Duration: 20:31

97 views

2

Presented at SAAM Mobility 2021 (https://events.eclipse.org/2021/saam-mobility/)

Authors: Anand Prakash, Lukas Krawczyk and Carsten Wolff

Abstract: The computational demands of safety critical ADAS applications on autonomous vehicles have been ever-increasing. As a result, high performance computing nodes with varying operating frequencies, and inclusion of different sensors have been introduced, which has resulted in introduction of heterogeneous architectures with high complexity. This complexity has led to challenges in analysing a system’s timing behavior, such as determining the end-to-end response time of high-level functionality as well as a real-time application’s latency. The existing approaches in analysing the timing behavior of an ADAS application does not address the overall end-to-end response time. It is also worth mentioning that a practical timing verification of an ADAS in a real-life scenario is still an open issue.
This talk provides an overview of the existing RC car platforms along with its comparison to the proposed RC car demonstrator platform. It focuses on the model based architectural design of RC car demonstrator as well as the event-chain task model, which is coarsely based on Waters Challenge 2019. Further, it lays emphasis on deriving the timing parameter metrics from the generated trace file, which can be used to evaluate the system behavior against the timing requirements. Finally, the talk concludes with the importance on designing an ADAS with a model-based approach. This will enable a better, faster and modular analysis of a safety critical ADAS application.