The enclosed videos demonstrate an approach of integrating the optimality of two layers of autonomous vehicle trajectory design. We assume that some optimal control laws are available as a set of motion primitives (the lower layer) to address the vehicle dynamics. For the upper layer, the trajectories that achieve the task are determined solely through the primitives and do not reference the vehicle dynamics directly. We translate the task into a very special type of cost-to-go function, which is partially specified artificially and partially determined by an admissibility condition imposed by the set of primitives. The optimality feature of the primitives is formally extended to the final trajectory design. Four videos are enclosed. In two of them, the solutions were derived analytically in closed form. As a result, the designs require little computation for real-time implementations. The other two videos demonstrate learning based on our approach. For more details, please look for the upcoming paper of the authors in the Robotics and Autonomous Systems journal.