Accurate person detection is a key enabler for intelligent monitoring in smart home environments, supporting applications such as assisted living, activity recognition, and adaptive automation. 3D LiDAR sensors are particularly well suited for this task, as they provide precise geometric information while preserving privacy. However, robust indoor detection remains challenging due to cluttered interiors, occlusions, diverse human activities, and the limited availability of annotated datasets. We propose a novel LiDAR-based person detection approach that leverages cross-dataset transfer between deep feature learning and One-Class classification. Deep geometric features are first learned from the L-CAS 3D People Dataset and transferred to the SZTAKI LiDAR Gait-and-Activity dataset, where they are exploited by a lightweight One-Class SVM trained only on the “person” class. This enables the detection of non-person anomalies without requiring extensive negative samples. We report multiclass classification results on the L-CAS dataset, achieving over 98 % accuracy. On the more challenging SZTAKILGA dataset, the proposed method reaches up to 96 % accuracy and 96 %F1-score. Finally, the full pipeline is validated on a small in-house dataset collected in our laboratory, highlighting its practical feasibility in a realistic indoor setting and providing a foundation for future extensions toward multi-person tracking and advanced smart-environment applications.

While headsets dedicated to mixed reality are increasingly considered to virtualize workspaces in the office, the keyboard remains mostly unchanged and nevertheless is still envisioned as an interface for common professional activities heavily demanding in text entry. However, unlike headsets, a keyboard is intrinsically limited to immobile context during its use, due to the need for physical support. In this article, we investigate a radical change in the structural shape of the keyboard, switching from the plane to a cube, allowing users to fully benefit from their whole environment by entering text in a much wider variety of contexts. We designed a cubic layout based on the QWERTY and results from two user studies to define the best holding position as well as fingers preferences and reachability to the cubic shape. Then, we conducted two subsequent user studies and one longitudinal case study to compare and evaluate typing skill transfer from the keyboard, learning curve, fingers usage, usability, and workload in various contexts. The cubic-shaped text entry device finally showed its versatility with users having similar performances while standing with a mixed-reality headset or sitting in front of a laptop, and an input speed from an expert reaching 55.1 wpm.

Abstract Executive function operations (Formulate a Goal, Plan, Carry out the task and Verify goal attainment) are important for supporting independence, and are often impacted early in dementia, yet are seldom considered in the design of assistive technologies for cognition for older adults with dementia. This article introduces personas, i.e., fictitious, specific, concrete representations of target users, to support the design of assistive technologies for cognition from the perspective of executive dysfunction. We first categorized the assistance most appropriate to provide, based on a quantitative secondary analysis of annotated videos of 16 older adults who received assistance during a functional assessment. The annotations of the videos classified the assistance required for task completion into categories. We then designed the personas based on this quantitative secondary analysis. A persona was designed for each of the first three executive function operations: Formulate a goal, Plan and Carry out the task. No persona was designed for Verify goal attainment because assistance was seldom provided for this operation. Stimulate the thought process and clarification of instructions were the categories of assistance most frequently identified overall. Each persona is illustrated with examples of assistance. Stimulate the thought process was the category of assistance most frequently provided for goal formulation and plan, and motivational assistance for carry out the task. We defined several design recommendations to support the design of assistive technologies for cognition for older adults on the dementia continuum, including 1) stimulate the person to reason and act by themselves first; 2) design context-aware assistive technologies; 3) consider “goal formulation” and “plan” executive operation dysfunctions; 4) personalize assistive technologies to the specific needs of each individual; 5) do not rely only on personas to take individual needs into account. Personas created from real situations can serve as a tool to better understand the assistance this population requires in order to develop assistive technologies for cognition.

With the modernization of smart cities and the technological advancement of the Internet of Things, we are now reaching a point where technology can be weaved into the fabric of our cities. In this article, we tackle the challenge of using machine learning to recognize the profile of pedestrians based on their gait and silhouette with the use a thermal camera (FLIR Lepton) connected to a Raspberry Pi. We present the additional challenges faced by collecting gait data in a noisy environment, and by taking human identification a level of abstraction higher and recognizing categories of people. The far-reaching implications of such a system in terms of accessibility, inclusivity, and social participation of semi-autonomous populations are discussed. The hardware, software, and cost of the handmade prototypes used for data collection are described. In an effort to take a step toward sustainable smart cities, the possibility of powering this system using solar panels is investigated. This article aims to share the lessons learned throughout the creation and deployment of the system and to share the promising first results obtained by the team. We have reached an accuracy of 74.63% for binary age classification (adult, elder), 93.98% for mobility recognition (mobile subject, subject with reduced mobility), 85.77% for group size estimation (one subject, two subjects, or more), and 77.29% for observed gender recognition (male, female) using a 2-layer CNN without any preprocessing on very low-resolution thermal images.

