The Goals

For Unified Robotics IV: Navigation, I worked with two other students on developing 2D Simultaneous Localization and Mapping (SLAM) on a Grid-Based Map with TurtleBot, a robot with an open source platform. The system is built using the Robot Operating System (ROS). We utilized the Kalman filter for State Estimation and Monte Carlo for Localization. We programmed the robot to navigate an unknown map with a Frontier-based Exploration. We also developed calculation for deriving the Configuration Space and experimented with different optimization algorithm, for example A* path planning algorithm, to construct a path from one point to another. Using the GMapping package in ROS, we programmed the robot to explore an unknown area, create a 2D map, navigate back to the starting point with the smoothest and shortest path, and move to a given point in the map.

What I did

• Achieved autonomous map exploration and navigation using frontier-based exploration and SLAM GMapping algorithm in ROS.
• Implemented A* path planning algorithm and configuration space calculation for obstacle avoidance using Python.
• Optimized mapping and navigation performance by streamlining ROS node communication and implementing algorithms that minimize frequency of turning.