Christian Buckl

Embedded systems and software challenges in electric vehicles

The design of electric vehicles require a complete paradigm shift in terms of embedded systems architectures and software design techniques that are followed within the conventional automotive systems domain. It is increasingly being realized that the evolutionary approach of replacing the engine of a car by an electric engine will not be able to address issues like acceptable vehicle range, battery lifetime performance, battery management techniques, costs and weight, which are the core issues for the success of electric vehicles. While battery technology has crucial importance in the domain of electric vehicles, how these batteries are used and managed pose new problems in the area of embedded systems architecture and software for electric vehicles. At the same time, the communication and computation design challenges in electric vehicles also have to be addressed appropriately. This paper discusses some of these research challenges.

RACE: A Centralized Platform Computer Based Architecture for Automotive Applications

In the last couple of years software functionality of modern cars increased dramatically. This growing functionality leads directly to a higher complexity of development and configuration. Current studies show that the amount of software will continue to grow. Additionally, advanced driver assistance systems (ADAS) and autonomous functionality, such as highly and fully automated driving or parking, will be introduced. Many of these new functions require access to different communication domains within the car, which increases system complexity. AUTOSAR, the software architecture established as a standard in the automotive domain, provides no methodologies to reduce this kind of complexity and to master new challenges. One solution for these evolving systems is developed in the RACE project. Here, a centralized platform computer (CPC) is introduced, which is inspired by the well-established approach used in other domains like avionics and automation. The CPC establishes a generic safety-critical execution environment for applications, providing interfaces for test and verification as well as a reliable communication infrastructure to smart sensors and actuators. A centralized platform also significantly reduces the complexity of integration and verification of new applications, and enables the support for Plug&Play.

Analysis and optimization of fault-tolerant task scheduling on multiprocessor embedded systems

Reliability is a major requirement for most safety-related systems. To meet this requirement, fault-tolerant techniques such as hardware replication and software re-execution are often utilized. In this paper, we tackle the problem of analysis and optimization of fault-tolerant task scheduling for multiprocessor embedded systems. A set of existing fault-and process-models are adopted and a Binary Tree Analysis (BTA) is proposed to compute the system-level reliability in the presence of software/hardware redundancy. The BTA is integrated into a multi-objective evolutionary algorithm via a two-step encoding to perform reliability-aware design optimization. The optimization results contain the mapping of tasks to processing elements, the exact task and message schedule and the fault-tolerance policy assignment. Based on the observation that permanent faults need to be considered together with transient faults to achieve optimal system design, we propose a virtual mapping technique to take both types of faults into account. To the best of our knowledge, this is the first approach in fault-tolerant task scheduling that considers permanent and transient faults in a unified manner. The effectiveness of our approach is illustrated using several case studies.

Energy-Aware Task Allocation for Network-on-Chip Based Heterogeneous Multiprocessor Systems

Energy-efficiency is becoming one of the most critical issues in embedded system design. In Network-on-Chip (NoC) based heterogeneous Multiprocessor Systems, the energy consumption is influenced dramatically by task allocation schemes. Although various approaches are proposed to allocate tasks in an energy-efficient way, existing work does not well explore the tradeoff between the two major power consumers, namely the processors and network links, resulting in sub optimal mappings from a system point of view. In this paper, we first extend the existing Integer Linear Programming (ILP) formulation to take both processing and communication energy into account. Thereafter, we propose a Simulated Annealing with Timing Adjustment (SA-TA) heuristic to accelerate the optimization process. While the SA-TA algorithm achieves performance very close to the global optimum, significant improvement in computation speed is observed.

Services to the Field: An Approach for Resource Constrained Sensor/Actor Networks

More and more devices become network enabled and are integrated within one large, distributed system. The serviceoriented paradigm is the main concept to implement this approachand to cope with the heterogeneity of the underlying network. However, resource constraints imposed by the underlying hardware, such as 8-Bit micro controllers, require efficient protocols and often prohibit the use of technologies known from the Web service domain, the major implementation of the service-oriented paradigm. Nevertheless, a quick and seamless information flow between embedded devices and Web services is an important requirement for many application scenarios, e.g., real-time aware production management or the Internet of Things. Within this paper, we present an approach that allows to profit from the benefits of traditional SOA implementations, such as Web service interfaces and an IP compatible addressing scheme, and on the other hand can be implemented on resource constraint devices. The main idea is to use a data-centric processing paradigm at the device level and a gateway that mediates between the Web service and the embedded device world.

