An integrated framework to develop context-aware sensor grid for agriculture.

Context-Aware Computing involves sensing the context via sensors and other input channels, taking smart decisions, reflecting back the decisions and feedback tracking from the context. Agriculture is one of the key disciplines where context plays a vital role. Implementation of contextaware solutions in agriculture can be considered as an extension of Precision Agriculture. Very little work has been done in Context-Aware Agriculture. This involves transforming key agricultural concerns which follow traditional approaches to a context model governed via an I/O channel of sensors and actuators. The context model is a semantic model of the traditional practice. One novel approach to create context model is through building Ontology, use of wireless sensor and computing grids. In this paper, a context aware model of a typical watering process in agricultural fields have been developed and tested against the traditional practices. The results spanning over a large period of crop indicates that the context model clearly saves the scarce water resource and may have long term impact on improved decision making for agriculturists and other soil properties, directly affecting improving yield over a long period. As a cross disciplinary research, we have identified several technological issues and directions that make this novel approach viable under certain constraints, providing a pragmatic generic solution to solve many other problems like fertilization, pesticide spraying and early detection of various crop specific issues through the generic context-aware layer provided in our framework.