Languages
DE
We are delighted to have been an exhibitor at InnoTrans in Berlin for the first time this year! The world’s leading trade fair for transport technology and mobility took place from September 24 to 27 at the Messe Berlin exhibition center.
Many thanks to everyone who visited us at our stand in the AI Mobility Lab and for the great and exciting discussions we were able to have. The numerous visitors from industry and science were able to find out about our projects and get to know our research activities in the field of AI. The focus was on the joint project KI-LOK. Many thanks also to our project partners from Fraunhofer FOKUS, Heinrich Heine University Düsseldorf and Hitachi Rail for their support at our booth!
See you in two years at the next InnoTrans!
ITPower Solutions at IEEE AITEST 2024 in ShanghaiOur colleagues Nicolas Grube and Paul Wancura successfully represented ITPower Solutions at the 6th IEEE International Conference on Artificial Intelligence Testing (IEEE AITEST 2024) in Shanghai. In two presentations, they showcased pioneering approaches for testing AI systems:
1. Outline of an Independent Systematic Blackbox Test for ML Systems
2. On a Systematic Test of ML-based Systems: Experiments on Test Statistics
Their contributions highlight our commitment to developing innovative solutions for the reliable and precise testing of AI technologies.
ITPower Solutions is a partner and project coordinator in the medCS.5 research project. The aim of the project is to analyze the patient data security and resilience of 5G infrastructures in the medical sector, identify potential threats and, derived from this, formulate recommendations for the security, design and operation of such networks. For this purpose, various complementary methods, innovative technologies, and novel algorithms and tests will be developed, validated, and deployed. Furthermore, it will be shown how network resilience can be further strengthened by complementary new technologies and developments.
New communication technologies such as 5G/6G will play an increasingly important role in the reality of healthcare: due to modern diagnostic and treatment procedures as well as mobile patient monitoring. However, this also poses major challenges for the resilience of the infrastructure, as well as for data security, especially for the most sensitive data: health-related information of the individual. It is precisely for this purpose that sufficiently protected processing must be ensured.
The project’s work results are expected to increase patient data security through novel AI-based monitoring algorithms and quantum encryption, while at the same time
simultaneously open up new application areas for technologies such as LiFi in the eHealth domain. In addition, simulative resilience assessment of medical campus networks and related advances in science and technology represent a key step in the development of smart hospitals in Germany. In this context, not only the security of patient-related data in medical-grade 5G networks improves, but also the safety, quality of care and experience of the patients themselves.
In addition to ITPower Solutions, project partners are the BTU Cottbus-Senftenberg, the Fraunhofer Heinrich Hertz Institute and the Center for Connected Health Care as a subcontractor. Project sponsor is the Federal Office for Information Security (BSI).
The project started on 01 January 2023 and has a duration of 2 years.
In modern software development, short development cycles and increasingly complex architectures, combined with high demands on product quality, determine the picture. To master this challenge, the inustry uses Continuous Integration (CI). In this case, software components are continuously assembled and automatically tested overnight.
A consortium consisting of three partners from science and industry is researching over a span of two years within the framework of the project “Optimised Test Selection for Continuous Integration (OTS4CI), how this process can be improved by optimising and automating the selection of test cases with the help of machine learning techniques.
The main challenge of the project is to map the complex process of test selection and prioritisation, which is normally performed manually by experienced testers, using suitable AI methods.
The technique of reinforcement learning is to be used for this task. This approach is an attempt to emulate learning behaviour in nature. An algorithm learns by interacting with information from its “environment” and subsequently receives information about the new state of the environment as well as a “reward”. The algorithm, also called an agent, should thereby learn a behaviour that maximises its reward.
The goal is to be achieved through an extended information model, in which data that accumulates during the development process is systematically analysed and used for test case selection,
is systematically analysed and used for test case selection.On the other hand, an environment model of the CI process will be developed. This model is fed with data from the information model and is intended to simulate the potential success of a certain selection of test cases in finding defects.
The successful selection of suitable test cases should then lead to the creation of a highly effective test plan through a so-called scheduling algorithm.
Participating project partners are the BTU Cottbus-Senftenberg, KOSTAL Industrie Elektrik based in Hagen and ITPower Solutions from Berlin. The project is part of the KMU-innovativ funding programme of the Federal Ministry of Education and Research and will run until 30 September 2024.
