A cooperative learning framework for mobility-aware resource management in multi-inhabitant smart homes

The essence of pervasive (ubiquitous) computing lies in the creation of smart environments saturated with computing and communication capabilities, yet gracefully integrated with human users. ´Context Awareness´ is perhaps the most important feature of such an intelligent computing paradigm. The mobility and activity of the inhabitants play significant roles in forming the context at any instance of time. In order to extract the best performance and efficacy of smart computing environments, one needs a technology-independent, context-aware platform spanning over multiple inhabitants. In this paper, we have developed a framework for mobility-aware resource (in particular, energy consumption) management in a multi-inhabitant smart home, based on a dynamic, cooperative reinforcement learning technique. The inhabitants´ mobility creates uncertainty of his location and activity. Using the proposed cooperative game-theory based framework, all the inhabitants currently present in the house attempt to minimize this overall uncertainty in the form of utility functions associated with them. Joint optimization of the utility function corresponds to the convergence to Nash equilibrium and helps in accurate prediction of inhabitants´ future locations and activities. This results in adaptive control of automated devices and temperature of the house, thus providing an amicable environment and sufficient comfort to the inhabitants. Simulation results point out that our framework can adaptively control the smart environment, while reducing the energy consumption and enhancing the comfort.