Autonomous Ground vehicle needs to tackle a lot of problems encountered in their pathways which needs proper detection and identification for navigation purpose. Detection and identification of obstacles during navigation helps in defining the trajectories for vehicle to maintain it into a safe drivable zone. Hence, it is necessary to fuse the data from different sensors for proper navigation. In this paper, vision and radar sensors data are used for classification of objects in the field of view of vehicle and the relative distance of detection is made by the Radar sensor. 77GHz mmw radar data has been coupled with a camera data for detection and identification purpose. YOLOv3 architecture has been used for obstacle detection through vision subsystem. It is observed that the proposed system helps in detection and identification of objects in real time during navigation of vehicle. This system may be reliable and accurate even in environments with low visibility like foggy or dusty weather due to features extracted by radar sensor without any distortions in spite of less visibility observed by vision sensor.
Software And Hardware
• Hardware: Processor: i3 ,i5 RAM: 4GB Hard disk: 16 GB • Software: operating System : Windws2000/XP/7/8/10 Anaconda,jupyter,spyder,flask Frontend :-python Backend:- MYSQL