A flexible integration strategy for in-car telematics systems

In this paper we present the DriveLab IVIS testing platform which allows for the same experiments to be conducted both under simulator and real car conditions. Other key aspects of DriveLab is that it is highly modular (therefore allowing the exchange or integration of different components) and that it supports more than one driver. For example we show that the same IVIS devices and scenario can be used with two different 3D engines. The paper provides a technical overview and a brief example of use.

2018

Optimizing the software testing process has received much attention over the last few decades. Test optimization is typically seen as a multi-criteria decision making problem. One aspect of test optimization involves test selection, prioritization and execution scheduling. Having an efficient test process can result in the satisfaction of many objectives such as cost and time minimization. It can also lead to on-time delivery and better quality of the final software product. To achieve the goal of test efficiency, a set of criteria, having an impact on the test cases, needs to be identified. The analysis of several industrial case studies and also state of the art in this thesis, indicate that the dependency between integration test cases is one such criterion, which has a direct impact on test execution results. Other criteria of interest include requirement coverage and test execution time. In this doctoral thesis, we introduce, apply and evaluate a set of approaches and tools for test execution optimization at industrial integration testing level in embedded software development. Furthermore, ESPRET (Estimation and Prediction of Execution Time) and sOrTES (Stochastic Optimizing of Test Case Scheduling) are our proposed supportive tools for predicting the execution time and the scheduling of manual integration test cases, respectively. All proposed methods and tools in this thesis, have been evaluated at industrial testing projects at Bombardier Transportation (BT) in Sweden. As a result of the scientific contributions made in this doctoral thesis, employing the proposed approaches has led to an improvement in terms of reducing redundant test execution failures of up to 40 % with respect to the current test execution approach at BT. Moreover, an increase in the requirements coverage of up to 9.6 % is observed at BT. In summary, the application of the proposed approaches in this doctoral thesis has shown to give considerable gains by optimizing test schedules in system integration testing of embedded software development.

2009

Abstract Ensuring high object interoperability is a goal of integration testing for object-oriented (OO) software. When messages are sent, objects that receive them should respond as intended. Ensuring this is especially difficult when software uses components that are developed by different vendors, in different languages, and the implementation sources are not all available. A finite state machines model of inter-operating OO classes was presented in a previous paper.

Proc. of Convergence, 2004

Usually, the testing of today's ECU software follows a gut feeling approach, leading to test gaps and test redundancies. This paper presents a new, more systematic way of testing automotive control software. The central element of the approach is the Classification-Tree Method for Embedded Systems (CTM/ES). Using an interface description, which can be based on the specification and / or an executable model of the software, test scenarios can be derived systematically and described in a graphical way so as to provide the user with visual information about test coverage. The CTM/ES can be integrated into an overall test strategy for automotive control software developed in a model-based way. The approach opens up a new way of assuring quality for embedded control software which is especially designed for automotive software developers.

Industrial Information Technology, 2008

Usually, the testing of today's ECU software follows a gut feeling approach, leading to test gaps and test redundancies. This paper presents a new, more systematic way of testing automotive control software. The central element of the approach is the Classification-Tree Method for Embedded Systems (CTM/ES). Using an interface description, which can be based on the specification and / or an executable model of the software, test scenarios can be derived systematically and described in a graphical way so as to provide the user with visual information about test coverage. The CTM/ES can be integrated into an overall test strategy for automotive control software developed in a model-based way. The approach opens up a new way of assuring quality for embedded control software which is especially designed for automotive software developers. 1 General description of a test case which specifies a set of (detailed) test data or test data sequences. The work described was partially supported by the IMMOS project funded by the German Federal Ministry of Education and Research (project ref. 01ISC31D) www.immos-procject.de ... erroneous sub-domain ... test scenarios 2 Standardized procedures for selecting test scenarios on the basis of certain sources of information (e.g. functional specification, executable model, program code).

