Smart wheelchair guidance using optical flow

This paper proposes a method that uses optical flow information to help guide a powered wheelchair along the centre of corridors and through the centre of doorways. The Shi and Tomasi algorithm is used to find good points to track, then the optical flow vectors are calculated using the pyramidal Lucas-Kanade method and finally the optical flow vectors are translated into depth information and averaged. The results show that the proposed algorithm runs in real-time at a high level of accuracy enabling a wheelchair enough time to avoid a collision.