A robotic control system directs a robot to take an object from a human grasp by obtaining an image of a human hand holding an object, estimating the pose of the human hand and the object, and determining a grasp pose for the robot that will not interfere with the human hand. In at least one example, a depth camera is used to obtain a point cloud of the human hand holding the object. The point cloud is provided to a deep network that is trained to generate a grasp pose for a robotic gripper that can take the object from the human's hand without pinching or touching the human's fingers.

A robotic control system directs a robot to take an object from a human grasp by obtaining an image of a human hand holding an object, estimating the pose of the human hand and the object, and determining a grasp pose for the robot that will not interfere with the human hand. In at least one example, a depth camera is used to obtain a point cloud of the human hand holding the object. The point cloud is provided to a deep network that is trained to generate a grasp pose for a robotic gripper that can take the object from the human's hand without pinching or touching the human's fingers.

A human cooperation robot system includes: an external force detecting unit that detects an external force acting on a robot; a retreat operation commanding unit that commands a retreat operation for causing the robot to be moved in a direction such that the external force is decreased when the external force detected by the external force detecting unit is greater than a first threshold value; and an external force variation monitoring unit that stops the retreat operation when a variation width of the external force at a predetermined time after the retreat operation is commanded by the retreat operation commanding unit is smaller than a second threshold value.

A robotic control system directs a robot to take an object from a human grasp by obtaining an image of a human hand holding an object, estimating the pose of the human hand and the object, and determining a grasp pose for the robot that will not interfere with the human hand. In at least one example, a depth camera is used to obtain a point cloud of the human hand holding the object. The point cloud is provided to a deep network that is trained to generate a grasp pose for a robotic gripper that can take the object from the human's hand without pinching or touching the human's fingers.

A robot system includes a robot, a sensor, and a processor. The sensor is configured to detect an external force acting on the robot. The processor is configured to move the robot in a forward direction such that a representative point of the robot moves along a motion track in the forward direction; move the robot in a reverse direction such that the representative point moves along the motion track in the reverse direction opposite to the forward direction when the external force satisfies a first condition which includes a condition that the external force is larger than a first threshold force; and move the robot to reduce the external force when the external force satisfies a second condition which includes a condition that the external force is larger than a second threshold force even after the robot has been moved in the reverse direction.

A robot gripper includes: a drive unit to drive a powertrain with active elements, wherein each element has a working region arranged in a body-fixed manner relative to the robot gripper, a respective element being moveable in and capable of reaching the working region; a control unit to control the drive unit; and a sensor system connected to the control unit to ascertain forces/moments applied externally to individual elements, the control unit configured such that collision monitoring is capable of being carried out for the elements, and when a collision is detected for an element, the drive unit is actuated according to a specified operation, including: providing a defined region within the working region for the elements, and collision monitoring for the elements only when the elements are located outside the assigned region, and deactivating collision monitoring when the elements are located at least partly within the assigned region.

In the field of robot technologies, a robot control method and apparatus are for protecting safety of a human during interaction between the human and a robot. The method includes: detecting a current location of the human; determining at least two regions surrounding the human according to the detected current location; and controlling movement of the robot in any region of the at least two regions, to protect safety of the human during interaction with the robot. Because the region is set surrounding the human, movement of the human does not affect the interaction between the human and the robot. In addition, using a protected object as a target, the robot is controlled to move in regions that surround the human. Compared with a case in which the robot is limited in fixed space, no matter how the human moves, safety of the human may be effectively protected.

A sensor detection error at a joint of a robot arm is correctly detected. A joint structure that joins links and of a robot arm includes a sensor for determining force acting between the links. A driving apparatus that generates a driving force of a joint includes first and second driving parts. A constraining part that constrains the joint movable in a driving direction of the joint and be unmovable in another direction includes first and second supporting parts that are movable relative to each other in the driving direction of the joint. The driving part of the driving apparatus is fixed to the link, and the supporting part of the constraining part is fixed to the link. Also, the supporting part of the constraining part is fixed to the driving part of the driving apparatus. The sensor is fixed so as to link the supporting part and the link.

A robot system is provided with a robot, a hand, and a range sensor that acquires information about a distance to a workpiece. Claw parts of the hand include grip regions having a shape for gripping the workpiece. A controller stops the robot when the hand collides with the workpiece. The controller determines whether or not a target workpiece can be gripped by a part other than the grip regions of the hand. When it is determined that the target workpiece can be gripped by the part other than the grip regions, the controller performs control for gripping the target workpiece by using the hand, and for moving the target workpiece inside a container.

Provided is a highly safe robot system. A robot system includes: an arm operation control unit that controls operations of an arm; and a storage unit that records hazardous part information related to at least any one of a grip unit of the arm and a work target gripped with the grip unit, and the arm operation control unit causes the arm to operate such that the hazardous direction of the hazardous part is different from a moving direction of the grip unit on the basis of the hazardous part information.