ROS package accompanying "On Infusing Reachability-Based Safety Assurance within Probabilistic Planning Frameworks for Human-Robot Vehicle Interactions" by Karen Leung*, Edward Schmerling*, Mo Chen, John Talbot, J. Christian Gerdes, Marco Pavone) published in the 2018 International Symposium on Experimental Robotics. An extended journal version is to appear in the International Journal of Robotics Research.
This ROS package performs the prediction and planning for a pairwise traffic-weaving interaction detailed in our ICRA2018 paper, Multimodal Probabilistic Model-Based Planning for Human-Robot Interaction by Edward Schmerling, Karen Leung, Wolf Vollprecht, Marco Pavone. In a nutshell, it predictions are made about what the human-driven car will do in the future, and the robot selects an optimal action sequence to perform a traffic-weaving maneuver with the human-driven car. A desired trajectory (ROS msg) is produced. In our experiments, this trajectory is fed into a lower-level tracking controller. We use Pigeon.jl
.