Hold position determination device and hold position determination method
A holding position determination device includes a control unit that determines the position where an end effector adapted to hold a holding target contacts the holding target as a holding position. The control unit acquires an end effector model that identifies an area where a holding portion of the end effector can exist. The control unit acquires a rule map that includes a map defining the position of the holding target to be held by the end effector based on an image of holding target obtained by photographing the holding target. The control unit determines the holding position based on the end effector model and the rule map.
1 . A holding position determination device comprising: at least one processor a control unit configured to determine a position where an end effector, including two or more fingers, adapted to hold a holding target contacts the holding target holding position, the at least one processor control unit further configured to acquire an end effector model that includes a stroke model that identifies a stroke range which represents a maximum opening width of the fingers identifies an area where a holding portion of the end effector can exist, acquire a rule map that includes a map defining the position of the holding target to be held by the end effector based on an image of holding target obtained by photographing the holding target, wherein the rule map includes a surrounding environment map that identifies an object area, which represents an area where the holding target exists at a height of the position at which the holding target is held, project the end effector model, as a temporary holding position of the holding target, onto the rule map, and determine that the end effector model cannot be projected when a position of one of the fingers overlaps the object area, determine the holding position based on the end effector model and the rule map, and generate control signals to actuate the end effector at the holding position to physically grasp the holding target and move the holding target from a work start platform to a work target platform.
2 . The holding position determination device according to claim 1 , wherein the at least one processor is further configured to acquire the rule map based on at least one of shape data of the holding target and/or depth data associated with the image of the holding target.
3 . The holding position determination device according to claim 1 , wherein the rule map includes a height map generated based on the position of the holding target in the height direction at which the holding target is to be held.
4 . The holding position determination device according to claim 1 , wherein the at least one processor is further configured to;
calculate an appropriate value representing appropriateness of holding position for a plurality of holding positions based on the end effector model and the rule map, and
determine the holding position based on the appropriate value.
5 . The holding position determination device according to claim 4 , wherein
the end effector model and the rule map are represented by values assigned to each position that indicate appropriateness as the position for holding the holding target, and
the at least one processor is further configured to calculate the appropriate value by calculating the values assigned to the each position when the end effector model is overlapped with the rule map.
6 . The holding position determination device according to claim 4 , wherein
the rule map comprises a plurality of different types of maps, and
the at least one processor is further configured to calculate the appropriate value based on the plurality of maps included in the rule map and a map coefficient determined for each map of the plurality of maps.
7 . The holding position determination device according to claim 1 , wherein
the at least one processor is further configured to:
acquire information on annotations for a plurality of holding positions and
correct an appropriate value based on the information on annotations.
8 . The holding position determination device according to claim 1 , wherein
the rule map includes a surrounding environment map that defines a range in which the end effector can move when the end effector holds the holding target, and
the at least one processor is further configured to determine the holding position from a plurality of positions of the end effector that conforms to the surrounding environment map.
9 . The holding position determination device according to claim 8 , wherein
in the surrounding environment map, a first area in which the end effector can exist and a second area in which the end effector cannot exist are defined, and
the second area is defined as at least the holding target and/or a surrounding area of the holding target, and in the surrounding area, the second area is assigned such that the proportion of the second area in the first area decreases as separating from the holding target.
10 . The holding position determination device according to claim 1 , wherein
the rule map includes an object map representing appropriateness of positions for the holding target, and
the at least one processor is further configured to determine the holding position from a plurality of positions of the end effector that conforms to the object map.
11 . The holding position determination device according to claim 10 , wherein the object map is defined based on properties possessed by the holding target.
12 . The holding position determination device of claim 11 , wherein the object map is defined based on at least one of the shape, material, texture, weight, and/or friction coefficient of the holding target.
13 . The holding position determination device according to claim 1 , wherein
the rule map includes a contact map representing appropriateness of a position of a contact portion of the end effector with the holding target as a position to contact the holding target, and
the at least one processor is further configured to determine the holding position from a plurality of positions of the end effector relative to the contact map.
14 . The holding position determination device according to claim 13 , wherein the contact map is defined based on a shape of the contact portion of the end effector with the holding target and a shape of the holding target.
15 . The holding position determination device according to claim 1 , wherein
the end effector is a gripper that grasps the holding target by sandwiching it between at least two fingers,
the end effector model defines a movement range of the fingers, and
the at least one processor is further configured to determine a combination of positions where the fingers contact the holding target when the gripper grasps the holding target as the holding position.
16 . The holding position determination device according to claim 1 , wherein
the end effector includes a suction portion that holds the holding target by suction,
the end effector model defines a range in which the suction portion interferes with other objects when the end effector sucks the holding target, and
the at least one processor is further configured to determine the position where the suction portion contacts the holding target when the end effector sucks the holding target as the holding position.
17 . The holding position determination device according to claim 1 , wherein the at least one processor is further configured to:
acquire an image and depth data of the holding target obtained by photographing the holding target,
generate a mask image representing an outline of the holding target based on the image and depth data,
generate the rule map comprising a plurality of different types of maps based on the mask image.
18 . A holding position determination method, comprising: acquiring, by a holding position determination device that determines the position where an end effector, including two or more fingers, adapted to hold a holding target contacts the holding target at a holding position, an end effector model that includes a stroke model that identifies a stroke range which represents a maximum opening width of the fingers identifies an area where a holding portion of the end effector can exist; ac quiring, by the holding position determination device, a rule map that includes a map defining the position of the holding target to be held by the end effector based on an image of holding target obtained by photographing the holding target, wherein the rule map includes a surrounding environment map that identifies an object area, which represents an area where the holding target exists at a height of the position at which the holding target is held, projecting the end effector model, as a temporary holding position of the holding target, onto the rule map, and determining that the end effector model cannot be projected when a position of one of the fingers overlaps the object area; determining, by the holding position determination device, the holding position based on the end effector model and the rule map, and generating control signals to actuate the end effector at the holding position to physically grasp the holding target and move the holding target from a work start platform to a work target platform.
19 . A holding position determination device comprising:
at least one processor configured to:
determine a position where an end effector adapted to hold a holding target contacts the holding target as a holding position,
acquire an end effector model that identifies an area where a holding portion of the end effector can exist,
acquire a rule map comprising a plurality of different types of maps that includes a first map defining the position of the holding target to be held by the end effector based on an image of holding target obtained by photographing the holding target,
calculate an appropriate value representing appropriateness of holding position for a plurality of holding positions based on the plurality of different types of maps included in the rule map and a map coefficient determined for each map of the plurality of different types of maps,
determine the holding position based on the appropriate value, and
generate control signals to actuate the end effector at the holding position to physically grasp the holding target and move the holding target from a work start platform to a work target platform.