Research on progressive dementia increased significantly in the past years due to the urgency of the aging population. Patients suffering from such dementia, for instance Alzheimer's disease, lose efficiency in cognitive spheres such as memory, planning skills, initiative and perseverance. Some researchers tried to evaluate the potential of close-to-reality simulations and generic video games for brain training to stimulate the cognitive abilities of AD patients. Using recent advances in artificial intelligence such as learning, activity recognition and guidance to enhance this concept of training, we are proposing, in this paper, a detailed explanation of an adapted serious game we designed for this purpose. A prototype has been developed showing how to exploit AI techniques to create an affordable and accessible tool for cognitive training and allowing in-game estimation of the patient's cognitive performance.
With the increasing demand in terms of non-intrusive appliance load monitoring (NIALM), more and more smart meters and smart analyzers were released on the market to extract well-defined load signatures and/or for performing autonomously the various monitoring operations as needed. Nevertheless, this hardware proves to be very expensive and not necessarily accessible to all. Moreover, most applications resulting of the use of these smart devices simply refer to energy saving and costs reducing of energy consumption. Thus, this paper proposes a new algorithmic method for an application field that is still very lightly exploited, i.e. the activity recognition of reduced-autonomy residents living in a smart habitat through load signatures. This one is based on steady-state operations and signatures and its extraction process of load signatures of appliances is carried out in a three-dimensional space through a single power analyzer which is non-intrusive (NIALM). This approach has been tested and verified rigorously through daily scenarios reproduced in the smart home prototype in a laboratory.. Hence, we can affirm that, with an exceptionally minimal investment and the exploitation of especially limited data, our method can recognize the use of appliances with high precision and low-cost allowing us to compete with other approaches which are much more expensive and require supplementary equipment.
<p class="MsoNormal" style="text-align: left; margin: 0cm 0cm 0pt;" align="left"><span class="text"><span style="font-family: ";Arial";,";sans-serif";; font-size: 9pt;">Assistance to people suffering from cognitive deficiencies in a smart home raises complex issues. Plan recognition is one of them. We propose a formal framework for the recognition process based on lattice theory and action description logic. The framework minimizes the uncertainty about the prediction of the observed agent’s behaviour by dynamically generating new implicit extra-plans. This approach offers an effective solution to actual plan recognition problem in a smart home, in order to provide assistance to persons suffering from cognitive deficits. An implementation of this model was incorporated in our smart home laboratory, in order to validate the approach. We currently planning the experimentation phase of the system, which will be based on a set of real case scenarios.</span></span><span style="font-family: ";Arial";,";sans-serif";; font-size: 9pt;"></span></p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.