Closed BoSmallEar closed 5 years ago
Sorry, I just saw there is a tutotial-2 for this issue. I will re-open if needed.
Actually I want to modify the path-planning part of the rbprm module and keep other parts in the pipeline unchanged, however this module is very complicated and it run several servers. Therefore how should I modify this module to add another path planner?
There are several steps to follow.
ps.loadPlugin("plugin.so")
to your script.For the last step, you have to be more precise. What step of the RBPRM planner do you want to modify ? the algo for planning a trajectory of the root, the algo for planning a trajectory of the root or any other ?
I want to change the algo for planning a trajectory of the root(the path used for contact generation), I think the server side of this module seems to be very different from the original corbaserver.
Then it should be sufficient to create a planner class and use selectPathPlanner. For testing purpose, you can do it in the source file containing the main function, in hpp-rbprm-corba.
The base class Path-planner will run oneStep until a solution is found. Does that mean the planner will stop when the start node and the goal node are connected and I have to implement the planner with a explicit roadmap?
Method oneStep fills the roadmap with configurations and paths. It is responsible neither for checking whether the problem is solved nor for computing the solution path.
I have to implement the planner with a explicit roadmap?
I don't understand what you mean.
I hope you solved your problem. I close this because it has been inactive for some time. @BoSmallEar re-open if necessary.
Hi, I want to use my own path planner (based on A*) with the HPP framework. I'm a little confused about how to do so. Do I have to inherit from the hpp::core::PathPlanner class and override the methods? Do I need to make any modifications on the corba server side? I noticed that there is a void hpp::corbaserver::Problem::selectPathPlanner method for the client.