HybridMotionPlanning

Overview

In this project, we implemented Informed RRT* in C++ and Python to find a path on a map in Gazebo. The map was mapped using laser scan and a cost map was generated to perform planning. A turtlebot was used to move from the current location to the goal location. Custom plugins were made for local and global planners.

---

Results

Map used for planning

Paths found by Informed RRT

---

Tools and Frameworks

ROS Version: Noetic

Gazebo Version: 11.9.1

IDE: VS Code
(Any IDE of youe choice)

---

Building

Install point-cloud to laser scan package through ros noetic to run gmapping using ros

Install ros-gmapping

rosrun gmapping slam_gmapping scan:=/laser_scan _base_frame:=ego_vehicle _map_update_interval:=0.5

Install Turtlebot3 packages

sudo apt-get install ros-noetic-turtlebot3
sudo apt-get install ros-noetic-turtlebot3-gazebo
sudo apt-get install ros-noetic-base-local-planner
sudo apt-get install ros-noetic-nav-core
sudo apt-get install ros-noetic-dwa-local-planner

---

Usage

Launch the world using

roslaunch hybrid_planner_sim  turtlebot3_world.launch

To see an example of planning:

rostopic pub /move_base_simple/goal geometry_msgs/PoseStamped "header:
  seq: 0
  stamp:
    secs: 0
    nsecs: 0
  frame_id: 'base_footprint'
pose:
  position:
    x: 1
    y: 0
    z: 0.0
  orientation:
    x: 0.0
    y: 0.0
    z: 0.0
    w: 1" 
    

---