Stick/slip is a severe type of torsional drillstring oscillation that affects the efficiency of the drilling process and can cause bit damage as well as drillstring failure. Different approaches to mitigate stick/slip oscillations are used in the field, ranging from procedures for optimizing drilling parameters, to passive downhole damping devices, and active surface controllers. To improve performance, some of these systems use mathematical models of the rotary system. Since the drilling process is subject to permanent changes, model-based control systems can only perform adequately when constantly updated with the actual conditions. To date, surface data has been used for this, but some key drilling process characteristics can only be identified with downhole data.This paper presents an innovative model-based stick/slip control system. It is based on a model of the drilling process that predicts the intensity of downhole vibration with respect to drilling parameters. Using downhole measurements, the model is adapted to the actual drilling process and employed to find optimal drilling parameters. To overcome the limited bandwidth of the typically used mud-pulse telemetry systems, models are identified downhole in the dynamics measurement tool.The paper presents field tests performed to evaluate the stick/slip control system and the investigation of how model parameters can be identified from measurements as well as how well the identified models represent the drilling process. Results show that this model can be used to predict the intensity of torsional oscillations and determine optimal drilling parameter values.The presented stick/slip control system can provide quantitative recommendations on changing drilling control parameters to mitigate stick/slip and can be used in an automated mode by directly connecting to the rig control system for monitoring drilling conditions and adjusting drilling parameters. Proper handling of downhole vibrations can significantly increase reliability.