A nut runner, an attachment unit, and a sliding member are provided in a second arm that is a leading arm of the robot and has a leading end shaft movable along the direction of an elevation axis. The nut runner includes a screw fastening driver, a drive shaft rotationally driven by the screw fastening driver, and an extension bar that is so connected to rotate together with the drive shaft in the circumferential direction and to move in the axial direction. The attachment unit fixedly connects the screw fastening driver to the second arm so that a screw fastening axis parallel to the elevation axis serves as a rotation axis of the screw fastening driver. The sliding member is connected to the leading end shaft and supports the extension bar so as to move together in the axial direction and are rotatable relatively in the circumferential direction.

A robot includes, a holding unit configured to hold an object, an image pickup unit, and a predetermined first portion of the robot. The image pickup unit picks up images of the holding unit and the object using the first portion as a background.

A machine learning device for learning the fastening operation of a screw by a screwdriver includes a state observation unit for observing state variables consisting of at least one of a rotational speed of the screwdriver, a rotational direction of the screwdriver, a position of the screwdriver and an inclination of the screwdriver and at least one of a fastening quality of the screw fastened by the screwdriver and a fastening time for which the screw is fastened by the screwdriver, and a learning unit for learning at least one of the rotational speed, the rotational direction, the position and the inclination, observed by the state observation unit and at least one of a change of the fastening quality and a change of the fastening time, observed by the state observation unit in association with each other.