Decomposed perturbation approach using memory based learning for compliant assembly tasks

A robotic system for performing a positional search process in an assembly task, the robotic system comprising:

• a robotic arm;
• and
• a control system communicatively coupled to the robotic arm, the control system storing executable program instructions for causing the robotic arm to perform the steps of:
• decomposing, by the robotic system, a perturbation motion into a plurality of actions, the perturbation motion being a motion for an assembly position searched by the robotic system, each action of the plurality of actions related to a specific direction;
• performing reinforcement learning by selecting an action among decomposed actions and assembly movement actions at each step of the positional search process based on corresponding force-torque data received from at least one sensor associated with the robotic system;
• and
• outputting a best action at each step for completion of the assembly task as a result of the reinforcement learning.

A robotic system, comprising:

• a robot constructed to assemble a first component to a second component;
• at least one sensor coupled to the robot;
• a control system communicatively coupled to the robot and to the at least one sensor, the control system storing executable program instructions for executing assembly motion of the robot;
• a learning device communicatively coupled to the control system and operative to direct the robot via the control system, the control system decomposing a perturbation motion into a plurality of actions, the perturbation motion being a motion for an assembly position searched by the robotic system, each action of the plurality of actions related to a specific direction, and selecting, via reinforcement learning, an action among decomposed actions and assembly movement actions at each step of a positional search process based on corresponding force-torque data received from the at least one sensor;
• and
• an output device for outputting a best action at each step for completion of an assembly task as a result of the reinforcement learning.