OBCA

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.

Short Description

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.

Examples

OBCA for Quadcopter Navigation

OBCA for Autonomous Parking

Backwards Parking

Parallel Parking

Parking of Truck with Trailer