This chapter discusses the control aspects of continuous crystallization processes. Common control objectives for continuous crystallization are related to crystal product quality, process stabilization, economic performance, and environmental impact. Supersaturation is often used as controlled variable to obtain desirable crystal quality attributes, although direct approaches with a crystal quality attribute as controlled variable have also been developed. Sensors to measure crystal quality attributes or supersaturation in situ are readily available, which makes the application of automated feedback control loops attractive. A mixed-suspension mixed-product-removal crystallizer has limited options for process actuation unless fines dissolution is employed. Novel plug-flow crystallizers allow for the adoption of different control strategies (e.g., controlled cooling profiles with seeding). Model-based controllers in combination with state observers can handle time-varying model uncertainty, input constraints, sensor and actuator faults and asynchronous measurements, whereas model-predictive control has the unique capability to enforce multiple process constraints and is most effective when dealing with complicated interactions between multiple inputs and outputs. State observers can also be used to design effective filters for actuator fault detection. The rapid development of dynamic process models, advanced analytical techniques and improved numerical methods are main drivers of the current trend towards model-based control strategies for continuous crystallization.