Pervasive behavior tracking for cognitive assistance

Abstract

The current and prospective situation of cognitively impaired people entails great human, social, and economical costs. Smart homes can help to maintain in place cognitively impaired people, to improve their autonomy, and accordingly to alleviate the burden put on informal and professional caregivers. The research performed at DOMUS lab aims at turning the whole home into a cognitive prosthetic, especially by providing cognitive assistance. In this process, behavior tracking is a fundamental piece. After sketching the infrastructure, two cognitive assistants are used to illustrate how activity recognition can help to address four kinds of cognitive deficits (initiation, attention, planning, and memory). An experimentation of one of these asistant involving people with intellectual deficiencies is finally shortly described.

Reducing temporal tensions as a strategy to promote sustainable behaviours

“…A cooking navigation system helps novice users to cook two dishes in parallel by optimizing time and processes, helped by text, video and audio from a touchscreen device (Hamada et al, 2005). Other research examines the use of HCI to overcome cognitive deficits via pervasive behaviour tracking and information (Giroux, Bauchet, Pigot, Lussier-Desrochers, & Lachappelle, 2008). The panavi system integrates wireless sensors to measure temperature and pan movements, and this data feeds the system that displays the situated instructions, helping users to master culinary arts (Uriu et al, 2012).…”

Activity tracking

“…Using low-cost sensors and tracking devices, researchers have begun to make important inroads into understanding complex symptomology associated with both physiological and neurological conditions [3,5,6,7]. Specifically, we hope to build on results of studies focusing on geriatrics [2], vulnerable users [11], chronic disease management [8], and dementia [1,9,10] to aid individuals experiencing chronic illness.…”

A Planning Approach to the Recognition of Multiple Goals

“…It is a ubiquitous task in the artificial intelligence and pervasive computing research. With the growing maturity of sensor and sensor‐network technologies, advanced applications are emerging speedily in diverse fields such as assisted cognition intelligence, sophisticated computer games, intrusion and terrorism detection, and military applications . In the real world of plan recognition, however, the multigoal recognition (MGR) problem is still a major challenge due to its inherent complexity, where an actor may have the intention to accomplish multiple goals within a single sequence of actions where goals are achieved concurrently and the actions that achieve them are interleaving.…”

“…Type 3 shows another special case where goal transitions between Goal 1 and Goal 2 happen frequently. In type 4, Goal 1 and Goal 2 are pursued concurrently and they share the middle two actions a 2 , a 3 . Goal 1 is achieved by three actions a 1 , a 2 , a 3 ahead of Goal 2.…”