Marijo Alilović

An algorithm for generating Real Drive Emissions (RDE) cycles using driven data. All legislative requirements are user inputs. It is possible to optimize for high or low dynamic cycles where one the dynamic parameters is near its boundary limit.

I developed a tool that creates routes within user-specified boundaries to maximize scenario coverage for testing Advanced Driver Assistance Systems (ADAS) on public roads. This includes a route generation algorithm, a fast API, and a user-friendly graphical user interface (GUI). This tool enables the easy creation of successful and cost-efficient public road testing campaigns for testing ADAS features.

In this project, my role encompassed the systematic data extraction from the database, followed by a meticulous tidying process to enhance its usability. Subsequently, I delved into an exploratory phase to gain deeper insights into the data's intricacies. Leveraging machine learning techniques, I navigated through various models to accurately predict the state of battery health. This iterative process allowed for a comprehensive understanding of the data landscape and facilitated the development of robust predictive models.

My primary responsibility in the project was to extract raw data from the database, then meticulously clean and organize it to ensure its accuracy and consistency. Subsequently, I crafted diverse transformations and aggregations to make the data suitable for visualization and key performance indicator (KPI) calculations. Utilizing R Markdown, I presented these insights in HTML format, enabling stakeholders to grasp and interpret the findings easily. This comprehensive process ensured that our visualizations and KPIs were not only informative but also accessible and visually compelling.

In this project, I took the lead in designing and implementing a cutting-edge tool for data enrichment. This tool provides an interface enabling integration of APIs into the workflow. This integration facilitates querying for supplementary metadata, enriching the dataset for more comprehensive analysis in subsequent stages. Through this strategic approach, we ensured that our analysis was bolstered by a wealth of additional information, enhancing the depth and accuracy of our insights.