Design and simulation of mobile robot with slam and path planning algorithm

In this thesis, the simulation-based investigation of a Turtlebot mobile robot in an environment created in ROS using methods such as an Occupancy Grid SLAM (Simultaneous Localization and Mapping) and performed navigation using RRT* path planning. Also, the ability of an autonomous robot to navigate in conflict situations from one point to another in an unknown environment has been addressed while assuring collision avoidance. The navigation of the autonomous robot in an environment with random and live obstacles is one of the main challenges faced by mobile robot technology. Robotic navigation path planning is an axial problem because it is complex and time-consuming for the robot to select an ideal path. Before starting the navigation, the robot must achieve self-localization in an unknown environment known as SLAM. Simultaneous Localization and Mapping (SLAM) and path planning are the main factors that decide efficient path planning for an autonomous robot. SLAM problems can be solved using various methods but mainly rely on the sensors' accuracy, such as camera, LIDAR, etc. Even after successfully mapping the environment, navigating through the obstacles to find the optimum path to a goal location is challenging. In this simulation, a map is constructed using Gazebo software. Then, localization of the robot on the map is achieved using the LIDAR sensor data and using the Occupancy grid SLAM method while controlling the robot manually. The next step was to develop the ROS nodes needed for navigation using python language, which establishes the communication between the client and the navigation server. Different ROS nodes required to perform the navigation are initiated by creating the navigation launch file. The RRT* algorithm is implemented to find the global path from an initial location to a goal location. The navigation of the mobile robot is performed, and the data obtained during the navigation are stored in a ROS bag file. These data are then plotted to understand the robot's behavior under different conditions