A Context-Aware Driving Assistance system links drivers to the environment in which they drive, both from the perspective of the relationship between the human and his/her car, as well as from the perspective of the traffic situation. Existing ADAS solutions each focus on one only aspect of driving.

The purpose of a context-aware ADAS system is to provide an overall understanding of the current traffic conditions (at a semantic level) (for example, a driver behind another one wants to overtake under light rain conditions and in a broad-bend road). A context-aware robotic personal assistance system uses and reasons using various sensors mounted on the robot and in enclosures to provide personalized, efficient robotic services with greater robustness than could be provided by the stand-alone robot. In robotic assistance scenarios for people with special needs as well as in cases of assistance in public spaces, the robots used must always be aware of the situation they are in, the environment in which they operate, and especially the users whom they serve.

In context-aware services, an essential issue is to determine what kind of information becomes context information for an application, and how it can be used in conjunction with other information.


The purpose of the ROBIN-CONTEXT project is to create a support platform for semantic definition / representation and easy and efficient management of data that becomes context in personalized robotic assistance scenarios and ADAS systems. The platform will define a well-defined flow of context data (from their acquisition, to dissemination / consumption) and will provide support libraries for inferring situations with high semantic level by combining knowledge-based and data-based techniques.

The specific objectives of the project are:

  • The definition of a data representation format that is expressive and can be easily used in semantic inference, in event processing systems, and in inference models based on machine learning.
  • The design and implementation of the support platform for implementing the acquisition – inference – dissemination flow needed in the processing and management of context data for centralized applications such as context-aware ADAS.
  • The design and implementation of the support platform for the realization of the acquisition – inference – dissemination flow needed in the processing and management of context data consumed as web services.
  • The development of component for semantic processing of data to define the detection constraints in some situations.
  • The development of a component for making inferences in the form of rapid event processing.
  • The development of a component for inferring context using libraries of algorithms based on probabilistic graphical models.
  • The design, implementation and testing of tools and services for economic agents.

Originality and innovation:

The originality of the proposed solution in the ROBIN-Context project consists of the creation of an extensive and modular framework for processing context, which can be easily adapted to the needs of a wide range of possible applications. The proposed platform is designed in such a way that its components can be used independently of each other (for example the different inference engines) and within the architecture where they will support the whole cycle of data acquisition – processing – dissemination of context information for projects in ROBIN.