Data-Efficient Learning from Human Interventions for Mobile Robots
Zhenghao Peng , Zhizheng Liu , Bolei Zhou
University of California, Los Angeles
TL; DR
We train two mobile robots in real world in real time 
via human-in-the-loop learning! Our method:
Learns from online human intervention and demonstration!
Trains from scratch, without reward!
Safe Navigation
A delivery robot (Unitree Go2) learns to navigate safely in a real-world environment, avoiding collisions with static or dynamic obstacles. Even though the observation is RGB-D image, training can be done in 20 minutes and the robot can generalize to unseen environments.
Human Following
We train a quadruped robot (Unitree Go2) to follow a human subject in a real-world environment. Training is completed within 10 minutes.
Zero-shot Deployment
We have successfully deployed the learned policies to unseen environments.
Safe Navigation:
Human Following: