This class provides an introduction to the fundamentals of robotics, covering topics such as kinematic models, motion planning, control systems, and robot perception.

The class had us work on a skid-steer robot platform, implementing various algorithms to enable autonomous navigation using the onboard camera and IMU sensors. The environment used for development and testing was ROS2 (Robot Operating System) running on Ubuntu.

Unfortunately, due to the class being largely graded on observations and footage, I cannot share much of the work publicly. However, the main projects included:

1. Implement and run a kinematics-based waypoint–following controller on the Rover/Rubik PI using a provided waypoint file.
2. Turn an open-loop controller into a vision-based closed-loop controller using landmarks to improve localization while driving a specified waypoint path.
3. Implement an EKF-based SLAM system that builds a landmark map and robot trajectory in a 10×10 ft environment from an unknown start pose.
4. In a landmarked 10×10 ft environment with a central obstacle, design and justify two path planners where one prioritizes safety and one minimizes time/distance.
5. Perform environment mapping using a SLAM approach in a random obstacle field.