OBSTACLE AVOIDANCE FOR TELEOPERATED ROBOTS FOR LIVE POWER LINES MAINTENANCE, USING ARTIFICIAL VISION. R. Aracil*, F. M. Sánchez*, D. García*, J.M. González*,L.M. Jiménez**, J.M.Sebastián*. * DISAM, Universidad Politécnica de Madrid, MADRID, Spain ** DISC, Universidad Politécnica de Alicante. ALICANTE, Spain Abstract: Nowadays uninterrupted power supply has become a must for electrical companies, due to the increasing technological advances, present in our society, that lead to higher demands of electricity. This strong need is obliging the electrical companies to develop processes and techniques for outage-free maintenance of electrical power supply, whose major application lies in aerial distribution lines. In these conventional techniques workers have to do their job on a live electrical power line indirectly, with various kinds of hot-sticks or directly touching the line with rubber-insulated gloves. Therefore, work is performed in a hazardous environment with both, the risk of electric shock and the danger of falling from a high place. In addition, workers have to be very skilled and work cooperatively under very demanding tasks. For all these reasons, the European Community foresees the suppression of this operation in a short term. This paper presents a teleoperated system, developed at DISAM*, capable of performing electrical power lines maintenance and inspection, tasks. The general features of the project are explained, paying special attention to the artificial vision system, which is an obstacle avoidance system capable of preventing collisions between robots and the electric cable. Such a system is the result of several years of research on computer vision. Keywords: Obstacle Avoidance, Teleoperation, Artificial Vision, Stereo Vision, Calibration, Path Planning.