Low Latency, Low Power in A. I. Perception Modules Development and Implementation for Autonomous Driving

Abstract

Self-service vehicles can combine data to boost the understanding of that of other cars, and thus improve safety drive and identification performance. However, it is burdensome to share in between autonomous vehicles, because due to the quantity of data generated by different vehicle types of sensors. In the search for ever faster and more efficient computing, researchers and manufacturers are busy exploring novel processing architectures. Among these, neuromorphic engineering, the emulation of brain function inside computer chips are showing particular promise for applications involving deep learning, an increasingly common form of artificial intelligence (AI) that uses neural networks inspired by brains to uncover patterns in large datasets. In this research, we will examine an ultra-low-power protocol related to low-latency data exchange and Deep Learning Neural Network (DLNN) using neuromorphic computing for addressing AI perception issues in autonomous driving.