FAITH: Fast Iterative Half-Plane Focus of Expansion Estimation Using Optic Flow

Course estimation is a key component for the development of autonomous navigation systems robots. While state-of-the-art methods widely use visual-based algorithms, it worth noting that most fail to deal with complexity real world. They often require obstacles be highly textured improve overall performance, particularly when obstacle located within focus expansion (FOE) where optic flow (OF) almost null. This study proposes FAst ITerative Half-plane (FAITH) method determine course micro air vehicle (MAV). achieved by means an event-based camera, along fast RANSAC-based algorithm uses OF FOE. The performance validated benchmark on simulated environment and then tested dataset collected indoor avoidance. Our results show computational efficiency outperforms while keeping high level accuracy. has been further demonstrated onboard MAV equipped thus showing FAITH can online suitable avoidance MAVs.