drone_planning

Project for Motion Planning Methods and Algorithms course.

The goal of the project is to compare 3 different methods of motion planning for drone. The environment in which drone moves is represented by Octomap.

Tasks of project:

• prepare octomap environment
• prepare drone (mesh)
• launch chosen methods of motion planning for defined problem
• visualize planned path and motion of the drone

Dependencies:

• libccd https://github.com/danfis/libccd
• fcl https://github.com/flexible-collision-library/fcl
• octomap https://github.com/OctoMap/octomap
• MoveIt! https://moveit.ros.org/install/

Bag file with octomap environment used in project:

• bag file: https://drive.google.com/open?id=11DyT1_8go2_tEr34aZ9rSi7Hpfpf1Mm4

Launch with:

• play bag file with octomap in loop mode
• $ roslaunch drone_planning drone_planning.launch

Choosing planner:

• make sure that library of planner you want to use is imported.
• change type of planner declaration in drone_planning.hpp.

std::shared_ptr <ompl::geometric::LazyPRMstar> planner;

• change type of planner definition in drone_planning.cpp in configure() function

planner.reset(new ompl::geometric::LazyPRMstar(si));

• change planner clearing function in planPath() depending if you use multi-query planner or single-query planner

// if you use multi-query planner uncomment line below and comment line with clear() method
planner->clearQuery();
// if you use single-query planner uncomment line below
planner->clear();