Research project to strengthen confidence in automated systems

Research project to strengthen confidence in automated systems

Feature articles |
By Christoph Hammerschmidt

Human drivers are usually trusted to make better decisions in road traffic than software. The consortium of the project “Product security for cross domain reliable dependable automated systems” (SECREDAS) has set itself the goal of strengthening trust in such networked automated systems. This is not only about physical security, but also about trust in data privacy. 69 partners from 16 European countries are participating in the project, including the Fraunhofer Institute for Experimental Software Engineering IESE.

In autonomously driving cars, neural networks play an increasingly important role in control and situation recognition. The difficulty here is that the way in which these neural networks make their decisions cannot always and never be fully understood. The researchers are therefore developing a safety supervisor that monitors the decisions of the neural network live, so that, if necessary, regulatory intervention can be carried out on the basis of these assessments. This supervisor is based on algorithms that make use of classical approaches. Using these, the researchers do not record the overall situation like the neural networks, but critical key points. In this context, the researchers’ first concern in the project is to identify suitable metrics; the introduction of suitable countermeasures to control the risk will then be the subject of further work.

The researchers explain how this is done exactly using the example of a real intersection situation: The neural network is designed to capture the overall situation: Which right-of-way rules apply, is the traffic light red or green, are pedestrians within the danger zone, do other cars cross the planned future roadway? Instead, the Safety Supervisor’s algorithms rely on specific metrics. These would be, for example, the “General-time-to-collision (GTTC)”, i.e. the time to a collision taking into account the expected trajectory, or the “Worst Case Impact Speed” metric for assessing the severity of damage based on the expected collision speed. If the car is heading towards another road user who should have escaped the neural network, the algorithms of the safety supervisor recognize that the distance is shrinking to a dangerous degree. They can take command and brake the car if the autonomous control fails. In a simulation, the researchers evaluated the suitability of these metrics for various dangerous situations. The result is positive for the researchers: “The approach of checking the neural networks at any time and live using classical approaches, together with dynamic risk management, can significantly increase safety,” says Mohammed Naveed Akram, scientist at Fraunhofer IESE.

If another driver has used the car, it is often necessary to adjust the seat and mirrors to suit, select one’s own favorite music, enter one’s personal favorite locations in the navigation system, and the like – only then can the car be set up. Although it is possible and practical to save such information so that all settings automatically fit. But not everyone would like to use it. Many shy away from it for reasons of data protection. It becomes even more tricky if the car also records medical data, such as blood sugar levels or heart rate – in order to issue a warning to the driver or call for help if necessary. This is because it is currently difficult for users to determine whether the data will remain in the car or be processed in a cloud. User confidence in a system perceived as opaque varied greatly from person to person.  “One-fits-it-all is therefore not a solution,” says Arghavan Hosseinzadeh da Silva, software developer at Fraunhofer IESE. “In general, the more data you release, the more service you get. However, how much data someone wants to share in which case varies greatly from person to person”.

The researchers are therefore developing a framework that can be used to restrict the use of all personal settings according to situation and preferences. One would like to have the Whatsapp messages displayed on the car display – unless one is not alone in the car? You want the same contacts and playlists to be displayed in the rental car as in your own vehicle and the seat, steering wheel and mirrors to be set to match? Health data, such as heart rate measurements, should remain in the car and not be sent to a cloud – unless urgent help is needed, which should then be called for automatically, for example after an accident? In the future, users will be able to set such things themselves via an app, and these privacy specifications will be transferred via smartphone to every vehicle that the user is currently using, whether it is a company car, rental car or private car.

The necessary framework components will be integrated into the car. A “Policy Decision Point PDP” will check whether the transfer of data is permitted. Within the framework of SECREDAS, the IESE researchers intend to develop a prototype for the framework, which is expected to be ready by the end of 2020. In the long term, the SECREDAS consortium would like to establish a standard for data usage control in cars, which should be adopted by all car manufacturers if possible, in order to enable the informational self-determination of car users.


More information:

Fraunhofer IESE

Secredas project homepage


Related articles:

Automotive industry outlook: Software passes hardware on the fast lane

Can we trust our cars?

Security and Connected Cars


If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :    eeNews on Google News


Linked Articles