Applications of moving windows technique to autonomous vehicle navigation

A ‘Moving Window’ scheme for detecting lanes, obstacles and corridor environments from the images captured by a CCD camera in an automobile or mobile robot is proposed. Processing the input dynamic images in real time requires high performance hardware as well as efficient software. In order to relieve these requirements for detecting the useful information from the images in real time, a ‘Moving Window’ scheme is proposed. For each image frame, the ‘Moving Window’ is newly defined and it is moved in a certain direction that is predicted by the Kalman filtering technique. By detecting the useful information, it becomes possible to search the obstacles within the driving lane of an automobile or mobile robot. The obstacle can be verified through the correlation between the stochastic characteristics of the suspected obstacle and the actual obstacle in a database. The feasibility of the proposed algorithm is demonstrated through the simulated experiments of the freeway and the corridor driving.