Continuous Real Time POMCP to Find-and-Follow People by a Humanoid Service Robot
Abstract
This study describes and evaluates two new methods for finding and following people in urban settings using a humanoid service robot: the Continuous Real-time POMCP method, and its improved extension called Adaptive Highest Belief Continuous Real-time POMCP follower. They are able to run in real-time, in large continuous environments. These methods make use of the online search algorithm Partially Observable Monte-Carlo Planning (POMCP), which in contrast to other previous approaches, can plan under uncertainty on large state spaces.
We compare our new methods with a heuristic person follower and demonstrate that they obtain better results by testing them extensively in both simulated and real-life experiments. More than two hours, over 3 km, of autonomous navigation during real-life experiments have been done with a mobile humanoid robot in urban environments.
Experimental Videos
Adaptive HB-CR-POMCP Follower
To perform the experiments, we update the belief in each iteration. When the person is visible, the Heuristic Follower is used to follow the person, but at the same time the belief of the person's position is updated using POMCP. When the person is not visible, a 2D histogram from the belief of the position of the person is generated. From this matrix the cell with the highest count is used as the robot's next goal.
Heuristic Follower
The robot's goal position is set just in front of the person. When the target is not visible anymore, the robot keeps going to the last person's position. When reaching this position, the robot stays and waits for the person to appear again.
To see the high resolution videos, visit the normal video page.