| Autors: Trendafilov, I. T., Gancheva, V. S. Title: Neuromorphic Assisted Sensor Grids Keywords: artificial intelligence, intelligent sensors, IoT Abstract: A particular problem in using artificial intelligence techniques in the sensor grid is the high power consumption. Remote sensors are usually limited by the amount of power available, thus our general goal is to minimize it while preserving accurate experimentation results. Using emerging technologies like spiking neural networks and neuromorphic hardware, we can create sensor grids that perform data processing with preserving low power consumption. In this paper, we demonstrate a remote node with integrated visual sensor that works on less than 10mW of energy, while performing continuous scene monitoring, object detection and classification. The system has intelligent power management and the ability to send data wirelessly. References Issue
|
Вид: публикация в международен форум, публикация в реферирано издание, индексирана в Scopus
