Your browser doesn't support javascript. This means that the content or functionality of our website will be limited or unavailable. If you need more information about Vinnova, please contact us.

Improved Impact of Collision Avoidance by Steering Technology on Real Life Safety

Reference number
Coordinator Zenuity AB
Funding from Vinnova SEK 5 200 000
Project duration January 2015 - December 2019
Status Completed
Venture Traffic safety and automated vehicles -FFI
End-of-project report 2014-05621.pdf(pdf, 487 kB) (In Swedish)

Purpose and goal

The overall objectives with this project is to identify methods for significantly higher safety benefits of active safety systems. Even if most of the important accident types are adressed by todays system, they still suffer from significant limitations. In specific, methods for improving the accuracy of prediction of vehicles future paths, are sought, as well as an evaluation of these methods effects on the overall safety impact of the relevant systems.

Expected results and effects

From a thorough review study of existing strategies carried out within the project, it was concluded that to significantly improve the benefit from collision avoidance systems relying on automatic steering, the predictions of the vehicles future paths need to be considerably improved. Two different methods have been studied and applied. Promising results when applying artificial neural networks to predict possible lane departures, have been achieved. Compared to today’s state-of the art methods, the performance was significantly improved.

Planned approach and implementation

Initially, an analysis of limitations and underlaying causations of today’s system, as well as a thorough review of all existing methods and their pros and cons, were carried out. Based on this analysis, it was primarily studied how to best use a neural network, where access to large data sets containing real world driving was an important precondition. The methods have been implemented and evaluated in simulation but with real data as input, making the results highly relevant.

The project description has been provided by the project members themselves and the text has not been looked at by our editors.

Last updated 22 April 2020

Reference number 2014-05621

Page statistics