1997

Software architectures can play a role in improving the testing process of complex systems. In particular, descriptions of the software architecture can be useful to drive integration testing, since they supply information about how the software is structured in parts and how those parts (are expected to) interact. We propose to use formal architectural descriptions to model the "interesting" behaviour of the system. This model is at a right level of abstraction to be used as a formal base on which integration test strategies can be devised. Starting from a formal description of the software architecture (given in the CHAM formalism), we first derive a graph of all the possible behaviours of the system in terms of the interactions between the components. This graph contains altogether the information we need for the planning of integration testing. On this comprehensive model, we then identify a suitable set of reduced graphs, each highlighting specific architectural properties of the system. These reduced graphs can be used for the generation of integration tests according to a coverage strategy, analogously to what happens with the control and data flow graphs in unit testing.

Journal of Automotive Software Engineering, 2021

Autonomous driving and e-mobility are swiftly becoming not only the work of science fiction or popular science, but a reality. A key focus of manufacturers and suppliers in the automotive domain is of course to specify systems that implement this reality. Often, scenarios at type-level are used throughout the development process to specify system behavior and interaction within the car, as scenario models are comparatively easy to understand and can easily be subjected to manual validation. However, autonomous driving and e-mobility require interaction not just of systems within the same car, but collaboration between multiple cars as well as between cars and miscellaneous road infrastructure (e.g., smart road signs). The car becomes a Cyber-Physical System that dynamically forms collaborating networks at runtime with other Cyber-Physical System to create functionality that goes beyond the scope of the individual vehicle (e.g., resolve a traffic jam). Consequently, a plethora of possible compositions of such a network exist and must be specified and validated completely to assure their adequate and safe execution at runtime. Doing this at type-level with scenario models becomes prohibitively tedious, error prone, and likely results in unrealistic development cost. To combat this issue, we investigate the use of multi-level Message Sequence Charts to allow for specifying interaction scenarios between collaborative Cyber-Physical System in a network of collaborating automotive Cyber-Physical System. To assist the developer in systematically defining multi-level Message Sequence Charts, we propose two processes. The resulting diagrams use a mixture of type and instance-level abstractions within one conceptual diagram. This allows reducing the required effort to manually validate the adequacy of scenarios to a manageable amount because information within the scenarios can be validated in batches. At the same time, instance-level defects become more obvious. Evaluation results from a controlled experiment show that multi-level Message Sequence Charts contribute to effectiveness and efficiency of manual validation for collaborative automotive Cyber-Physical System.

Safety Science

In the European 6th Framework Programme, fundamental research in the field of technology assessment regarding behavioural effects was conducted in the HASTE (Human machine interface And the Safety of Traffic in Europe) project. Methods, metrics and scenarios for the exploration of safety problems related to IVIS (in-vehicle information systems) were defined and applied to the evaluation of surrogate systems in the laboratory originally and real systems subsequently, in laboratories and field studies. The results of the laboratory and field studies led to the formulation of guidelines for a standardized test regime regarding existing and newly developed IVIS. In this way the HASTE project has succeeded in the development of a valid, reliable and efficient tool that may aid testing authorities in their safety evaluation of IVIS.

Systematic Test of Embedded Automotive Communication Systems is the focus of the so-called STEACS project 1-a partnership between academia and industry. The main emphasis in respect of automotive communication systems is laid on the communication protocol FlexRay. The goal is to achieve a systematic test approach and to evaulate the feasibility by corresponding prototype implementations. This paper presents one of the results of the project, a generic tool basis supporting monitoring and diagnosis methods for FlexRay based systems, with interfaces to other protocols, such as CAN, and digitial and analog input signals.

Software Quality Journal, 2008

This paper considers the problem of integration testing the components of a timed distributed software system. We assume that communication between the components is specified using timed interface automata and use computational tree logic (CTL) to define communication-based coverage criteria that refer to send-and receive-statements and communication paths. The proposed method enables testers to focus during component integration on such parts of the specification, e.g. behaviour specifications or Markovian usage models, that are involved in the communication between components to be integrated. A more specific application area of this approach is the integration of test-models, e.g. a transmission gear can be tested based on separated models for the driver behaviour, the engine condition, and the mechanical and hydraulical transmission states. Given such a state-based specification of a distributed system and a concrete coverage goal, a model checker is used in order to determine the coverage or generate test sequences that achieve the goal. Given the generated test sequences we derive a partial test-model of the components from which the test sequences were derived. The partial model can be used to drive further testing and can also be used as the basis for producing additional partial models in incremental integration testing. While the process of deriving the test sequences

