Achievement

Trainee Piotr Szczurek of Computer Science worked with faculty members from Computer Science and Civil and Materials Engineering on a platform for developing and evaluating vehicular ad-hoc networks (VANET) safety warning applications. Examples of such applications include the emergency electronic brake light and the highway merge warning. A major issue with such applications is the impact of false warnings, which can lessen any safety benefits the application provides. A high number of false warnings will lead to driver desensitization to the warnings, which will reduce or completely eliminate any potential safety benefits. The platform developed by Piotr and Bo can be used to evaluate received alerts in terms of their relevance for the given vehicle. The platform provides a set of tools based on the machine learning for learning and testing relevance estimation modules. Simulation evaluation verified the feasibility of using it for various VANET safety applications.