Figure 3 - uploaded by Bruno Bouchard
Content may be subject to copyright.
Source publication
Smart home technologies have become, in the last few years, a very active topic of research. However, many scientists working in this field do not possess smart home infrastructures allowing them to conduct satisfactory experiments in a concrete environment with real data. To address this issue, this paper presents a new flexible 3D smart home infr...
Context in source publication
Similar publications
Smart home technologies have become, in the last few years, a very active topic of research. However, many scientists working in this field do not possess smart home infrastructure allowing them to conduct satisfactory experiments in a concrete environment with real data. To address this issue, this paper presents a new flexible 3D smart home infra...
Smart home technologies have become, in the last few years, a very active topic of research. However, many scientists working in this field do not possess smart home infrastructure allowing them to conduct satisfactory experiments in a concrete environment with real data. To address this issue, this paper presents a new flexible 3D smart home infra...
Citations
... Bouchard et al. [33] proposed SIMACT, a simulator that allows third-party components to connect to the simulator's database to receive and store real-time sensor readings. The simulator includes a set of pre-recorded scenarios (agent traces) to ensure data consistency. ...
One of the challenges in the field of human activity recognition in smart homes based on IoT sensors is the variability in the recorded data. This variability arises from differences in home configurations, sensor network setups, and the number and habits of inhabitants, resulting in a lack of data that accurately represent the application environment. Although simulators have been proposed in the literature to generate data, they fail to bridge the gap between training and field data or produce diverse datasets. In this article, we propose a solution to address this issue by leveraging the concept of digital twins to reduce the disparity between training and real-world data and generate more varied datasets. We introduce the Virtual Smart Home, a simulator specifically designed for modeling daily life activities in smart homes, which is adapted from the Virtual Home simulator. To assess its realism, we compare a set of activity data recorded in a real-life smart apartment with its replication in the VirtualSmartHome simulator. Additionally, we demonstrate that an activity recognition algorithm trained on the data generated by the VirtualSmartHome simulator can be successfully validated using real-life field data.
... However, this tool is not publicly accessible. Another 3D smart home simulator is SIMACT [27]. This tool has a series of prestored scenarios created from data collected from medical studies, which can be used to generate datasets for activity recognition. ...
This paper describes a methodology to optimize the home sensor network to measure the Activities of Daily Living (ADLs) of older people using Machine Learning (ML) applied to synthetic data generated via a newly developed Smart Living Environment (SLE) simulation tool. A home sensor network consisting of Passive InfraRed (PIR) and door sensors allows people to age in place, avoiding invasiveness of the technology by keeping track of the older users' behaviour and health conditions. However, it is difficult to identify a priori the optimal sensor network configuration to measure users' behaviour. To ensure better user acceptability without losing measurement accuracy , the authors proposed a methodology to optimize the home sensor network consisting of simulating human activities, and therefore sensor activations, in the reconstructed SLE and analysing the datasets generated through ML. Four ML classifiers, namely the Decision Tree (DT), Gaussian Naïve Bayes (GNB), Support Vector Machine (SVM) and K-Nearest Neighbors (KNN), were tested to measure the accuracy of ADL classification. Optimization analysis was made, providing the most suitable home sensor network configuration for two home environment case studies by exploiting the DT classifier results, as it proved to achieve the highest mean accuracy (over 94%) in measuring ADLs.
... Model-based approaches for the simulation data for the purposes of generating synthetic devices data include specific activity models that determine the arrangement of the events, the likelihood of an event taking place, and the time that the performance of specific activities takes. Bouchard et al. [17] propose an approach used within the SIMACT simulator. The tool provided a form-based interface for the specification of scripts that detail the series of steps involved in the performance of activities within a SH. ...
... The literature on how to model activity [18,20,55,65], how to learn these models [4,23,32,44,61,62], and how to test and simulate ADLs [2, 3,14,27,34,35,41,56,57], is quite vast. As we said, scientists have proposed many different ways of modeling an activity and activities library. ...
... In the literature, we can find several works that focus on how to generate datasets or simulate behaviors in smart homes [2, 3,14,34,41,56,57]. For instance, our team proposed in 2012 an open-source smart home simulator [14]. ...
