of the contact body.

First claim

The invention claimed is:

1. A tactile sensor comprising:
a chip comprising a plurality of stress sensors and a memory with a classification scheme; and
at least one contact body,
wherein the stress sensors are integrated into the chip and are distributed over an area of the chip,
wherein the stress sensors are configured to detect a load pattern applied on a detection surface of the contact body, and
wherein the classification scheme assigns a set of output values of the plurality of stress sensors to a predefined load pattern.

2. The tactile sensor according to claim 1 ,
wherein the load pattern comprises a static tactile force and/or a dynamic tactile force, and
wherein the dynamic tactile force varies with a frequency between 1 Hz and 1000 Hz, inclusive.

3. The tactile sensor according to claim 1 , wherein the classification scheme is generated by a machine learning algorithm.

4. A method for operating a tactile sensor comprising a chip having a plurality of stress sensors and a memory with a classification scheme, and at least one contact body, wherein the stress sensors are integrated into the chip and are distributed over an area of the chip, the method comprising:
applying a load pattern to a detection surface of the contact body;
transducing, by the stress sensors, the load pattern to a set of output values; and
assigning, by the chip, the set of output values to a predefined class of load pattern by the classification scheme.

5. The method according to claim 4 , wherein load patterns resulting from static and dynamic tactile forces on the detection surface are classified by the same classification scheme.

6. The method according to claim 4 , wherein dynamic tactile forces are classified without spectral analysis of the output values.

7. A method for generating the classification scheme according to claim 4 , the method comprising:
predefining classes of load patterns;
performing multiple representative measurements of each predefined class of load pattern with a reference tactile sensor; and
defining a decision tree ensemble by assigning sets of output values to each predefined class of load pattern respectively.

8. The method according to the claim 7 , further comprising generating the classification scheme for a classification of load patterns resulting from static tactile forces and dynamic tactile forces.

9. A tactile sensor comprising:
a chip comprising a plurality of stress sensors and a memory with a classification scheme; and
at least one contact body,
wherein the stress sensors are integrated into the chip and are distributed over an area of the chip,
wherein the stress sensors are configured to detect a load pattern applied on a detection surface of the contact body,
wherein the stress sensors are transistor-based stress sensors and comprise a p-MOS transistor and an n-MOS transistor or p-MOS transistors and/or n-MOS transistors; and
wherein the classification scheme assigns a set of output values of the plurality of stress sensors to a predefined load pattern.

Assignees

• Univ Freiburg Albert Ludwigs

Inventors

• Kollmitz Jan
• Kuhl Matthias
• Manoli Yiannos

Classifications

• G01L5/162

  of piezoresistors · CPC title

• G01L1/205 Primary

  using distributed sensing elements · CPC title

• A61B2562/0261

  Strain gauges · CPC title

• G06N20/00

  Machine learning · CPC title

• A61B5/7267

  involving training the classification device · CPC title