I need a simple planner for finding a possible collision free path for a human

asked 2017-07-06 06:40:14 -0500

dmkarage gravatar image

updated 2017-07-07 06:29:55 -0500


I am looking for a simple path planner that takes into account a start pose, a goal pose and the (global) costmap of the environment and plans the shortest collision free path from start to goal. I know that there are available planners like A* or Dijkstra in the navigation stack, but they are connected to the move_base node. I need the planner to compute the shortest path that a human would probably follow to his destination. So I do not want to design a path for the robot but for the human.

I was wondering if there is a standalone node of such a planner that simply computes a path from start to end given the costmap.

Thank you

edit retag flag offensive close merge delete


Take a look at using Open Motion Planning Library http://ompl.kavrakilab.org/ directly, not sure about existing standalone nodes (subscribe to goal and current position, publish plan, or service version of same).

lucasw gravatar image lucasw  ( 2017-07-06 08:01:19 -0500 )edit

@dmkarage Have you got the answer? I am also looking for the answer.Can you advice me, please?

Heba gravatar image Heba  ( 2019-04-07 01:24:48 -0500 )edit

@Heba@dmkarage I am also trying to design a path planning node without using move_base. Could you help?

alperenkeser gravatar image alperenkeser  ( 2023-02-10 03:05:29 -0500 )edit

The code in https://github.com/AtsushiSakai/Pytho... may have what you are looking for, and is designed to have minimal dependencies (it doesn't use ros for example, but it's easy to use the code there in a ros node)

lucasw gravatar image lucasw  ( 2023-02-14 10:00:47 -0500 )edit