Virtual CAN Lines in an Integrated MPSoC Architecture

The standard solution for automotive control networks is the Control Area Network (CAN) bus. Almost any vehicular computer system comprehends at least one CAN line. For the past two decades, software development for control system has been strongly connected to the properties and interfaces of the CAN bus. Currently, the automotive industry is in the middle of a technology leap towards an information-based industry. New technologies are getting ready to fulfill newly emerging requirements for innovative products such as hybrid engine control, intelligent energy management, and advanced driver assistance systems. Integrated Multi-Processor-on-a-Chips (MPSoCs) will be one part of the solution to provide an adequate computing infrastructure for these newly emerging systems. The established technologies like the CAN bus will have to be reconsidered. In this work, we propose a virtual CAN overlay that abstracts the communication interfaces of an MPSoC to provide the Application Programmer Interface (API) of CAN to programmers. The overlay provides the standard behavior of a CAN line and works transparently over chip boundaries. The major implications is that the programmers can continue their used software development approaches and tools when introducing a new computing infrastructure. The main benefit is that the productivity can be maintained during this critical phase. In summary, our solution helps to mitigate the effects from a technology shift to integrated MPSoCs. Our approach is fully compliant with new automotive software development approaches like AUTOSAR.

[1]  C.E. Lin,et al.  Reliability and stability survey on CAN-based avionics network for small aircraft , 2005, 24th Digital Avionics Systems Conference.

[2]  Henning Wallentowitz,et al.  Strategien in der Automobilindustrie : Technologietrends und Marktentwicklungen , 2009 .

[3]  Roman Obermaisser,et al.  Reuse of CAN-Based Legacy Applications in Time-Triggered Architectures , 2006, IEEE Transactions on Industrial Informatics.

[4]  Hermann Kopetz,et al.  Real-time systems , 2018, CSC '73.

[5]  Christian El Salloum,et al.  The ACROSS MPSoC -- A New Generation of Multi-core Processors Designed for Safety-Critical Embedded Systems , 2012, 2012 15th Euromicro Conference on Digital System Design.

[6]  Lars-Berno Fredriksson,et al.  CAN for Critical Embedded Automotive Networks , 2002, IEEE Micro.

[7]  Roman Obermaisser,et al.  The time-triggered System-on-a-Chip architecture , 2008, ISIE 2008.

[8]  Hermann Kopetz,et al.  A Comparison of CAN and TTP , 2000 .

[9]  Armin Wasicek,et al.  Enhancing security in CAN systems using a star coupling router , 2012, 7th IEEE International Symposium on Industrial Embedded Systems (SIES'12).

[10]  Roman Obermaisser,et al.  From a Federated to an Integrated Automotive Architecture , 2008 .

[11]  S Latham,et al.  A reference book of driving cycles for use in the measurement of road vehicle emissions , 2009 .

[12]  Roman Obermaisser,et al.  The time-triggered System-on-a-Chip architecture , 2008, 2008 IEEE International Symposium on Industrial Electronics.

[13]  R. Obermaisser,et al.  Fault containment in a reconfigurable Multi-Processor System-on-a-Chip , 2011, 2011 IEEE International Symposium on Industrial Electronics.

[14]  Manfred Broy,et al.  Engineering Automotive Software , 2007, Proceedings of the IEEE.