Grip device and controlling method thereof
A grip device includes a first finger with a groove on a gripping surface thereof; a second finger facing the first finger; a middle member inserted into the groove of the first finger and covering the gripping surface; a plurality of mono-axial force sensors each positioned on one of a plurality of inclined surfaces of the groove and configured to detect a force applied to the middle member; a driving motor configured to adjust a gripping force of the first finger and the second finger; and a processor. The processor is configured to calculate a force vector applied to the middle member based on sensing values received from the force sensors, and, based on a force component of the force vector in a direction parallel to a surface of the middle member being less than a preset value, apply a greater gripping force by the first and second fingers.
1 . A grip device comprising:
a first finger comprising a groove formed as a plurality of inclined surfaces on a gripping surface of the first finger;
a second finger facing the first finger;
a middle member comprising a protrusion region inserted into the groove on the gripping surface of the first finger, the middle member covering the gripping surface of the first finger, a first surface of the middle member facing the gripping surface of the first finger, a second surface of the middle member being opposite to the first surface;
a plurality of mono-axial force sensors, each mono-axial force sensor being respectively positioned on an inclined surface of the plurality of inclined surfaces and configured to detect a force only in one direction, the one direction being normal to the inclined surface, and the force being applied to the middle member by an object gripped between the first finger and the second finger,
wherein the middle member is formed of a rigid material configured to transmit the force to the plurality of mono-axial force sensors without loss;
a driving motor configured to adjust a gripping force of the first finger and the second finger; and
a processor configured to:
calculate a force vector applied to the middle member based on the forces detected by each of the plurality of mono-axial force sensors in the direction normal to each inclined surface, using sensing values received from the plurality of mono-axial force sensors, and
based on a first force component of the force vector in a direction parallel to the second surface of the middle member being less than a preset value when the first finger and the second finger grip an object by a first gripping force, control the driving motor such that the first finger and the second finger grip the object by a second gripping force that is greater than the first gripping force.
2 . The grip device of claim 1 , wherein the processor is further configured to, based on a second force component of the force vector in a second direction being less than a predetermined value when the first finger and the second finger move in a first direction opposite to the second direction while gripping the object, control the driving motor such that the first finger and the second finger grip the object with the second gripping force.
3 . The grip device of claim 1 , wherein the plurality of inclined surfaces comprise a first inclined surface and a second inclined surface, and
wherein the plurality of mono-axial force sensors comprise first and second mono-axial force sensors provided on the first and second inclined surfaces, respectively.
4 . The grip device of claim 3 , wherein the first inclined surface and the second inclined surface are angled to face at least partially vertically.
5 . The grip device of claim 1 , wherein the plurality of inclined surfaces comprise a first inclined surface, a second inclined surface, and a third inclined surface forming a triangular pyramid shape, and
wherein the plurality of mono-axial force sensors comprise first, second, and third mono-axial force sensors provided on the first, second, and third inclined surfaces, respectively.
6 . The grip device of claim 1 , wherein the plurality of inclined surfaces comprise a first inclined surface, a second inclined surface, a third inclined surface, and a fourth inclined surface forming a shape of a quadrangular pyramid, and
wherein the plurality of mono-axial force sensors comprise first, second, third, and fourth mono-axial force sensors provided on the first, second, third, and fourth inclined surfaces, respectively.
7 . The grip device of claim 1 , further comprising a camera configured to capture an image of the object,
wherein the processor is further configured to:
identify at least one of a shape, a weight, a physical property, and a volume of the object based on the captured image, and
determine the first gripping force based on a result of the identifying.
8 . The grip device of claim 7 , wherein the processor is further configured to:
determine a threshold gripping force at which the object begins to break based on the result of the identifying, and
based on the second gripping force being greater than or equal to the threshold gripping force, control the driving motor so that the first finger and the second finger decrease the gripping force.
9 . The grip device of claim 7 , further comprising a memory storing a learning model trained to identify a feature of the object based on an input of image information of the object,
wherein the processor is configured to identify the at least one of the shape, the weight, the physical property, and the volume of the object further based on the learning model.
10 . The grip device of claim 1 , wherein the second surface of the middle member is formed as a plane.
11 . A method of controlling a grip device comprising a first finger comprising a groove formed as a plurality of inclined surfaces on a gripping surface of the first finger, a second finger facing the first finger, a middle member comprising a protrusion region inserted into the groove on the gripping surface of the first finger, the middle member covering the gripping surface of the first finger, a first surface of the middle member facing the gripping surface of the first finger, a second surface of the middle member being opposite to the first surface, and a plurality of mono-axial force sensors, each mono-axial force sensor being respectively positioned on an inclined surface of the plurality of inclined surfaces and configured to detect a force only in one direction, the one direction being normal to the inclined surface, the force being applied to the middle member by an object gripped between the first finger and the second finger, the middle member being formed of a rigid material configured to transmit the force to the plurality of mono-axial force sensors without loss, the method comprising:
gripping an object between the first finger and the second finger by a first gripping force;
receiving sensing values from the plurality of mono-axial force sensors;
calculating a force vector applied to the middle member based on the forces detected by each of the plurality of mono-axial force sensors in the direction normal to each inclined surface, using the received sensing values;
confirming slip based on a comparison of a first force component of the force vector in a direction parallel to the second surface of the middle member and a preset value; and
based on a determination that slip has occurred, adjusting a gripping force of the first finger and the second finger to a second gripping force greater than the first gripping force.
12 . The method of claim 11 , wherein the confirming of the slip further comprises confirming slip based on a second force component of the force vector in a second direction being less than a predetermined value when the first finger and the second finger move in a first direction opposite to the second direction while gripping the object.
13 . The method of claim 11 , wherein the grip device further comprises a camera configured to capture an image of the object, the method further comprising:
identifying at least one of a shape, a weight, a physical property, and a volume of the object based on the captured image; and
determining the first gripping force based on a result of the identifying.
14 . The method of claim 13 , wherein the grip device further comprises a memory storing a learning model trained to identify a feature of the object based on an input of image information of the object, and
wherein the identifying of the at least one of the shape, the weight, the physical property, and the volume of the object is further based on the learning model.
15 . The method of claim 13 , further comprising:
determining a threshold gripping force at which the object begins to break based on the result of the identifying; and
based on the second gripping force being greater than or equal to the threshold gripping force, decreasing the gripping force applied by the first finger and the second finger.