Data Analyst Leslie Wedraogo Advances Autonomous Vehicle Safety Through Ohio’s DriveOhio Project

As the race to perfect autonomous vehicle technology accelerates across the United States, one of its most important frontiers isn’t the test track—it’s the data lab. In Ohio, a growing hub for mobility innovation, Leslie Wedraogo, a Data Analytics Student Co-op, has been at the forefront of this effort through his work on the DriveOhio Project. His contributions have helped transform raw vehicle performance data into actionable insights that enhance the safety, reliability, and readiness of autonomous systems operating on public roads.

Working with large-scale datasets from autonomous vehicle testing in Athens and Vinton counties, Wedraogo conducted descriptive and diagnostic analytics on “disengagement events”—instances when human drivers must take control from automated systems. These brief but revealing moments offer a window into system limitations, environmental challenges, and operational risks. By identifying and analyzing patterns across geography, weather, and system states, Wedraogo’s analyses provided DriveOhio engineers and researchers with critical feedback to strengthen vehicle reliability.

Beyond analyzing what had already happened, Wedraogo’s work also looked ahead. Using k-nearest neighbors (KNN) and logistic regression models, he developed predictive frameworks capable of forecasting disengagement events before they occur. These models offered a data-driven approach to mitigating safety risks, enabling teams to anticipate vulnerabilities in real time. In an industry where even incremental improvements in safety can save lives, this predictive capability marked a significant step forward.

Equally impactful was how Wedraogo communicated his findings. Recognizing that complex analytics must be understandable to be useful, he designed interactive Tableau dashboards that visualized disengagement data across multiple variables. These tools allowed stakeholders—from researchers to industry partners—to explore patterns and trends intuitively, supporting faster, evidence-based decision-making.

Wedraogo’s technical foundation was rooted in Python-based analysis within Jupyter Notebook, where he performed preprocessing and exploratory data analysis on dynamic, large-scale datasets. This approach allowed him to capture subtle, time-based patterns that static analyses often miss—insights essential for optimizing system performance and supporting continuous learning in autonomous platforms. His methods reflected a growing awareness within the field that data structure and analysis design are just as vital as algorithm selection.

The co-op experience also demanded adaptability. Wedraogo applied agile research methodologies to bridge collaboration between academic teams and industry partners, ensuring that analytical insights were both rigorous and timely. In the fast-paced world of autonomous vehicle research, his ability to align data deliverables with iterative development cycles helped maintain project momentum and operational relevance.

By the end of his co-op, Wedraogo had made a tangible impact on Ohio’s DriveOhio Project and on the broader field of autonomous vehicle analytics. The frameworks and tools he helped create hold potential applications far beyond Ohio, informing transportation safety initiatives across the United States.

At a time when public confidence in self-driving technologies depends on measurable safety progress, Wedraogo’s work stands as a reminder that the future of transportation won’t be shaped by hardware alone—but by the intelligence drawn from the data behind it. Through his efforts, the DriveOhio Project moves closer to a safer, more connected autonomous future.