FLOBOT is an EU funded Horizon 2020 project. The University of Lincoln is responsible for developing the human tracking for safety software module, and leading the environment reasoning and learning software module. In this repository, you find these two modules as a whole, and specifically:
-
bayestracking is a library originally developed by Dr. Nicola Bellotto.
-
detector_msg_to_pose_array, ground_plane_estimation, and upper_body_detector was originally developed by STRANDS project.
-
bayes_people_tracker is also from STRANDS but adapted according to the FLOBOT needs.
-
flobot_tracker_bringup, object3d_detector, pose_grid_map, and rgbd_leg_detector are originally developed by FLOBOT.
A detailed view of the proposed system in UML diagram is shown below:
In which leg_detector (and people_msgs) can be found here: https://github.com/wg-perception/people. For a more intuitive understanding, please refer to the following video.
Our system has been tested on Ubuntu 14.04 + ROS Indigo, as well as Ubuntu 16.04 + ROS Kinetic. For the latter:
$ sudo apt-get install ros-kinetic-people*
$ sudo apt-get install ros-kinetic-velodyne*
$ cd catkin_ws/src
$ git clone https://github.com/LCAS/FLOBOT
$ cd ../catkin_ws
$ catkin_make$ roslaunch flobot_tracker_bringup flobot_tracker.launchIf you are considering using this repository, please reference the following:
@article{zhimon2020,
author = {Zhi Yan and Simon Schreiberhuber and Georg Halmetschlager and Tom Duckett and Markus Vincze and Nicola Bellotto},
title = {Robot Perception of Static and Dynamic Objects with an Autonomous Floor Scrubber},
journal = {CoRR},
volume = {abs/2002.10158},
year = {2020},
url = {http://arxiv.org/abs/2002.10158},
archivePrefix = {arXiv},
eprint = {2002.10158}
}