Characterization of Transient Properties of Fault-Tolerant ADAS Control Systems

Staff Mentor:

prof. J. Hellendoorn

Other Mentor(s):

Dr. Arturo Tejada Ruiz (TNO)


Transportation and infrastructure; Robust control



The Integrated Vehicle Safety (IVS) department at TNO develops technology for automated driving and cooperative mobility for intelligent vehicle applications. An important aspect in all such applications (e.g., cooperative vehicle platooning, autonomous urban vehicles) is the ability of the participating vehicles to safely withstand platform failures.

Platform failures (e.g., computer failures, sensors failures, network failures, etc.) can be handled by an array of off-the-shelf solutions. The most common solution is replication: critical functions are executed by parallel replicated components so that in case of one component failure, the other can take its place. However, switching components leads to transient behavior of the system, which may in turn lead to unsafe vehicle operation.

Assignment description:

The main goal of this assignment is to model and analyze the interplay between mechanisms for platform fault tolerance (hardware replication, software exception traps, rollback, analytic redundancy, etc.) and the performance and transient behavior of advanced driver assistance systems (ADAS) and autonomous vehicles. The analysis should lead to criteria that improves the joint design of ADAS and platform fault-tolerant systems.

Main Elements

• Modeling of the effect of platform fault tolerant mechanism on ADAS systems

• Analysis of ADAS transient behavior generated by fault-tolerant mechanisms

• Analysis of ADAS performance when deployed on fault-tolerant platforms

Added value for the project:

• Criteria for the joint design of ADAS and in-vehicle platform fault-tolerant systems

Added value for the intern:

• Know how real-life ADAS systems

• Working in a dynamic environment on automated driving functions

© Copyright Delft Center for Systems and Control, Delft University of Technology, 2017.