Enhancing the Automotive E/E Architecture Utilising Container-Based Electronic Control Units

Date

2021-01

Advisors

Journal Title

Journal ISSN

ISSN

DOI

Volume Title

Publisher

De Montfort University

Type

Thesis or dissertation

Peer reviewed

Abstract

Over the past 40 years, with the advent of computing technology and embedded systems, such as Electronic Control Units (ECUs), cars have moved from solely mechanical control to predominantly digital control. Whilst improvements have been realised in terms of passenger safety and vehicle efficiency, there are several issues currently facing the automotive industry as a result of the rising number of ECUs. These include greater demands placed on power, increased vehicle weight, complexities of hardware and software, dependency on software, software life expectancy, ad-hoc methods concerning automotive software updates, and rising costs for the vehicle manufacturer and consumer. As the modern-day motor car enters the autonomous age, these issues are predicted to increase because there will be an even greater reliance on computing hardware and software technology to support these new driving functions.

To address the issues highlighted above, a number of solutions that aid hardware consolidation and promote software reusability have been proposed. However, these depend on bespoke embedded hardware and there remains a lack of clearly defined mechanisms through which to update ECU software. This research moves away from these current practices and identifies many similarities between the datacentre and the automotive Electronic and Electrical (E/E) architecture, demonstrating that virtualisation technologies, which have provided many benefits to the datacentre, can be replicated within an automotive context. Specifically, the research presents a comprehensive study of the Central Processor Unit (CPU) and memory resources required and consumed to support a container-based ECU automotive function. The research reveals that lightweight container virtualisation offers many advantages. A container-based ECU can promote consolidation and enhance the automotive E/E architecture through power, weight and cost savings, as well as enabling a robust mechanism to facilitate future software updates throughout the lifetime of a vehicle. Furthermore, this research demonstrates there are opportunities to adopt this new research methodology within both the automotive industry and industries that utilise embedded systems, more broadly.

Description

Keywords

Citation

Rights

Research Institute

Collections