Mark Jennings

Supporting Multidisciplinary Vehicle Analysis Using a Vehicle Reference Architecture Model in SysML

Abstract To develop competitive vehicles with ever increasing complexity, automotive designers need to improve their ability to explore a broad range of system architectures efficiently and effectively. Whereas traditional vehicle systems are based on internal combustion (IC) engines, todays environmentally conscious vehicle manufacturers must consider alternatives to the IC engine- only systems such as hybrid or electric systems. To design a good vehicle, it is necessary to analyze each of these system architectures from a variety of perspectives including performance, fuel economy, or even thermal behavior. Creating all the necessary analysis models for all possible system architectures manually is very time-consuming, expensive, and error-prone. To overcome such challenges, a novel approach has been developed for partly automatically generating subsystem model templates to support the integration of analysis models in a consistent and convenient fashion. The approach starts from a Vehicle Reference Architecture (VRA) model defined in the Systems Modeling Language (OMG SysMLTM). After specialization of this VRA into a specific vehicle program model, this SysML model is automatically transformed into Modelica and Simulink templates for the corresponding analysis models. These templates embody interfaces that fit into a system-level integrated model so that individual subsystem experts can focus on modeling the physical or controls behavior of their particular subsystem without having to worry about subsequent integration issues. The subsystem template models guarantee consistency in the integration phase. The entire approach introduced in this paper is called the Vehicle Architecture Modeling Framework (VAMF), which includes the SysML VRA model, the corresponding analysis templates, and the transformation tools developed to support the approach. Throughout this paper, a specific (realistic but sanitized) vehicle program and a full pedal acceleration analysis test scenario are used as demonstration examples.

Vehicle System Modeling for Computer-Aided Chassis Control Development

As the complexity of automotive chassis control systems increases with the introduction of technologies such as yaw and roll stability systems, processes for model-based development of chassis control systems becomes an essential part of ensuring overall vehicle safety, quality, and reliability. To facilitate such a model-based development process, a vehicle modeling framework intended for chassis control development has been created. This paper presents a design methodology centered on this modeling framework which has been applied to real world driving events and has demonstrated its capability to capture vehicle dynamic behavior for chassis control development applications.

Model Driven Vehicle System Engineering: Requirements, Issues and Challenges

The requirements, issues and challenges of establishing a model-driven vehicle system engineering (MD-VSE) process are explored. The needs for new methodologies, processes and infrastructure to address complex system level design issues are highlighted. The problem is viewed from the perspective of the relation between vehicle engineering domains and vehicle attribute and function requirements. With this view, an incremental approach to establishing a MD-VSE process is outlined. The approach consists of developing modeling frameworks to address critical attribute and function trade-off scenarios. The modeling frameworks would allow more formal and up front optimization and sensitivity analysis than is currently realized. In order to sustain these frameworks, a modeling cascade process has also been outlined. This would place formal requirements on VED’s to deliver models to support the MD-VSE process.Copyright

A Model Repository for Automotive Embedded Control Systems Design

Model-based product development methodologies are becoming more widely used by developers of automotive embedded control systems. This paper presents a model repository intended to provide configuration management, reuse, and sharing infrastructure in support of this trend. An initial set of repository requirements is presented and then augmented with lessons-learned from a pilot realization of the system. This pilot realization is discussed with respect to implementation and application. A repository data model is also described.Copyright

Vehicle System Modeling for HEV Systems Development

The sheer complexity of engineering propulsion systems for hybrid electric vehicles (HEV) demands the use of model-based development processes supported by comprehensive, robust vehicle system models. A Vehicle System Modeling (VSM) process has been developed to provide high-quality, application-appropriate vehicle system models in time to support critical HEV engineering activities. The process seeks to manage the complexity of the large number of model variants that are required to support a vehicle program. Additionally, it drives model development and aligns modeling activities with program timing. This paper describes the key elements of the VSM process and presents an application example. The application example illustrates the process by which a highly detailed HEV system model is created from an initial, base conventional vehicle system model via integration of high fidelity component models into a re-usable vehicle system modeling framework. The component models come from a variety of modeling tools and environments, which introduces additional complexity that must be managed. Results generated from the model show the complex system interactions that must be addressed by the vehicle control strategy. This re-enforces the notion that such modeling is required to achieve robust system designs.Copyright

Integration of Traffic Simulation and Propulsion Modeling to Estimate Energy Consumption for Battery Electric Vehicles

The introduction of battery electric vehicles (BEV) creates many new challenges. Among them is driving a vehicle with limited driving range, long charging time and sparse deployment of charging stations. This combination may cause range anxiety for prospective owners as well as serious practical problems with using the products. Tools are needed to help BEV owners plan routes that avoid both range anxiety and practical problems involved with being stranded by a discharged battery. Most of these tools are enabled by algorithms that provide accurate energy consumption estimates under real-world driving conditions. The tools, and therefore the algorithms must be available at vehicle launch even though there is insufficient time and vehicles to collect good statistics. This paper describes an approach to derive such models based on the integration of traffic simulation and vehicle propulsion modeling.