Tecnologia e Inovação personalizada para condução segura

Distraction and drowsiness monitoring technologies are already a reality in innovative vehicles. Most of the current vehicle technology is based only on driving performance variables, such as lane deviance, steering wheel angle, or time cumulative of the journey. Other systems are being developed based on biometric data collected by non-intrusive equipment, which is the case of the WingDriver system. Despite many of these detection and warning systems being shown to measure driver alertness levels with a high degree of confidence, they make no distinction as to the driver profile and states, leading to a high number of false and unnecessary alarms that impact driver’s compliance (Naujoks et al., 2016). The mobility sector demonstrates continuous and growing dynamism with various technological and business model innovations, making it one of the most significant growth markets and highly competitive. This latter leads to the delivery of products for the market that are not entirely validated, and it is also difficult to embrace specific conditions such as diseases or population groups. Indeed, the European Union general safety regulations for motor vehicles (2019/2144) require that all new vehicles of category M (cars and buses) and N (trucks) be equipped with driver drowsiness and attention warning systems. From 2023 onwards, direct driver monitoring will be required for a vehicle model to get a full score in the Euro NCAP rating. The project TEC4Safe aims to contribute with novel knowledge on driver fatigue to support innovative technology towards an accurate driver monitoring system tailored to the driver characteristics. The specific objectives of the project are: O1: Understand the accident risk of OSA disease on patients as well as on undiagnosed individuals, comparing with healthy individuals; O2: Understand the accident risk of professional drivers and youngers, both representing a population group with specific characteristics associated with higher accident risk, for distinct reasons; O3: Develop models in the view of fatigue detection and prediction, considering the different characteristics of the population groups selected and thus, outperforming the existing ones; O4: Analyse the effect of countermeasures on fatigue of the distinct population groups; O5: Evaluate and validate the alarm setting to prevent accidents accurately; O6: Analyze real-world data to a relative comparison of both environments; O7: Provide data and knowledge for commercial driver monitoring, supporting the system validation and accuracy. The project TEC4Safe goes beyond the state of the art through embracing distinct driver profiles and focusing on OSA disease not only on patients but also on undiagnosed cases. These project aspects will allow the development of tailored models to detect, predict, and warn the driver based on data quality, i.e., accuracy, validity, completeness, and consistency, ensuring data meets the system-specific needs and is trustworthy for analysis. A data-driven approach involving techniques like statistical analysis and machine learning to uncover trends, correlations, and anomalies that support models' development. The novel tailored framework for models’ development will be submitted for a patent. To achieve this ambitious goal, the project TEC4Safe joins distinct disciplines, which are transport engineering, informatics, medicine, and data science.