A method for assembling an operating member and an adapting member by a robot is provided. According to the method, the operating member is moved to the adapting member in an assembly direction until the operating member reaches the adapting member. A moving mode for adjusting the operating member is determined, in which the moving mode includes one or both of a moving-in-plane mode and a posture-adjusting mode. The operating member is moved in the determined moving mode and applied a force continuously in the assembly direction during movement. It is determined whether a preset condition for a completion of the assembling of the operating member and the adapting member is satisfied. And the movement of the operating member in the determined moving mode is stopped or the applied force is stopped when it is determined that the preset condition is satisfied.

Techniques are disclosed for training and applying machine learning models to control robotic assembly. In some embodiments, force and torque measurements are input into a machine learning model that includes a memory layer that introduces recurrency. The machine learning model is trained, via reinforcement learning in a robot-agnostic environment, to generate actions for achieving an assembly task given the force and torque measurements. During training, experiences are collected as transitions within episodes, the transitions are grouped into sequences, and the last two sequences of each episode have a variable overlap. The collected transitions are stored in a prioritized sequence replay buffer, from which a learner samples sequences to learn from based on transition and sequence priorities. Once trained, the machine learning model can be deployed to control various types of robots to perform the assembly task based on force and torque measurements acquired by sensors of those robots.

An adjustment support device includes: a storage unit for storing, with force state data and position data in an operation when performing force control of the industrial robot as a state variable and with data indicating a result of determining whether a result of the force control is success or failure based on predetermined criteria as determination data, a learning model generated by machine learning; an analysis unit for analyzing the learning model to analyze, for a control parameter used when the force control of the industrial robot has failed, an adjustment method of the control parameter for improving a degree of success of the force control; and an adjustment determination unit for determining, based on a result of the analysis by the analysis unit, an adjustment method of the control parameter in the force control used when the force control has failed and outputting the adjustment method.

A fixturing system for processing components includes a work surface, such as a table formed from a ferromagnetic material, a tool fixture having a magnet operable for securing the tool fixture to the work surface and having at least one alignment surface to hold and/or align a component to the tool fixture and to the work surface. The system further includes a storage location spaced from the work surface for storing the tool fixture when not in use on the work surface, a work piece, and optionally a robotic arm with a controller to move the work piece and robotic arm, and a computer having stored therein component location data. The computer is configured to control the work piece to retrieve the tool fixture from the storage location and to place the tool fixture on the work surface at a tool fixture location based on the component location data for aligning, and optionally holding the two components with the tool fixture while being processed.

To facilitate mounting of a second member on a first member having a plate shape while the first member is supported in a state of being deformable due to gravity. A method for manufacturing a component includes: a step (S13) where a supporting robot supports a skin having a plate shape in a state where the skin is deformable due to gravity, and a support state of the skin is changed based on teaching data based on a deformed shape of the skin, which is recorded in advance in a storage unit and which is obtained at the time of mounting a frame on the skin; and a step (S15) where a mounting robot makes the frame overlap with the skin where a support state is changed.

A first characteristic portion of a first workpiece and a second characteristic portion of a second workpiece are previously determined. A characteristic amount detection unit detects a first characteristic amount related to the position of the first characteristic portion and a second characteristic amount related to the position of the second characteristic portion in an image captured by a camera. A calculation unit calculates, as a relative position amount, the difference between the first characteristic amount and the second characteristic amount. A command generation unit generates a movement command for operating a robot based on a relative position amount in the image captured by the camera and a relative position amount in a predetermined reference image.

A robot control system includes a relative relationship calculating section configured to calculate a relative relationship between a first member and a second member at least at one of a plurality of points based on data acquired by a vision sensor, a contact point determination section configured to determine a contact point between the first member and the second member based on the calculated relative relationship, a control point setting section configured to set a control point based on the determined contact point, and a fitting control section configured to control fitting of the plurality of points based on the set control point.

A component manufacturing method includes a step of calculating, for a plurality of keyholes formed on a skin and disposed in a row along a first axial direction on the skin, a first imaginary line that passes an average position in a second axial direction perpendicular to the first axial direction and is parallel to the first axial direction, a step of calculating, for a plurality of keyholes formed on a frame and disposed in a row along a third axial direction on the frame, a second imaginary line that passes an average position in a fourth axial direction perpendicular to the third axial direction and is parallel to the third axial direction, and a step of superimposing the skin and the frame such that the first imaginary line and the second imaginary line coincide.

An adjustment support device includes: a storage unit for storing, with force state data and position data in an operation when performing force control of the industrial robot as a state variable and with data indicating a result of determining whether a result of the force control is success or failure based on predetermined criteria as determination data, a learning model generated by machine learning; an analysis unit for analyzing the learning model to analyze, for a control parameter used when the force control of the industrial robot has failed, an adjustment method of the control parameter for improving a degree of success of the force control; and an adjustment determination unit for determining, based on a result of the analysis by the analysis unit, an adjustment method of the control parameter in the force control used when the force control has failed and outputting the adjustment method.

A control device includes a processor that is configured to execute computer-executable instructions so as to control a robot provided with a manipulator, wherein in a case where the processor is configured to cause an end effector connected to the manipulator assemble a first object held by the end effector to a second object and a third object, the processor is configured to: cause the first object to come into contact with at least one of the second object and the third object; rotate the first object around a second rotation axis intersecting a first rotation axis while rotating the first object around the first rotation axis to assemble the second object and the first object to each other; and thereafter, rotate the first object around a third rotation axis intersecting the first rotation axis to assemble the third object and the first object to each other.