EasyLab: Model-Based Development of Software for Mechatronic Systems

Model-based development tools are one possible solution to handle the increasing complexity of mechatronic systems. While traditional approaches often separate design of hardware and software, especially in mechatronic systems hardware/software interaction is the most critical component. Hence, both aspects must be considered in this context. The goal is a model-based development tool for software/hardware co-design including the generation of efficient code for the respective target platforms. EasyLab is a modular and easily expandable development tool especially suitable for such applications. Its objectives are to facilitate reusability and to accelerate the development process. It raises the level of abstraction and thus simplifies the development of mechatronic systems even for unexperienced users. A graphical user interface provides various modeling languages that are easy to use. By employing platform optimized generation of the code, efficiency of the resulting programs can be guaranteed, which we demonstrate on a set of experimental mechatronic systems.

Towards adaptable manufacturing systems

Life cycles of many products are becoming shorter. In addition, the number of variants of one product is growing. As a fact, volume of one specific product that is being manufactured is decreasing. This leads to more frequent modifications of production lines. To cope with these changes, adaptable manufacturing systems are required. Current manufacturing systems can only be adapted to certain (predefined) situations. Other changes require high effort accompanied with high cost and setup time. In this paper, we focus on adaptivity with respect to IT systems. To increase the adaptability of IT systems for automation, we propose a model-based plug&play approach for integrating new stations. This helps in reducing changeover time and efforts. We propose different models describing stations and their capabilities, the setup of the factory, and the production plans. The system is then monitored automatically and the production is planned using models@run-time. To abstract from different platforms and communication technologies data transfer is handled by a middleware. We evaluate our approach using an industrial production system used for educational purposes.

The software car: Building ICT architectures for future electric vehicles

Disruptive technologies have the potential to change markets dramatically. The switch from internal combustion engines to electrical engines is such a change. But electric engines for vehicles are only the catalyst for the real change. Most significantly, the architecture and role of information and communication technology (ICT) will change for the vehicle of the future. This paper discusses the results of a study conducted in Germany on the role of ICT architectures. Furthermore, it will present an experimental platform that implements the vision of this study.

∈SOA - Service Oriented Architectures adapted for embedded networks

The development of embedded networks poses several challenges: complexity and size of the network, heterogeneity of nodes, and infrastructure dynamics. From other IT domains it is known that many of these challenges can be met by a Service Oriented Architecture (SOA). In the context of embedded networks, an application can be interpreted as a set of interacting services that produce, consume or manipulate data. However, the hard boundary conditions of embedded networks, such as limited resources and real time requirements, have to be taken into account. This paper discusses an embedded SOA (∈SOA) concept based on the definition of an embedded service (∈Service) term and the differences to traditional Web services based SOAs. The paper describes a middleware platform that supports the execution and development of embedded network applications by employing model based code generation and a pattern based service composition model. The advantages of the approach are showcased using an application from the building automation sector, focusing on the energy management of smart buildings.

Generating a Tailored Middleware for Wireless Sensor Network Applications

Wireless sensor networks are characterized by resource constraints. Therefore, todays sensor networks are implemented from scratch emphasizing code efficiency. This development strategy leads to relatively complex code and bad code reusability in further projects. To improve reusability and development efficiency, it is state-of-the-art in the development of standard information systems to divide applications into at least two parts, the application-logic, providing all the functions to solve a given problem and a reusable distributed middleware providing a container for the application. After developing the middleware once, the developer of further projects need to focus only on the application-logic. Thereby, the development times can be reduced considerably. However, a generic middleware layer replacing code implemented from scratch is not practicable in sensor networks due to resource constraints. Within this paper, we will present a model-driven approach in combination with a template-based code generator to get the best of both development strategies. This approach enables us to generate a tailored middleware for our application including interface-stubs for the application-logic. In contrast to other component-based approaches, the templates can be adopted easily to fulfill specific platform needs. We will demonstrate the practicability of this approach by implementing the control of a model railway.