Data driven scene classification to analyze ADAS

Abstract

In every sphere of life Big Data will be transformative. Data Visualization and Analytics plays an important role in decision making in various sectors. In autonomous vehicles data from various sensors and RADAR are stored in data logger, which is huge in size. To evaluate the performance of specific sensor manually is tedious task. This paper proposes an idea to create an interactive GUI framework to analyze the vehicle data and sensor data using big data visualization method. The framework contains various plots and plots are made interactive to analyze data in depth for all the scenarios of ADAS. It can be used to analyze the behavior of the vehicle at each instance of time interactively and time synchronized image frames are also incorporated with framework to see behavior of the plots. The paper proposes a Framework to analyze the huge amount vehicle data and sensor data which can be used to analyze the behavior of ADAS application.