Robust Obstacle Detection from Stereoscopic Image Sequences Using Kalman Filtering

In this paper a new approach for video based obstacle detection for a mobile robot is proposed, based on probabilistic evaluation of image data. Apart from the measurement data, also their uncertainties are taken into account. Evaluation is achieved using Kalman filter technique combining the results of video data processing and robot motion data. Obstacle detection is realised by computing obstacle probability and subsequent application of a threshold operator. The first experiments show remarkably stable obstacle detection.