Skip to content

ROS wrapper for the Expressive Numerical Heuristic Search Planner (ENHSP)

License

Notifications You must be signed in to change notification settings

lsa-pucrs/ros_enhsp

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ros_enhsp

ROS wrapper for the Expressive Numerical Heuristic Search Planner (ENHSP). Part of project published as "Energy-Aware Path Planning for Autonomous Mobile Robot Navigation".

Description

This package integrates the ENHSP planner with the ROS framework (kinetic version). It can be used to solve planning problems with linear and non-linear numerical expressions, for example. It has four components (three nodes and one service server), inspired by the ROSplan environment:

  • Problem generator: Service server which generates STRIPS problems in PDDL
  • Problem interface: Service client which calls the problem generator
  • Planner interface: Node which gathers relevant information from ROS topics, as well as the STRIPS domain and problem, calls ENHSP to obtain a plan and publishes the plan in a ROS topic
  • Planner dispatch: Gets the plan, parses it and executes the instructions (optional)

Installation

The package can be installed via CATKIN. The planner interface node also requires the repository's local copy of ENHSP to be installed. To do this, open a terminal, navigate to the planner's folder and run "./install".

Dependencies

The package has no explicit dependencies, except for ROS itself and ENHSP. However, ROS and the planner's dependencies still have to be met.

If running the application described below, then the turtlebot stack must be installed, as well as the Turtlebot 2 simulator stack and ROS' navigation stack.

Application

The package is currently used for energy-efficient path planning in a simulated Turtlebot 2 robot. The domain present in the "common" folder reflects this planning problem, which considers high-energy zones when choosing a path to a goal. The energy-increasing and energy-decreasing actions are specific to a Jetson TX2 embedded computer, hard-coded in the planner dispatch node. More information in this paper.

Acknowledgements

To-Do:

  • Fix minor movement bug in planner dispatch (steering angle jumps to -360 when robot yaw is close to 180 degrees, due to atan2 discontinuity at 180 degrees).
  • Change planner interface node to load domain from ROS parameter.
  • Generalize problem generator node to allow it to work with any planning problem.

About

ROS wrapper for the Expressive Numerical Heuristic Search Planner (ENHSP)

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Java 47.5%
  • Python 44.0%
  • CMake 7.4%
  • Other 1.1%