Preference learning on the execution of collaborative human-robot tasks

Abstract

We present a novel method to learn human preferences during, and for, the execution of concurrent joint humanrobot tasks. We consider tasks realized by a team of a human operator and a robot helper that should adapt to the human's task execution preferences. Different human operators can have different abilities, experiences, and personal preferences, so that a particular allocation of activities in the team is preferred over another. We cast the behavior of concurrent multi-agent cooperation as a semi Markov Decision Process and show how to model and learn human preferences over the team behavior. After proposing two different interactive learning algorithms, we evaluate them and show that the system can effectively learn and adapt to human preferences.

Peg-in-Hole Assembly Based on Two-phase Scheme and F/T Sensor for Dual-arm Robot

“…In the past few years, focus on compliant dual-arm robots has increased due to their safety and human-like manipulation [ 1 , 2 , 3 , 4 ]., Compliant dual-arm robots are often limited to low positioning accuracy, and thus usually applied to complete a daily task [ 4 , 5 ] or an industrial task that does not require high positioning accuracy [ 6 , 7 ]. In order to extend a compliant dual-arm robot to complete a precision task, such as precision peg-in-hole assembly, a reasonable assembling strategy and sensor information are required.…”

Implementation of a Laboratory Case Study for Intuitive Collaboration Between Man and Machine in SME Assembly

“…Other systems based on Artificial Intelligence are based on a learning framework to construct an optimal task-sharing schedule. In works like Munzer et al (2017) and Mitsunaga et al (2008), the authors propose an online learning algorithm which adapts to the operator behavior during the task-sharing procedure. Given an initial task schedule, it is possible to adapt the robot's actions based on comfort and discomfort measurements gathered from the sensing system.…”

Physical Human–Robot Collaboration: Robotic Systems, Learning Methods, Collaborative Strategies, Sensors, and Actuators

“…Besides the autonomous mode of operation, a robot can collaborate with humans physically or remotely to perform a task. The collaboration perhaps includes shared knowledge, experience, and belief about the task goal to be achieved [2].…”