PREPER dataset available for public use

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.

The PREPER dataset that is now publicly available strives to fill this gap by providing data of 160 pre-crash sequences involving two passenger cars in rural and city scenarios recorded at the AstaZero track in December 2020 and January 2021. It comes with a set of collision warning indicators for evaluating the object detection performance of AI perception systems with respect to safety requirements. The goal of this unique dataset is to support researchers to have easy access and use of validation 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 by the project “Safety-driven data labelling platform to enable safe and responsible AI” funded by the FFI Road Safety and Automated Vehicles programme. The dataset can be downloaded from https://github.com/AsymptoticAI/PREPER. For details please contact jorg.bakker@16.171.23.164.

A second dataset containing public road data is planned to be published subsequently complementing the dataset released today. For further information please send an e-mail to yinan.yu@16.171.23.164 or samuel.scheidegger@16.171.23.164.

Introducing joint venture AIXIA

We are excited to announce a new evolutionary step for our company. The CGIT and Asymptotic AI joint venture AIXIA will offer machine learning operation solutions to companies and organizations working with AI.

The first tools that will be offered through AIXIA are AiQu and SnapXS.

Aiqu is an advanced scheduler specifically developed for AI workloads that will drastically reduce complexity when planning and using valuable compute resources. This will make daily a lot easier for deep learning teams and IT organizations by reducing reducing complexity and at the same time increase efficiency.

SnapXS is a data management software suite that enables efficient use of big data in AI ecosystems. It provides easy access to petabytes of data and its annotations. It enables easy search, filtering and data sharing with only a few lines of code. SnapXS will significantly increase the data related return of investment by maximizing the value of data, reducing the development cost and shortening the time to production.

Together, SnapXS and AiQu offers a turnkey solution to give you a head start in your AI development.

We are very excited to be the first Swedish company to offer a complete and proven AI infrastructure solution and machine learning operations platform both offered as a service or on-premise.

Contact us for more information.

Check out our video introduction of AiQu and SnapXS

Asymptotic AI joins IGLAD accident database consortium to help strengthen automotive safety

We are now part of the accident data project IGLAD (Initiative for the Global harmonisation of Accident Data) where international safety experts from industry, government and research have build up an unparalleled database of global in-depth real world accident data within the last decade. Asymptotic is proud to be part of this unique source of safety expertise helping strengthen the development of safe automatic driving functionality based on real world data.

Asymptotic joins SAFER to contribute to a sustainable and safe transport system

We are excited to join SAFER – Vehicle and Traffic Safety Centre at Chalmers to contribute to a sustainable and safe transport system. We will aid SAFER in their vision towards zero accidents by providing our tools for data quality checking, robustness control, edge case analysis and anomaly detection available in the collaborative research projects at SAFER.

Asymptotic aims to discuss and help applying machine learning to research projects in the SAFER consortium and they plan to disseminate and utilize SAFER’s research by bringing knowledge to real-world use.

MALIN LEVIN, SAFER

Innovative Startups project: Fast and scalable pipeline to enable AI with big data

We are proud to have finalized our project funded by the Vinnova Innovative Startups initiative.

Thanks to the support from Vinnova, we have achieved our goals of enriching the technical functionalities of our AI platform and expanding our network. As a result, we have implemented two key components using the state-of-the-art deep learning technology and established several strategic partnerships with complementary expertise. We are looking forward to our journey ahead together with our partners!

The full public report can be found at Vinnova’s webpage.