Precise localization is essential for the operation of Connected and Automated Vehicles (CAVs) in urban scenarios. Camera and LiDAR-based solutions are currently used in some of the CAVs around the world, but they entail an expensive performance in terms of computational time and economic cost. Due to this, Global Navigation Satellite System (GNSS) based solutions remains the preferred solution for positioning Automated Vehicles in different Operational Design Domains (ODDs). Although several improvements in system reliability have been made recently, in case of positioning failures most of the architectures rely on system redundancy or automation disengagements to achieve minimal risk conditions. This paper presents a fail-safe decision architecture for mitigating positioning failures of automated vehicles in urban scenarios. The proposed architecture utilizes a combination of sensor fusion localization and decision-making modules to ensure the safe and efficient operation of the vehicle in the event of a positioning failure. The proposed approach is evaluated through simulation in a representative urban scenario and is shown to effectively handle positioning failures, improving the localization accuracy provided by each information source.

The SELFY project has received funding from the Horizon Europe programme under grant agreement No. 10106974