Event

The theme for the SAFER Thursday webinar week 43 is pre-studies! This session features results from two SAFER pre-studies with speakers from Halmstad University and Skövde University.  
_____________________________________________________________________________

Cooperative Automated Driving Use Cases for V2X Communication
Speaker: Stephanie Alvarez Fernandez, Ph.D. Electrical Engineering, Research Engineer at Halmstad University

This pre-study aims to answer the following central research question: What are the latency requirements for collective perception systems to enable Advanced Driver Assistance Systems function to fulfil the state-of-the-art safety norms in a realistic traffic use case? With the goal of solving the above question, the pre-study focusses on establishing a tool chain data-scenarios-simulation-requirements that help us to understand the needed pieces/shortcomings of the data and tools. Such a study would enable to set the requirement on latency of Vehicle-to-Everything communication systems based on safety requirement of Advanced Driver Assistance Systems functions. Halmstad University is responsible for the pre-study and the co-production involve two parties, namely Viscando AB and Zenseact AB, that are interested on the achievement of the present pre-study.

Multimodal Data to Support Safe Driving Patterns 
Speaker: Yacine Atif, Professor of Informatics, Skövde University

In order to safely operate their vehicle, drivers must analyze and integrate  large amounts of heterogeneous information. This continuous process provides drivers with current and forthcoming conditions that influence their driving intentions. Advanced Driver Assistance Systems shipped with today’s vehicle offload drivers from this cognitive burden and pave the way to automated decisions and autonomous driving actions. These systems are further expected to forecast unsafe conditions and correct driver intentions to prevent collisions. However, care must be taken to ensure the driver is provided with the appropriate amount of information to make effective decisions. Safety scores condense current situation awareness and future driver intention to associate driver manoeuvres with risk assessment. 

We provide a context-aware and intention-aware approach to safety score that correlates driver-error in knowledge, performance and intent. Context corroborating to driver error is derived from descriptive analytics of driver state and driving behaviour factors, whereas driver intentions use predictive analytics to categorise intended manoeuvres. Subsequently, the inferred safety score could be used to guide assistive measures that reduce driver-error margins, accordingly.

Welcome!
_____________________________________________________________________________________

This seminar is only open for SAFER partners. Missing your invitation? Please contact Mikael von Redlich.