Event

The SAFER Thursday webinar week 45 features an Open Research project and a SAFER pre-study with speakers from Asymptotic and Chalmers. Welcome!
 

Safety dataset for training and validation of AI perception functions in AD and ADAS
Speaker: Jörg Bakker, Asymptotic AB

Deep learning algorithms are becoming the state-of-the-art technology used in the perception functions of AD and ADAS. With the safety assurance of level 5 AD systems the perception of the environment and especially dynamic objects becomes a central component. Data-driven AI algorithms are to a large extent determined by the training dataset and need to be tested and validated especially for safety critical edge cases which can be derived from real world scenarios that lead to accidents caused by human drivers. However, current available public data is limited to normal driving situations. In this project we strive for creating a dataset that is fully aware of safety requirements and develop an initial version of a public dataset of driving scenarios relevant for safety studies along with a set of key performance indicators (KPIs) to evaluate the perception system of AD/ADAS functions. The goal is to support researchers to have easy access and use of verification data while triggering research and development on safety aspects of AI perception systems and eventually help improve the robustness of AI perception systems used in AD/ADAS. 

The project is supported by the “Open Research Programme” by Chalmers, RISE, and SAFER in collaboration with AstaZero and also by the
project “Safety-driven data labelling platform to enable safe and responsible AI” funded by the FFI Road Safety and Automated Vehicles programme.

Looking for adipose tissue material model parameters
Speakers: Hosein Naseri, Ph.D., Håkan Johansson, Ph.D., Dept Mechanics and Maritime Sciences at Chalmers

An important restriction in the current Finite element human body models (FE-HBMs) is the material models implemented to represent the soft tissues. This is to some extent that for soft materials, biofidelity is sacrificed to accommodate computationally stable models. But also, because experimental data on tissue level is much lacking for the development of useful material models. The long-term aim of this work is to provide experimental data and updated constitutive models of human soft tissues when exposed to high strain rate loading. Especially noteworthy is the compressibility of soft tissues at different strain rates, as it has a huge influence on the mechanical response as well as computational stability.

In this talk, we present a review of experimental works on determining the mechanical properties of adipose tissue. Of concern is to what extent animal tissues can represent humans, as well as the influence of temperature. We also propose a design for a test setup that is under development in the context of a SAFER pre-study, with support from industrial partner Autoliv and academic partners in the SAFER community. 

Welcome!
 

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