Iterative-Sim-to-Real Transfer of Human-Robot Interaction

Home

Discover

Bookmarks

Keep Up With Latest Trending Papers. Computer Science, AI and more.Subscribe

CC BY

Source arXiv

Robotics

Iterative-Sim-to-Real Transfer of Human-Robot Interaction

i-Sim2Real: Reinforcement Learning of Robotic Policies in Tight Human-Robot Interaction Loops

We present an iterative-sim-to-real (i-s2r) method for training robotic policies that are proficient at interacting with humans upon deployment.

I-s2r bootstraps from a simple model of human behavior and alternates between training in simulation and deploying in the real world.

In each iteration, both the human behavior model and the policy are refined.

We evaluate our method on a real world robotic table tennis setting, where the objective for the robot is to play cooperatively with a human player for as long as possible.

We present results on an industrial robotic arm that is able to cooperatively play table tennis with human players, achieving rallies of 22 successive hits on average and 150 atbest.

For 80% of players, rally lengths are 70% to 175% longer compared to the sim-to-real baseline.

Authors

Saminda Abeyruwan, Laura Graesser, David B. D'Ambrosio, Avi Singh, Anish Shankar, Alex Bewley, Pannag R. Sanketi

◐ Recommended Follows

◐ Latest Activity