People Find-and-Follow Behavior for Service Robots using Continuous Belief Maps
Alex Goldhoorn, Anaís Garrell, Fernando Herrero, René Alquézar and Alberto Sanfeliu
Abstract— Find-and-follow is an important behavior for social
robots which assist people. We introduce an on-line planning
method for the robot: Continuous Real-time POMCP (CR-POMCP). It uses Partially Observable Monte-Carlo Planning
(POMCP), which, in contrast to most other planning algorithms,
can plan under uncertainty, on large state spaces, and real-time.
Our method, CR-POMCP, also works in continuous state spaces,
and moreover takes into account sensory noise, false negatives,
and false positives. Comparisons have been done with a Heuristic
Follower, which simply follows the person. Two extensions to
CR-POMCP were tested to improve the robot’s performance
when doing real-life experiments. All variants tested
in simulation were found to be working better than the Heuristic
Follower. Dynamic obstacle were introduced in simulation by
adding randomly walking people.
Real-life experiments have been
done during a week with a mobile service robot in three urban
environments of Barcelona with other people walking around.
On this page videos are shown which give an overview of the experiments done with the find-an-follow methods. Like explained in the article, the robot has to search and follow the person. The person to follow is recognized using AR Markers.
An explanation of the the different areas of the video is given below.
More details about the used urban environments are given on
this page, where they also can be downloaded.
Heuristic Follower and Adaptive HB-CR-POMCP Follower (HB-CR-POMCP Searcher & Tracker):
The Heuristic Follower goes to the last position where it saw the person,
and when it reached that position it waits until it sees the person.
In contrary, the other methods, such as the Adaptive HB-CR-POMCP Follower have
a belief (i.e. 'memory') with the location of the person.
The video shows two experiments on the FME lab, where first the Heuristic Follower does not move until the person is visible. The Adaptive HB-CR-POMCP Follower, in contrary, does move, trying to find the person actively.
Adaptive HB-CR-POMCP Follower (HB-CR-POMCP Searcher & Tracker) in the BCN lab:
The method is shown to work in different large urban environments.
The robot finds and follows the person.
The video shows a part of several large experiments on the Barcelona Robot Lab. An example is shown where, when not seeing the person, the robot goes the other direction than the person goes. The robot followed the highest belief point, but the belief stays consistent with the person's movement.
Find:
This video shows that the method is to find the person, even if he/she has not yet been seen before.
Group:
The method is also shown to work when other people are 'disturbing' the robot's vision.
Long run in the Barcelona Robot Lab:
The person is being followed in the Barcelona Robot Lab.
Second long run in the Barcelona Robot Lab (Telecos Square):
The person is followed in another urban environment, the Telecos Square on the campus.
In the video the experiment is shown in three areas:
Video: shows the scenario.
Map: map as shown by ROS rviz; which shows the following:
Dabo: blue body, white head;
Obstacles: black and dark gray;
Laser detections: white line/dots;
People detection:
Leg detection: shown by white dots;
AR Marker detection: blue dot;
Last used person location: red dot, this is a combination of leg detection and AR Marker detection.
Belief Map: shows the robot's belief (i.e. probability matrix) of the person's location:
Dabo: the blue circle;
Obstacles: black squares;
Probability matrix: the probability of the person being on a certain location is shown with the colors white to red, the light blue color indicates a probability of 0, white is a low probability and red high. Note that in some videos these cells are bigger because the resolution of the probability matrix is set lower than of the discritized map.