Xavier Vision: Pioneering Autonomous Vehicle Perception with YOLO v8 on Jetson Xavier NX

The real-time detection and identification of objects is critical to the efficient and safe navigation of autonomous vehicles (AVs). It is the initial phase of developing autonomous vehicles. Using low-power, tiny processing units along with edge computing has shown promise for object detection in autonomous vehicles. In this article, we look at and talk about the advantages and disadvantages of using the Jetson Xavier NX board for object detection in autonomous vehicles. We provide a synopsis of the most sophisticated deep learning model for object detection, called YOLO v8, discuss whether it can be used with the Jetson board, and investigate the next steps for integrating the two for edge computing. According to research, Jetson Xavier can effectively use the YOLO v8 with a mAP of up to 93%, an F1 score of 89.29%, and a response time as low as 6.7 ms. To solve the issues and guarantee the safe and responsible deployment of autonomous cars with object detection capabilities, more research and development is necessary.