Sensor Technology and Fusion in Autonomous Vehicle
Loading...
Files
Date
2022
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Department of Electrical Engineering, NITS
Abstract
As the globe is moving forward towards the era of technological advancement, artificial intelligence and ease of access to the use of individual technology, it is evident that the automobile industry is evolving itself into autonomous technology and transforming itself to the generation of Autonomous Vehicle. These autonomous vehicle will not just be a smart transportation system but will also help in increase the accuracy of driving scenarios by making smart decisions in just a blink of an eye, reduction in road accidents and most importantly reducing emissions benefitting the environment. Therefore, to achieve all this a proper selection of multiple sensors is necessary. This project deals with a novel approach towards the selection and integration of multiple camera and radar sensor to achieve an enhanced field of view. This Multi-Sensor fusion approach has been carried out to achieve a more scalable and robust fusion architecture which helps in reduction of processing time of environmental input data to the fused output sensor data. To achieve the desired fusion results clustering algorithm techniques like Density Based Spatial Clustering Algorithm with Noise (DBSCAN) and Nearest Neighbor (NN) is used for radar and camera respectively. For data association, Joint Probabilistic Data Association Technique is implemented which helped to get the lateral and longitudinal position of radar and camera separately along with fused result of both radar and camera.