Heterogeneous social sensing edge computing system for deep learning based disaster response: demo abstract

Social sensing has emerged as a new application paradigm where measurements about the physical world are collected from humans or devices on their behalf. One of the representative application of social sensing is disaster damage assessment (DDA) that automatically identifies damage severity of impacted areas from imagery reports reported by eyewitness in the aftermath of a disaster (e.g., earthquake, hurricane, landslides). In this demo, we present a Social Sensing based Edge Computing system (SSEC) that can coordinate the privately owned IoT devices in close proximity of the disaster scene to collect, process and report the real-time status of the disaster. We showcase a supply chain-based resource management framework for SSEC that tames the pronounced run-time and hardware heterogeneity of the IoT devices at the edge to provide reliable sensing and computing power. The system is demonstrated on a real-world hardware platform consists of a diverse set of heterogeneous embedded systems.