Robotic touch perception
An apparatus such as a robot capable of performing goal oriented tasks may include one or more touch sensors to receive touch perception feedback on the location of objects and structures within an environment. A fusion engine may be configured to combine touch perception data with other types of sensor data such as data received from an image or distance sensor. The apparatus may combine distance sensor data with touch sensor data using inference models such as Bayesian inference. The touch sensor may be mounted onto an adjustable arm of a robot. The apparatus may use the data it has received from both a touch sensor and distance sensor to build a map of its environment and perform goal oriented tasks such as cleaning or moving objects.
1. A method for operating a robot using robotic touch perception, the method comprising:
receiving, at a processor from a touch sensor, touch data associated with an object, the touch data including measurements of force feedbacks received by the touch sensor when forces are applied by the robot to the object, wherein generating the touch data comprises:
directing a part of the robot carrying the touch sensor towards the object;
attempting to move the object for a distance; and
detecting force values in attempting to move the object;
determining, based on the touch data associated with the object, whether the object is movable;
moving, in response to a determination that the object is movable, the object by the robot from an original location to another location; and
controlling the robot to perform a goal directed action relative to the original location.
2. The method of claim 1 , further comprising:
detecting one or more attributes of the object;
storing the attributes of the object in an object library; and
marking the object as movable.
3. The method of claim 2 , wherein the one or more attributes comprise image data and touch data.
4. The method of claim 2 , wherein the one or more attributes comprise an expected amount of force required to move the object.
5. The method of claim 1 , wherein attempting to move the object for the distance comprises:
moving an arm of the robot towards a surface of the object until pressure is first detected at a position of the arm;
attempting to move the arm further towards the surface until the pressure reaching a specific maximum allowable pressure; and
returning the arm to the position of the arm in which the pressure is first detected.
6. The method of claim 1 , further comprising determining a contour of the object.
7. The method of claim 6 , wherein determining the contour of the object comprises:
directing a part of the robot carrying the touch sensor towards a surface of the object; and
directing the part carrying the touch sensor to move in two or more directions of travel across the surface while maintaining contact of the touch sensor with the surface.
8. The method of claim 1 , wherein the goal directed action corresponds to a movement of a part of the robot along a surface.
9. The method of claim 1 , wherein the goal directed action corresponds to a cleaning task.
10. The method of claim 1 , further comprising generating a three-dimensional map of touch data characteristics of the object.
11. The method of claim 1 , further comprising
attempting to move a second object;
determining, based on touch data associated with the second object, that the second object is beyond a capacity of the robot to move the second object; and
causing the robot to cancel an attempt to move the second object.
12. A robot comprising:
a robotic arm;
a touch sensor carried by the robotic arm, the touch sensor configured to generate touch data associated with an object, the touch data including measurements of force feedbacks received by the touch sensor when forces are applied by the robot to the object, wherein generating the touch data comprises:
directing the robotic arm carrying the touch sensor towards the object;
attempting to move the object for a distance; and
detecting force values in attempting to move the object; and
a processor in communication with the robotic arm and with the touch sensor; and
memory coupled to the processor, the memory storing instructions, the instructions, when executed by the processor, cause the processor to perform steps comprising:
determining, based on the touch data associated with the object, whether the object is movable;
directing, in response to a determination that the object is movable, the robotic arm to move the object from an original location to another location; and
directing the robotic arm to perform a goal directed action relative to the original location.
13. The robot of claim 12 , wherein the steps further comprise:
detecting one or more attributes of the object;
storing the attributes of the object in an object library; and
marking the object as movable.
14. The robot of claim 13 , wherein the one or more attributes comprise image data and touch data.
15. The robot of claim 13 , wherein the one or more attributes comprise an expected amount of force required to move the object.
16. The robot of claim 12 , wherein attempting to move the object for the distance comprises:
moving an arm of the robot towards a surface of the object until pressure is first detected at a position of the arm;
attempting to move the arm further towards the surface until the pressure reaching a specific maximum allowable pressure; and
returning the arm to the position of the arm in which the pressure is first detected.
17. The robot of claim 12 , wherein the steps further comprise determining a contour of the object.
18. The robot of claim 17 , wherein determining the contour of the object comprises:
directing the robotic arm carrying the touch sensor towards a surface of the object; and
directing the robotic arm carrying the touch sensor to move in two or more directions of travel across the surface while maintaining contact of the touch sensor with the surface.
19. The robot of claim 12 , wherein the goal directed action corresponds to a movement of a part of the robot along a surface.
20. The robot of claim 12 , wherein the goal directed action corresponds to a cleaning task.
21. The robot of claim 12 , wherein the steps further comprise generating a three-dimensional map of touch data characteristics of the object.
22. The robot of claim 12 , wherein the steps performed by the processor further comprise:
attempting to move a second object;
determining, based on touch data associated with the second object, that the second object is beyond a capacity of the robot to move the second object; and
causing the robot to cancel an attempt to move the second object.