Optimization-Based Collision Avoidance - a path planner for autonomous navigation
Paper describing the theory can be found here.
Note: An OBCA version specialized towards autonomous parking can be found at H-OBCA.
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.