Vehicle Localization Kalman Filtering for Traffic Light Advisor Application in Urban Scenarios.

Area of Science:

• Intelligent Transportation Systems (ITS)
• Automotive Engineering
• Control Systems

Background:

• Advanced Driver Assist Systems (ADASs) offer significant potential for traffic management, safety, and environmental benefits.
• Accurate vehicle localization is critical for the effective operation of systems like the Traffic Light Advisor (TLA).
• Global Positioning System (GPS) limitations in urban environments necessitate robust localization solutions.

Purpose of the Study:

• To propose a novel vehicle localization technique using Kalman filtering for enhanced Traffic Light Advisor (TLA) system performance.
• To address the challenges of GPS signal loss in urban areas for precise vehicle positioning.
• To validate the TLA system's ability to optimize speed for traffic light passage, improving safety and reducing energy consumption.

Main Methods:

• A multi-rate sensor fusion approach employing Kalman filtering is proposed.
• The localization technique integrates map matching with a simple one-dimensional kinematic model.
• The method is designed to overcome Global Positioning System (GPS) limitations in urban scenarios.

Main Results:

• Experimental results show an estimation error below 0.5 meters on urban roads, even in areas with GPS signal loss.
• The proposed localization technique is suitable for Traffic Light Advisor (TLA) applications.
• Validation confirmed a 40% reduction in energy consumption compared to unassisted driving.

Conclusions:

• The developed Kalman filter-based localization technique provides accurate vehicle positioning essential for TLA systems.
• The system effectively mitigates GPS signal loss issues in urban environments.
• The Traffic Light Advisor (TLA) system demonstrates significant energy-saving benefits, enhancing driving efficiency and safety.