Skip to content

ori-drs/trajectory_prediction_ros

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

37 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

trajectory_prediction_ros

This package implements the human trajectory prediction method as described in the paper:
"Motion Planning in Dynamic Environments Using Context-Aware Human Trajectory Prediction", Mark Nicholas Finean, Luka Petrović, Wolfgang Merkt, Ivan Marković, and Ioannis Havoutis, submitted to The International Journal of Robotics Research.

Link to: Preprint
Link to: Video
Link to: Dataset

Prerequisites & installation

If both GTSAM and GPMP2 are installed correctly, then this package can be built within a catkin workspace using:

catkin build

To run

Place the dataset bags in the bags folder. Place the corresponding map.yaml in the scripts folder. Edit the launch file to point to the correct bag. Run roscore and rviz. Launch the roslaunch file, which runs the map server, static_tf_publisher, trajectory_prediction node and plays the bag you selected in the launch file.

Inputs and outputs

The trajectory_prediction node subscribes to the following topics:

/vicon/person_1/person_1

It publishes the following:

/desired_goal

which is a PointStamped that publishes the current desired goal obtained by our intention recognition method,

/predicted_trajectory

which is a PoseArray that publishes the currently predicted trajectory obtained by running gpmp2 starting from the current position with the desired goal.

Launch commands

roslaunch trajectory_prediction trajectory_prediction.launch global_frame:=odom use_empty_distance_field:=true roslaunch trajectory_prediction hsr_sim.launch global_frame:=odom use_empty_distance_field:=true

Citing


If you use this work, please cite following publications:

@INPROCEEDINGS{Finean2022,
  author={Finean, Mark Nicholas and Petrović, Luka and Merkt, Wolfgang and Marković, Ivan and Havoutis, Ioannis},
  title={Motion Planning in Dynamic Environments Using Human Trajectory Prediction}, 
  year={2022},
  }

License


This repository is licensed under the BSD-3-Clause license.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published