Embedded Systems Technology enables a new automotive engineering process that couples verification with design so that the concurrent development and testing of ECUs in the context of their subsystems, subsystems in the context of the overall control system, and the automotive vehicle in the context of traffic is seriously and systematically addressed - prior to physical realization.

GRAHAM HELLESTRAND
Chief Executive Officer

Home > Solutions > Automotive Systems > Verification > The Current State of Automotive Verification

The Current State of Automotive Verification

An Industry View of the State of Verification

“Current automotive systems are still largely engineered as a set of discrete subsystems with minimal interactivity. Current tools and processes do not support the inevitable growth of interconnectivity, interaction and optimization. Safe, reliable and predictable operation must be ensured as the complexity of the systems increases. Interactions between previously separate subsystems (e.g. brakes and powertrain) create coupled modes which must be assessed and managed with respect to safety, reliability and availability” 1

As is evident from the quotation above, the verification problem is not solved well in the current automotive engineering process. This is largely because the available technologies were not capable of fulfilling the required role in the process. Faults not discovered in the verification process are reported to be the cause of a number of recent highly publicized failures of expensive cars in the middle of highways.

The potential liability risks arising from the failure to adequately verify and calibrate a vehicle’s critical control systems - which govern stability and safety, and increasingly, the usurping of control from the driver - are enormous.

Historical Causes of Poor Verification

These complex, engineered automotive systems, which are predominantly driven in dense urban traffic, are perhaps, when part of traffic, the most complex systems-of-systems that humans have devised. They have the potential and history of being everyday dangerous – being involved in ~1.3 million deaths per year and between 20-50 million injuries.

As Stephens’ statement above indicates, current automotive systems:

  1. are produced piece-meal - largely as discrete subsystems (typically with a single ECU) - in heavily distributed supply chains
  2. predominantly use natural language requirements documents that are inherently ambiguous
  3. are not rigorously tested for correct and incorrect operation (function, timing and interference) across the boundaries of multiple communicating ECUs and subsystems, and
  4. have intellectual property owned by various entities in the supply chain that is not available in sufficient detail for the testing in (c), or for the final integration and safety testing and subsequent fault diagnosis.

An article on this website The Case for Automotive Model-Based Design, notes that as recently as 2011 both NASA and the US National Highway Transport and Safety Administration (NHTSA) have commented on the lack of rigor in automotive systems engineering.

EST Bridges the Stephen’s Gap

Embedded Systems Technology (EST) enables a new automotive engineering process that couples verification with design so that the concurrent development and testing of ECUs in the context of their subsystems, subsystems in the context of the overall control system, and the automotive vehicle in the context of traffic is seriously and systematically addressed – prior to physical realization.

The process is model-based and enabled by EST’s ESSE technologies and methodologies that include a specification notation, a scalable, high performance, multi-core, distributed simulation engine (DSE), and unique design, architecture and optimization methodologies.

EST’s technologies and methodologies, in Stephen’s terms : (i) bridge the gap between functional requirements and end design, (ii) enable the analysis of behaviour in subsystems and systems that exhibit state, mode transition and continuous behaviour, (iii) provide tools for design, analysis and implementation to manage complexity, and (iv) provide many of the necessary connections between software engineering, control systems engineering and physical systems engineering.

1 Craig Stephens, Powertrain Research and Advanced Engineering Ford Motor Company. Cyber Physical Systems Workshop, Detroit 3 April 2008. p3

Continue on Verification: