Capturing and interpreting the volumes of data generated by people and systems and drawing reliable conclusions is a growing challenge for conventional systems. Technologies and procedures from the field of artificial intelligence can support companies in using available information profitably. According to a recent Bitkom study, three quarters of the companies surveyed are aware that AI is one of the most important technologies of the future – but according to their own statements, the majority need support with the practical implementation.
MicroNova Consulting’s AI portfolio
Since every company has different goals and issues when introducing AI-based technologies, MicroNova focuses on comprehensive consulting for the use of AI and ML solutions. First, its experts support companies in selecting the right process for the intended application. The focus is on aspects such as costs, efficiency, and flexibility of the system.
After determining requirements and selecting the appropriate tool, the team of experts for data science and machine learning provide support throughout the entire project, from prototype development to the productive implementation of the solution within the organization. Methods from agile software development are usually deployed for this. Reliable, high-quality solutions are created on the basis of customer requirements.
Implementation of AI projects
The consultants not only assist with the analysis of the problem, concept development, and the construction of prototypes, but also support companies during the implementation of the developed solutions. All relevant stakeholders are involved in the process, and a clear definition of goals and requirements as well as an orderly structure ensure that the project runs smoothly.
“Nowadays, many processes can be made much more efficient through the correct use of artificial intelligence and data science. We see great potential in this area, especially with regard to the validation of autonomous driving functions with their vast number of variables,” explains Dr. Klaus Eder, COO of MicroNova. “A thorough inventory and preparation of the data is an important prerequisite for identifying the corresponding improvement opportunities and the development of AI systems.”
Application examples in the automotive industry
There are numerous applications for the use of artificial intelligence in the automotive sector. Four examples of AI applications are briefly explained below.
The creation of a so-called digital twin accelerates vehicle development because it enables control unit software to be tested in a virtual environment – without the need for a real prototype. This approach requires a large number of simulation models. With regard to autonomous vehicles in particular, such models can cover many different scenarios and factors virtually. Here, neural networks are particularly suitable for vehicle components that are difficult to reproduce with physical modeling, such as individual parts of an engine. MicroNova Consulting supports companies in data analysis, in the selection of suitable AI technologies for modeling, and in the integration and validation of individual model components.
Testing AI systems
Validating AI systems in vehicles – especially driver assistance systems and autonomous vehicle functions – requires new test methods, such as scenario-based testing. The necessary test environments must be able to integrate the control unit software and must also be massively scalable in order to perform the associated high test efforts in an acceptable time. Here, MicroNova Consulting helps with the design, introduction and further development of processes, methods, and tools for AI-supported systems for autonomous driving. Other areas of application include functional safety in accordance with ISO 26262 or the ASAM OpenSCENARIO standard.
Another challenge when testing autonomous vehicles is how to evaluate the large amounts of test results in a meaningful and practical way. MicroNova Consulting assists with the design and implementation of appropriate AI solutions.
Sensor data analysis
A vehicle must be able to recognize its surroundings and decide correctly which route to take or what response is required in order to move autonomously on the road. Sensors such as radar and lidar as well as cameras are used for this purpose as they complement each other. If, for example, weather conditions such as heavy rain or fog make it difficult to visually capture the environment via cameras, the system needs to be augmented with automotive radar technologies. These have been enhanced in recent years through new developments such as micro-Doppler automotive radars, standalone radar perception, and semantic radar. MicroNova Consulting supports companies in designing solutions for visual or radar-based object detection, e.g. traffic light detection or measurement of visibility in fog as well as the calculation of travel paths based on trajectories.
A printable photo of Dr. Klaus Eder, CEO MicroNova AG, is available for download here: https://bit.ly/31H5gjNy
- Full article: „Using AI Properly“ https://bit.ly/38bCbhk
The portfolio in detail:
The consulting portfolio in the field of AI includes different approaches in order to respond to the individual needs of the particular company.
Business understanding: The identification of problems or areas of application (use cases) usually marks the beginning of cooperation. The definition of the data required for the solutions is also part of the consulting service.
Data cleaning, data mining, data exploration: Data records are first prepared and cleaned up so that they are consistent in form and content. This makes it easier to derive relevant information from the data at a later juncture.
Predictive modelling: Comprehensive advice in this area in particular makes it easier to get started. MicroNova experts also provide support in post-processing results from existing models. Performance analyses of these models are also part of the service offering.
Feature engineering: Data mining techniques are used in this process to extract measurable properties of a dataset – the features – from the raw data. Combining and transforming these features allows new ones to be generated. The aim is to prepare the data in such a way that more patterns can be identified. Thanks to its feature engineering service, MicroNova supports companies in making more out of existing datasets than they appear to be at first glance.
Data visualization: The graphical representation of data sets is of crucial importance in most cases. It allows basic trends and characteristics of the data to be identified. Simulations, graphics and dashboard representations are among the most effective forms of data visualization. MicroNova provides support both in the selection of the appropriate tools for the creation of graphical representations and in their production.
Special workshops are held to concretize requirements and examine data already available in the company. These are precisely tailored to meet the needs of the company concerned and include, for example, an introduction to the principles and methods of artificial intelligence.