Search for a command to run...
Various forms of cloud computing principles and technologies are becoming important recently. This paper addresses cloud computing for automation and control applications. It's argued that the open Internet cloud idea has such limitations that its not appropriate for automation. Since automation is physically and geographically local, it is inevitable to introduce the concept of local automation clouds. It's here proposed that local automation clouds should be self contained an be able to execute the intended automation functionalities without any external resources. Thus providing a fence at the rim of the local cloud preventing any inbound or outbound communication. Such a local cloud provides possibilities to address key requirements of both todays and future automation solutions. Adding mechanisms for secure inter-cloud administration and data tranfere enables local automation cloud to meet IoT automation system requirements as: 1) Interoperability of a wide range of IoT and legacy devices 2) Automation requirement on latency guarantee/prediction for communication and control computations. 3) Scalability of automation systems enabling very large integrated automation systems 4) Security and related safety of automation systems 5) Ease of application engineering 6) Multi stakeholder integration and operations agility. How these requirements can be met in such a local automation cloud is discussed with references to proposed solutions. The local automation cloud concept is further verified for a compartment climate control application. The control application included an IoT controller, four IoT sensors and actuators, and a physical layer communication gateway. The gateway acted as host for local cloud core functionalities. The climate control application has successfully been implemented using the open source Arrowhead Framework and its supports for design and implementation of self contained local automation clouds.