Recent research allows envisioning what kind of sensory devices could be used for drivers’ navigation in the future, particularly for older adults. Population all around the world is aging, older adults will be more on the road. In this paper, we present a contact-less navigation system dedicated to this category of people based on a mobile head-up display. The aim of this system is to preserve their mobility in order to promote their autonomy and daily social activities. To do so, we interviewed older drivers to design a driving assistance system and we assessed the system under real road conditions (N = 34) with measurement of older drivers’ mental workload and the adequacy of users’ expectations. We emphasize the need to combine both measures of mental workload. The contribution aims at providing a richer understanding of how older people experience navigation technologies and to discuss the design recommendations of digital devices for older people.

The design of educational Serious Games (SG) remains a difficult operation which requires a tight weave between practices in instructional design and game design to be effective. Despite excellent works in the domain, the balance problem increases more significantly in the mobile learning system development such as Kids Smart Mobile School (KSMS) as a SG. KSMS is a school that aims to provide learning from K to 12 in Math and English as a Second Language to children without access to school in developing countries. This paper proposes a solution by designing a pedagogical pattern of a Learning Game Scenario, based on the educational Montessori approach mixed up with instructional engineering technique. This pattern is applicable to the various learning phases, making up the structure cognitive and pedagogical of KSMS. Moreover, this paper indicates how this pedagogical pattern makes easier the communication between members of an interdisciplinary team in different phases of design and development.

Recommender systems, deployed on the web and in mobile devices, help users in finding services and products that are addressing their needs, based on their profile. On the other hand, such systems help also organizations in promoting their products or services to a targeted market. Users in mobility in smart urban spaces are surrounded with products and services, such as mobile device applications. In this paper, we are presenting the result of a survey, where participants had to give their opinions on recommendation modalities in several scenarios of context-aware service recommendation in urban environments. These results are followed by an analysis and a discussion.

This document presents our proposal for a workshop on service delivery/provision in smart environment. The proposed type of workshop is a mini-track, with expected 10 to 12 authors/participants, divided in three presentation sessions.

Ubiquitous applications and smart environment technologies are complex to deploy, manage and use. Intelligibility, in ubiquitous computing applications, explains to users what a system did (outputs) and why it did it (inputs or contextual information). Making software more intelligible can reduce the complexity of a system for users. This paper presents our work on an intelligibility strategy for fuzzy logic systems, applied to a context-aware software organization and service provision (SOSP) middleware for smart environments. This fuzzy logic intelligibility strategy has been evaluated and tested with two groups of real users (technical and less technical users), and two versions of our prototype (with and without intelligibility).

Complexity in ubiquitous environments is an important problem that leads to high implementation and management costs. In some cases, the complexity of the environments' systems are so important that it become impossible for a single human to understand or manage them. The software self-organization simplifies this complexity by reducing the user manipulations and reasoning on the ubiquitous environments' system during software management and organization. This paper presents the self-organization middleware model and architecture, as well as results coming from tests and validation scenarios carried out on the middleware implementation.

Autonomic Computing Self-Configuration /Organization can simplify the complexity of software deployment, management and utilization in pervasive environments such as smart spaces. Software management complexity in pervasive environments is a serious problem, which leads to high management costs and slow down the broad utilization of the pervasive technologies. This paper presents our work on a self-organization middleware that reduce user manipulations and reasoning during software deployment and organization in smart spaces. Our approach is based on Fuzzy Logic and semantic description in order to support the self-organization reasoning on the environment context.

Context-aware software provides adapted services to users or other software components. On the other hand, Autonomic Pervasive Computing uses Context to reduce the complexity of pervasive system utilization, management and maintenance. This paper introduces two Context-awareness models, the macro and micro approaches, that define and integrate contextual views of individual pervasive components (micro level) and global knowledge of the system (macro level). These models are presented and compared with respect to their ability to meet the requirements of the Autonomic Pervasive Computing concept of the four selves.

The Autonomic Pervasive Computing can simplify the complexity of the configuration, maintenance and management of pervasive environments such as smart spaces. The Pervasive Computing is a paradigm where the information processes are distributed in the environment components, bringing processes more closer to the information context. On the other hand, the Autonomic Computing allows to simplify the complexity caused by the heterogeneity of the space components, there large number, the dynamic relations between them, etc.