... In the literature, we can find several works that focus on how to generate datasets or simulate behaviors in smart homes [2, 3,14,34,41,56,57]. For instance, our team proposed in 2012 an open-source smart home simulator [14]. It provided an interface to design the smart home and to write scripts. ...
Autonomy is a key factor in the quality of life of a person. With the aging of the population, an increasing number of people suffers from a reduced level of autonomy. That compromises their capacity of performing their daily activities and causes safety issues. The new concept of Ambient Assisted Living (AAL), and more specifically its application in Smart Homes for supporting elderly people, constitutes a great avenue of the solution. However, to be able to automatically assist a user carrying out is activities, researchers and engineers face three main challenges in the development of smart homes: (i) how to represent the activity models (ii) how to automatically constructs theses models based on historical data (iii) how to be able to simulate the user behavior for tests and calibration purpose. Most of recent works addressing these challenges exploit simple models of activity with no semantic, or use logically complex ones or else use probabilistically rigid representations. In this paper, we propose a global approach to address the three challenges. We introduce a new way of modeling human activities in smart homes based on Behavior Trees (BT) which are used in the video game industry. We then present an algorithmic way to automatically learn these models with sensors logs. We use a simulator that we have developed to validate our approach.
... The CASAS project [5], for example, focuses on the collection and sharing of smart home sensor data which is used to train recognition and prediction machine learning algorithms. Virtual smart home environments are also beginning to emerge due to their convenience; for example, OpenSHS provides a crossplatform 3D smart home simulator for dataset generation [6], and SIMACT is designed for research in the area of activity recognition [7]. Virtual Reality has also been used as a test-bed to design a home environment quickly and inexpensively, in order to collect data on which to create an activity recognition system [8], and has been combined with gesture sensors to conduct occupational therapy exercises and to collect important medical data [9]. ...
... It also needs to be able to construct an assistive step-by-step guidance protocol when something goes wrong [18]. Finally, researchers need to be able to simulate users' behaviour in order to help developing and testing smart home prototypes [8]. All these challenges have something in common: they first need to rely on a library of activities' models. ...
With the aging population, researchers around the world are investigating technological solutions to help seniors stay at home as long as possible. One of them is the concept of smart home, which is an intelligent house equipped with sensors and actuators. Aging people often suffers from physical and cognitive impairments, which limit their abilities to perform their Activities of Daily Living (ADL). Therefore, the smart home needs to be able to assist its resident in carrying out their ADL, when it is required. Recognising the ongoing ADL constitutes then a key challenge of the assistive services. Being able to simulate users’ behaviour is also an important issue, as well as being able to find an assistive step-by-step solution when something goes wrong. However, all theses challenges need to rely on a knowledge base of activities’ models. In the past, many researchers tried to make use of some logical encoding of the activities by exploiting, for instance, first order logic. These approaches work fine for the inferential process but they are very rigid, complex and time consuming. More recently, scientists in the field tried to represent the activities using stochastic models, such as Bayesian Networks or Markov Model. These probabilistic methods do not represent activities very naturally and are very static state-transition models. In this paper, we propose the use of Behaviour Trees (BT) as a means to represent the user’s ADL in a smart home. BTs are mainly used in the video game industry as a powerful tool to model the behaviour of non-player characters. BTs allow the modelling of activities with a flexible, well-defined approach.We will present a first exploitation of the behaviour trees in a smart home simulator.
... Smart homes can enhance comfort at home and help frail and disable people to recover autonomy. We can find several smart home simulator propositions in the literature [8][9][10][11][12][13][14][15][16][17][18]. These solutions help researchers to generate datasets to study issues and new approaches in the smart home domain. ...
... SIMACT is an open-source smart home simulator proposed in [15]. The authors use the model-based approach to design their simulator. ...
