Social-Aware Driver Assistance Systems for City Traffic in Shared Spaces

Area of Science:

• Urban planning and intelligent transportation systems.
• Human-computer interaction and multi-agent systems.
• Computer vision and predictive modeling for autonomous driving.

Background:

• Shared urban spaces with mixed mobility (pedestrians, cyclists, vehicles) pose challenges for current automated traffic systems.
• Traditional traffic models lack sufficient data, relying only on physical parameters like position and speed.
• Understanding human behavior and social context is crucial for accurate environmental modeling and prediction.

Purpose of the Study:

• To introduce the Social-Aware Driver Assistance System (SADAS) for integrating social information into traffic management.
• To develop a domain-specific modeling language for describing and processing social contexts in traffic.
• To enhance the predictive and explanatory capabilities of automated systems in complex urban environments.

Main Methods:

• Development of a domain-specific modeling language for social contexts and their dynamics.
• Formalization of relevant 'traffic social properties' from literature within the modeling language.
• Implementation of a multi-agent system architecture to manage social specifications and processing.
• Utilizing subscription-notification mechanisms for SADAS integration with existing traffic systems.

Main Results:

• Demonstrated the application of social knowledge for predicting pedestrian and vehicle movements in shared spaces.
• Showcased the system's ability to improve obstacle detection, tracking, and traffic signal management.
• Validated the SADAS approach through a case study in a complex urban traffic scenario.

Conclusions:

• Social factors significantly influence movement and interaction in shared urban spaces, necessitating their inclusion in automated systems.
• The SADAS approach provides a robust framework for integrating social context into traffic management, enhancing prediction and safety.
• Future intelligent transportation systems must incorporate social-aware modeling for effective operation in diverse urban environments.