Trajectory Optimization for Helicopter Unmanned Aerial Vehicles (UAVs)

This thesis explores the numerical methods and software development for optimal trajectories of a specific model of Helicopter Unmanned Aerial Vehicle (UAV) in an obstacle-rich environment. This particular model is adopted from the UAV Laboratory of the National University of Singapore who built and simulated flights for an X-Cell 60 small–scale UAV Helicopter. The code, which allowed the team to simulate flights, is a complex system of non-linear differential equations—15 state variables and four control variables—used to maneuver the state trajectories. This non-linear model is incorporated into a separate optimization algorithm code, which allows the user to set initial and final time conditions together with various constraints, and, using the same variable scheme, optimize a trajectory. The optimal trajectory is defined by using a cost function—the performance measure—and the system is subject to a set of constraints (such as mechanical limitations and physical three-dimensional obstacles). Simulations conclude that solutions are readily obtained; however, it is still very difficult to derive trajectories that are truly optimal, and our work calls for more future research in computational programs for optimal trajectory planning. All simulations in this thesis are modeled using the MATLAB program.