Advances in domains such as sensor networks and electronic and ambient intelligence have allowed us to create intelligent environments (IEs). However, research in IE is being held back by the fact that researchers face major difficulties, such as a lack of resources for their experiments. Indeed, they cannot easily build IEs to evaluate their approaches. This is mainly because of economic and logistical issues. In this paper, we propose a simulator to build virtual IEs. Simulators are a good alternative to physical IEs because they are inexpensive, and experiments can be conducted easily. Our simulator is open source and it provides users with a set of virtual sensors that simulates the behavior of real sensors. This simulator gives the user the capacity to build their own environment, providing a model to edit inhabitants’ behavior and an interactive mode. In this mode, the user can directly act upon IE objects. This simulator gathers data generated by the interactions in order to produce datasets. These datasets can be used by scientists to evaluate several approaches in IEs.
... SIMACT is a smart home simulator proposed in [9]. The authors use the model-based approach to design their simulator. ...
The advances in sensor networks, electronics and ambient intelligence make creation of intelligent environments (IEs) possible. However, on account of economic and logistic issues the implementation of physical IEs is difficult in research domain. That makes it harder for researchers to experiment new
approaches in IE domain. In this article, we propose a simulator to build virtual IEs. Simulators are a good alternative to physical IEs. Indeed, virtual IEs does not require expensive resources. Moreover, researchers
and designers can conduct experiments anytime and repeat scenarios easily. Our simulator provides users with a set of virtual sensors and actuators. Our virtual sensors try to reproduce behavior of physical sensors and to produce datasets with the same properties as those generated by real sensors. Our proposition
contains a tool to build a home from scratch and a model to define scenarios and behaviors of occupants. It also proposes an interface to control occupants directly. Virtual sensors collect data and generate datasets. Scientists and designers can use these datasets to evaluate and design new approaches in IE domain.
... Model-based approaches for data simulation for the generation of synthetic sensor data involve the specification of activity models that define the order of events, the probability of events occurring and the time taken for each event during the performance of specific activities. Bouchard et al. [55] provided an example of such an approach used within the SIMACT SH simulator. This tool provided a form-based interface for the specification of scripts that detail the series of steps involved in the performance of activities within a SH. ...
A globally ageing population is resulting in an increased prevalence of chronic conditions which affect older adults. Such conditions require long-term care and management to maximize quality of life, placing an increasing strain on healthcare resources. Intelligent environments such as smart homes facilitate long-term monitoring of activities in the home through the use of sensor technology. Access to sensor datasets is necessary for the development of novel activity monitoring and recognition approaches. Access to such datasets is limited due to issues such as sensor cost, availability and deployment time. The use of simulated environments and sensors may address these issues and facilitate the generation of comprehensive datasets. This paper provides a review of existing approaches for the generation of simulated smart home activity datasets, including model-based approaches and interactive approaches which implement virtual sensors, environments and avatars. The paper also provides recommendation for future work in intelligent environment simulation.
... Traditionally simulation of human activities has been based on an event-driven approach, which models human activities with consecutive events along a time line [7]. DiaSim [3], SIMACT [6], and Persim [2] are simulators that apply the approach. This approach requires detail design in a low level (by sensor events) for human activities to simulate with high realism and accuracy. ...
... Although it successfully examines the fidelity of applications, it lacks the ability to mimic a variety of human activities. Similarly, SIMACT [6], which provides 3D environment, lacks of designing modeling various human activities. By GUI, it increases realism for space, object, and activities, but the activities should be predefined in a script. ...
Human activities in smart spaces are traced by sensors and logged, as sensor events, in the form of sensory values, when the sensors detect elements of the activities. The event-driven approach that models the combination of the sensor events is one of the most common human activity simulation approaches. However, this approach is scalewise challenged as activities and spaces get more complex. A large volume of sensor events demands more human efforts in modeling, and requires higher processing overhead. We observe that rather than simulating by combining sensor events, semantical abstraction could offer a scalable alternative in managing such complexity. In our previous work, we proposed a context-driven such an approach, which scales well in complex simulation. The approach evaluates current state space and advances the simulation loop by units of context, not by sensor events. By changing the domain of simulation from event to context, we could measure a remarkable performance advantage. Through the experiment, we noticed that activity design was critical to end performance. In this paper, therefore, we focus on modeling activities for a better fit to the contextdriven approach. We introduce a new activity model along with associated algorithms to select and schedule the activities. We also provide an evaluation of the performance and computational complexity of the algorithms.