ITPower Solutions introduces certified quality management according to ISO 9001:2015In June 2022 ITPower Solutions certified its quality management system according to ISO 9001:2015, one of the most common and widespread quality management standards. For the implementation within the company, all existing, relevant processes were analyzed according to the specifications of ISO 9001 and adapted to the standard if necessary. These included quality policy, planning and management processes, resource management as well as the value-adding core processes. This is intended to ensure the key objective of the standard, to meet customer requirements and continuously optimize processes.
With its certification being applicable for the core business area – providing engineering services for the development of embedded software – ITPower Solutions demonstrates that the quality criteria that have been in place for years are compliant with the standard in all project phases and are therefore comprehensible and reliable for the customer.
“ITPower Solutions thus underlines its role as a specialist in the field of quality assurance of embedded software to provide our customers with services of excellent quality. It was important for us to show that the quality of our work results is demonstrably supported by the quality of our processes. For this reason we decided to align our activities in accordance with a professional quality management system and to have it certified by the accredited certifying body PÜG GmbH,” says Dr. Sadegh Sadeghipour, Managing Director of ITPower Solutions, commenting on the successful certification.
ITPower Solutions is a partner in a new research project investigating the use of new test methods in the development of AI-based systems. The aim of the project, called AI Locomotive Systems (KI-LOK), is to develop test procedures and methods for the safeguarding and certification of AI-based technologies for safety-critical applications in railway technology.
The research project is being carried out by a project network with a total of five partners. In addition to ITPower Solutions GmbH, the Fraunhofer Institute for Open Communication Systems FOKUS, Heinrich Heine University Düsseldorf, Thales Deutschland GmbH and neurocat GmbH are involved in the project implementation. The project is funded by the Federal Ministry for Economic Affairs and Energy within the framework of the funding guideline “New Vehicle and System Technologies” and is supported by TÜV-Rheinland.
The research project was motivated by the expected spread of AI solutions and machine learning (ML) in railway technology. The use of such technologies will not be limited to comfort functions, but will increasingly push into safety-critical areas. However, there are no industrially suitable methods and tools for ensuring the functional safety of AI systems in railway technology.
A central challenge here is the new properties of software for AI applications. The use of machine learning creates software products that are much more dynamic than the hard-coded algorithms of classic software. The behaviour of such systems is often not transparently comprehensible and is difficult to predict.
The project results expected by the partners can be summarised in three points. In a first step, methods and techniques for the verification and validation of AI systems will be developed. In a further step, the development of the KI-LOK methodology will provide a framework that brings together appropriate tools and processes. In the third step, a KI-LOK demonstrator platform will be developed to show and evaluate the benefits of the developed techniques.
The project starts on 01 April 2021 and runs over a period of three years.
ContinoProva is an efficient test tool for automated testing of embedded software. In order to offer new users a simple and quick access to ContinoProva, we have created the add-on BlinkTest Simulation, which allows you to get to know the features and behavior of ContinoProva within a given test scenario.
For this purpose, we have developed a simulation environment consisting of a test object and a tool environment. The simulation demonstrates how tests are specified, structured and executed. Errors in the test run are recognized by the tool and shown in the test report. Specifically, BlinkTest simulates an emergency braking situation of a vehicle to test if the emergency brake warning light operates correctly. During the test execution, the test components react to inputs from ContinoProva and provide the necessary data for test evaluation.
Although test cases and the test environment will be much more complex in real test projects, BlinkTest allows the user to experience a simplified but realistic test scenario.
The BlinkTest simulation environment is available now in the ContinoProva download area and on our evaluation website.
07.12.2020
ITPower Solutions GmbH constantly attaches great importance to confidentiality, availability and integrity in handling information. We are also obliged to meet the special requirements in terms of information security that apply in the automotive industry. We achieve this at the highest level by taking extensive measures to protect sensitive and confidential information.
Therefore, ITPower Solutions GmbH conducted the examination procedure for TISAX certification. TISAX stands for “Trusted Information Security Assessment Exchange” and is a standard for information security developed by the German Association of the Automotive Industry (VDA). It is based on the Information Security Assessment (ISA) developed by the VDA – a list of questions based on the ISO 27001 standard.
The TISAX assessment was conducted in November 2020 by the accredited testing institute operational services GmbH & Co. KG. TISAX and the TISAX results are not intended for the general public. The result is exclusively retrievable over the ENX Portal: https://portal.enx.com/de-DE/
TISAX is a registered trademark of the ENX Association