In the past few years the need for more flexibility in industrial production has implied, in the field of industrial robotics, a growing attention towards the possibility of making humans work directly in touch with robots. While totally out of reach just a few years ago, this scenario still cannot be completely achieved due to the presence of strict regulations about safety of human workers in an automated industrial environment. The final goal of this research project is to conceive, design and prototype a safety control system not only able to allow a fruitful human-robot interaction, such that it can be also certified with respect to actual and upcoming safety norms and regulations. In particular, the first goal is to enhance the perception capabilities of the typical control system of an industrial robot by integrating information coming from different kind of exteroceptive sensors and especially from vision sensors, like for instance common surveillance cameras and Microsoft Kinect. On the other hand, the project aims at developing new planning and control strategies that use the sensed information to allow a more structured human-robot interaction, while respecting as much as possible the standards and criteria established by safety norms and regulations.