In this survey paper some of contemporary research areas end relevant results in robotics are briefly presented. The first research area presented is the approach based on neural networks for learning and improving dynamic models of robots. The second area deals with the possibility of the application of fuzzy logic in robotics. As the third, very topical and widely investigated field in robotics, the problem of position/force control of robots is presented, or more broadly the problem of robotic tasks for robots working in contact with a dynamic environment. The last part contains active systems and constructions as well as problems related to controlling their dynamic performance.