RANSAC

An iterative algorithm (Fischler and Bolles, 1981) for fitting a mathematical model to data that contains a significant proportion of outliers. In accessibility-focused indoor navigation systems, RANSAC is commonly used to detect the floor plane from a LiDAR point cloud — points whose normal vectors align with the gravity direction and fall within a small height threshold of the detected plane are classified as floor, and everything else as obstacle or wall. This robust floor-plane detection is a prerequisite for building an occupancy grid map that a blind traveller's smartphone can use for obstacle-avoiding path planning.