In this project, you are provided with a quadrotor simulator written in Matlab. The simulator implements dynamics model of quadrotor and relies on the numerical solver ode45. You will need to implement controller.
In this project, you need to implement a trajectory generator. A carefully designed path planner can enable the quadrotor to operate aggressively and precisely. You need to find a natural way to command the quadrotor is to set waypoints that it needs to pass by.
In this project, two simple 2D grid maps which contain the obstacles, start and target locations will be provided. You need to implement a 2D A* path search method to find a optimal path with safty guarantee. Then you need to connect all the path points using your previous trajectory generator and control your drone to follow the trajectory.
In this project, you will be divided into 10 groups and every group will be provided with a quadrotor, which is equipped with a DJI N3 autopilot and a NVIDIA TX2 mini PC. Every group needs to implement your work on the NVIDIA TX2 and achieves your quadrotor with the ability of hovering and following a designed path stably.
In this project, you need to implement the algorithm on Lecture 7 to estimate camera’s poses with images. This is an individual project, which means you must complete it by yourself.
In this project, you need to implement the extender Kalman filter on your Lecture note 9. This is an individual project, which means you must complete it by yourself.
In this project, you need to integrate all the works you have done before onboard the drone. Which means you need to run the tag_detect-based localization and EKF on the TX2 to online localize itself. And use the controller to track a pre-defined trajectory.