Proceedings of the 40th International Conference on Software Engineering: Companion Proceeedings

Testing is an important quality assurance activity during development of automotive software. Automotive OEMs and suppliers use test case specifications to specify, mostly informal, test cases as well as supporting information like traces to requirements. While the quality of the test case specifications has a high influence on the quality of the subsequent testing, quality of informal automotive test case specifications have not been investigated yet. In this paper, we present 7 potential quality indicators, ranging from requirement coverage to contents of a test step. The quality indicators have been identified in a case study of 816 current test case specifications specified by an OEM and suppliers.

Proceedings of the 5th International Conference on Automotive User Interfaces and Interactive Vehicular Applications - AutomotiveUI '13, 2013

In this paper we present the DriveLab IVIS testing platform which allows for the same experiments to be conducted both under simulator and real car conditions. Other key aspects of DriveLab is that it is highly modular (therefore allowing the exchange or integration of different components) and that it supports more than one driver. For example we show that the same IVIS devices and scenario can be used with two different 3D engines. The paper provides a technical overview and a brief example of use.

New Trends and Developments in Automotive System Engineering, 2011

The need for automotive testing and validation is growing due to the increasing complexity of electronic control systems in modern vehicles. Since testing and validation is expensive in terms of prototypes and personnel, simply increasing the volume of the testing can be prohibitively costly. Moreover, since product development cycles must be shortened in order to reduce the time-to-market for new products, there is less time available for testing and validation. Consequently, more testing and validation work will have to be performed in less time in future automotive development projects. To some extent this challenge can be met through virtual product development techniques and simulation, but there will still be an increasing need for testing and validation of physical prototypes. This can only be accomplished by improving the efficiency of automotive testing and validation procedures, and the key to realizing this, we will argue in this chapter, is by introducing novel information and communication support tools that fundamentally transform the way automotive testing and validation is conducted. With the explosive proliferation of wireless communication technology over the last few years, new opportunities have emerged for accessing data from vehicles remotely, without requiring physical access to the vehicles. Special purpose wireless communication equipment can be installed in designated test vehicles, acting as gateways to the internal communication buses and to on-board test equipment such as flight recorders. With a fleet of test vehicles thus configured, sophisticated telematics services can be implemented that enable communication of virtually any kind of data to and from any vehicle, providing the bandwidth of the wireless connection is sufficient. This has an enormous potential of making automotive testing and validation more efficient, since much of a test engineer's time is spent finding the right data to analyse. By eliminating the need for the engineer to have physical access to the test vehicle, scarce vehicle prototypes can be made available for multiple simultaneous tests, reducing the overall need for physical prototypes. Moreover, the test vehicles can be accessed by the engineers irrespective of their geographical location, which makes a much broader range of test objects available for tests and frees up time for the engineers in scheduling a prototype for a test. The data resulting from the test can be uploaded from the vehicles to a server from where it can be accessed by any number of interested (and duly authorized) engineers. By having measurement data automatically collected into a central database, as opposed to being stored on the hard drive of each engineer's computer, the opportunities for reuse of data is greatly

1998

Abstract This paper addresses the problem of statically analyzing input command syntax as de ned in interface and requirements speci cations and then generating test cases for input validation testing. The IVT Input Validation Testing technique has been developed, a proof-of-concept tool MICASA has been implemented, and validation has been performed.

"Software Engineering for Automotive Systems" Workshop W14S - 26th International Conference on Software Engineering

This paper presents a test framework which interlaces the test process with the development process in the domain of embedded control systems. The framework shows the following characteristics: • Strong integration of test and development process by solving test management issues inside development tools instead of specialized test management tools • Test in very early development stages by assuming a model-based development process where executable models are available • Reuse of test sequences at different abstraction levels and tool flexibility by tool-coupling The paper demonstrates concept and realisation of the test framework, which offers a manual but comfortable test suite derivation where test cases can be archived in the requirements database and interlinked with the requirement specification modules. Once the test sequences are defined, an automatic test instrumentation, execution and evaluation can be done. The test instrumentation is flexible, such that the system under test can be either a model or the realized hardware/software system.