Systems and methods for a passive grasping surface on an active grasping robotic manipulator
One or more embodiments of the present disclosure relate generally to the field of robotic grasping systems, and in particular to an active robotic manipulator that includes a passive grasping component so that the robotic manipulator can grasp a wide variety of objects and simultaneously provide soft grasping features which reduce the risk of damage to objects.
1. A system for a robotic manipulator having an active grasping component that allows for passive grasping of an object, comprising:
a controller;
at least two fingers configured to provide an active grasp of the object;
at least two passive grasping surfaces, each one of the at least two passive grasping surfaces coupled to a respective one of the at least two fingers, wherein each of the passive grasping surfaces includes a deformable shell having an interior containing a medium;
a storage container configured to store the medium; and
a pump communicatively coupled to the controller, the pump configured to deliver the medium from the storage container to the interior of at least one shell of the passive grasping surfaces, the pump further configured to suction the medium from the interior of the at least one shell.
2. The system of claim 1 , wherein the pump is further configured to deliver fluid pressure to the interior of the at least one shell and to suction fluid from the interior of the at least one shell.
3. The system of claim 2 , wherein the fluid is selected from a group consisting of a liquid and a gas.
4. The system of claim 2 , wherein the medium is configured to compress when the pump suctions fluid from the interior of the at least one shell.
5. The system of claim 2 , wherein the at least one shell is configured to conform to a shape of the object when the pump suctions fluid from the interior of the shell.
6. The system of claim 1 , wherein the medium is selected from a group consisting of coffee grounds, sand, rice, metal fragments, rubber fragments, sawdust, flour, salt, and rocks.
7. The system of claim 1 , wherein the medium is selected from a group consisting of polymers, foam, elastomers, and silicone.
8. The system of claim 1 , further comprising an actuator communicatively coupled to the controller, wherein the controller is configured to control the actuator and the pump based on learned data generated from a machine learning model that has processed sensor data from the robotic manipulator.
9. The system of claim 1 , further comprising an actuator communicatively coupled to the controller for controlling movement of the fingers, wherein the controller is configured to independently control each of the fingers.
10. A system for a robotic manipulator having an active grasping component that allows for passive grasping of an object, comprising:
at least two fingers configured to provide an active grasp of the object;
at least two passive grasping surfaces, each one of the at least two passive grasping surfaces coupled to a respective one of the at least two fingers, wherein each of the passive grasping surfaces includes a deformable shell having an interior containing a medium;
a storage container configured to store the medium;
a pump communicatively coupled to a controller, the pump configured to deliver the medium from the storage container to the interior of at least one shell of the passive grasping surfaces, the pump further configured to suction the medium from the interior of the at least one shell; and
the controller, the controller configured to:
determine whether the at least two fingers are over-grasping an object; and
responsive to the determining that at least two fingers are over-grasping the object, instruct the pump to apply positive pressure to the interior of the deformable shell of at least one of the passive grasping surfaces.
11. The system of claim 10 , wherein the pump is further configured to deliver fluid pressure to the interior of the at least one shell and to suction fluid from the interior of the at least one shell.
12. The system of claim 11 , wherein the medium is configured to compress when the pump suctions fluid from the interior of the at least one shell.
13. The system of claim 11 , wherein the at least one shell is configured to conform to a shape of the object when the pump suctions fluid from the interior of the shell.
14. The system of claim 10 , further comprising an actuator communicatively coupled to the controller, wherein the controller is configured to control the actuator and the pump based on learned data generated from a machine learning model that has processed sensor data from the robotic manipulator.
15. The system of claim 10 , further comprising an actuator communicatively coupled to the controller for controlling movement of the fingers, wherein the controller is configured to independently control each of the fingers.
16. One or more instances of non-transitory computer-readable media storing instructions or data, wherein the instructions or data cause one or more processors to perform a method comprising:
determining whether at least two fingers of a robotic manipulator are over-graphing an object, wherein each of the at least two fingers are coupled to at least one passive grasping surface comprising a deformable shell having an interior containing a medium; and
responsive to a determination that the at least two fingers are over-grasping the object, causing a pump to change a pressure of the interior of the deformable shell of at least one of the passive grasping surfaces by:
delivering a medium from a storage container to at least one of the passive grasping surfaces; or
suctioning the medium from at least one of the passive grasping surfaces.
17. The one or more instances of non-transitory computer readable media of claim 16 , wherein the pump is further configured to deliver fluid pressure to the interior of the deformable shell of at least one of the passive grasping surfaces and to suction fluid from the interior of the deformable shell.
18. The one or more instances of non-transitory computer readable media of claim 17 , wherein the medium is configured to compress when the pump suctions fluid from the interior of the deformable shell.
19. The one or more instances of non-transitory computer readable media of claim 17 , wherein the at least one shell is configured to conform to a shape of the object when the pump suctions fluid from the interior of the deformable shell.
20. The one or more instances of non-transitory computer readable media of claim 16 , wherein the method further comprises:
causing the pump to change the pressure of the interior of the deformable shell based on learned data generated from a machine learning model that has processed sensor data from the robotic manipulator.