Optimization-Based Collision Avoidance - a path planner for autonomous parking

Paper describing the theory can be found here.

OBCA is a novel method for formulating collision avoidance constraints. It provides a smooth reformulation of collision avoidance constraints, allowing the use of generic non-linear optimization solvers.

OBCA can be used to in path planning algorithms to generate high-quality paths that satisfy the system dynamics as well as satefy constraints. We provide Julia-based implementations for a quadcopter navigation problem and for autonomous parking problems.