A viable autonomous passenger vehicle must be able to plot a precise and safe trajectory through busy traffic while observing the rules of the road and minimizing risk due to unexpected events such as sudden braking or swerving by another vehicle, or the incursion of a pedestrian or animal onto the road. The planner must be able to produce a plan within a small fixed time window. This project is working on a unified motion planning approach applicable to autonomous driving for highway and urban environments. Initial work adapted the state lattice framework pioneered for planetary rover navigation to the structured environment of public roadways. The main contribution of this work was a search space representation that allows the search algorithm to systematically and efficiently explore both spatial and temporal dimensions in real time. This allows the low-level trajectory planner to assume greater responsibility in planning to follow a leading vehicle, perform lane changes, and merge between other vehicles.
Our current work is on reducing computation through focused sampling and minimal replanning; and generation of smooth trajectories across the range of highway, urban, and evasive maneuver scenarios.