Describir: Velocity Planning via Model-Based Reinforcement Learning: Demonstrating Results on PILCO for One-Dimensional Linear Motion with Bounded Acceleration