In this paper an implementation of a smart predictive monitoring and adaptive control system for the public lighting have been carried out. The vehicular traffic flow acquired using a smart camera has been analyzed and several predictive methods have been studied. Then, a control strategy based on the given traffic forecasts and on the dynamical street class downgrade allowed by the law, has been implemented. Experimental results provided by a real life testbed showed that the proposed strategy has high potential energy savings without affecting safety. © 2017 The Authors.