A path planner for autonomous driving on highways using a human mimicry approach with Binary Decision Diagrams

This paper considers the problem of path planning for autonomous ground vehicles on highways with regular traffic. The goal is to select a desired trajectory from a set of parameterized candidate trajectories such that some criterion is optimized. This selection is subject to avoiding collisions, respecting the traffic rules, and eliciting smooth behavior for passenger comfort. The desired trajectory is fed as a state reference to a path follower which uses Model Predictive Control (MPC) to compute the steering and braking actions. This work proposes a human mimicry approach for the path planning problem by considering a Binary Decision Diagram (BDD). The binary decision diagram applies a tree search algorithm coupled with a truth table decision process. Simulation and experimental results are presented to verify the real-time feasibility and demonstrate the effectiveness of the proposed algorithm.