One of the main challenges in ubiquitous networks is localization. Many positioning methods have been proposed in the literature since this data is critical for numerous protocols and services. Choosing the right localization algorithm for a specific environment is a hard task. However, it is even more complex when several radio access technologies are available in evolving environments. This paper presents a framework based on autonomic networking concepts to achieve localization in ubiquitous networks. This approach uses ambient intelligence to provide flexibility, accuracy, and energy savings.