COAST: A Connected Open plAtform for Smart objecTs

Quick reactions and rescues are critical requirements in disaster and emergency management. However, the decision on how to react should be taken based on the correct judgment on the disaster situation. Nowadays ICT plays a pivotal role in the decision-making in disasters management. Particularly, the IoT technology is indispensable for information collection and analysis. Despite the rapid development in the ICT domain, the real-time analysis and response in certain emergency is still impossible. This is mainly due to the lack of computation power of thin smart objects and the vertical isolation among IoT applications. In this work, we will introduce the Connected Open plAtform for Smart objecTs (COAST) that gives smart objects, such as robots, drones and smartphones, the capability to autonomously negotiate and deploy additional resources on itself, on other objects or in the cloud. The overall goal of the COAST platform is to give the capability to smart object applications to self-adapt to the runtime context by providing an interoperable platform for real-time discovery, negotiation and on-the-fly allocation of additional resources. These resources can be provided by remote sensors, by on-demand deployment of additional - possibly 3rd party - services on the same or another smart object, or by outsourcing computation to edge cloud infrastructure tailored for high-performance and real-time computation. COAST provides the necessary platform services supporting run-time adaptation, monitoring, big data analytics. To allow the secure deployment of 3rd party application components on smart objects, the COAST provides a novel virtualization layer that supports high-performance execution through hardware accelerators. Expected performance will be reported to highlight the benefits of introducing such a platform to the IoT services in disaster management.