Abstract
The work examines, at a conceptual level, the changes in attitudes regarding the impact of technology, in particular ICT (Information and Communication Technology), in supporting people and favoring their independent life in the developing society, and highlights the complexity of the applications to be implemented, if the society wants to contribute to giving all people not only the opportunity to carry the activities necessary for living, but in such a way to promote their well-being in general. Innovative approaches depend on many possible components, both from the perspective of the individual and from the technological and social point of view. First, it must be considered that the attitudes to supporting people, all people, during their life have changed at the international level, for example, at the levels of the WHO (World Health Organization) and the UN (United Nations). This is due to a greater international awareness of people's problems, but also to fundamental changes in the availability of technology and consequently in the organization of society. For example, the ageing of the population is very important. It causes the presence in the societies of people who have more or less marked limitations. This implies difficulties with a different level of severity and varying in time for carrying out activities necessary for a comfortable life (well-being). This requires, for the design of support services, an approach different from the past, personalized, and dynamic: that is, able to manage differences between the users and changes over time.
Introduction
Innovative approaches to supporting people in their lives depend on many components, including individual's perspectives, societal approaches, and the development of technology. Attitudes towards supporting people, all people, throughout their lives have changed internationally, for example at the level of the WHO (World Health Organization) and the UN (United Nations) (UN, 1948, 2006), from an approach based on a fundamental interest in health, to the consideration that the possibilities of living an independent life in a situation of well-being must be guaranteed to all. The WHO (2001) has drawn up a document in which the activities that people must be able to perform in order to live independently and satisfactorily, and the abilities needed to carry them out, are classified. People must be cared for in such a way as to maintain their abilities as much as possible. The environment, at all levels, must be structured in such a way as to allow such activities not only for individuals with average abilities, but in principle for everyone, and must be flexible enough to adapt to any lack or decrease in certain abilities. This may be greatly helped by the technological innovations that support the development of intelligent environments (Augusto et al., 2010), where smart objects can be integrated and connected to each other to provide features that support most people to perform the activities necessary for life with their abilities.
This approach has been verified in prototype applications developed in the IFAC laboratory (Burzagli et al., 2014), the first for support in the kitchen to feed oneself and the second to implement a support for people's loneliness (Burzagli and Naldini, 2018). However, this development activity has also highlighted the increase in the technical complexity of the implementation of support applications. In fact, it is necessary (1) to develop a complete knowledge of the considered activities, (2) to identify the problems caused by the environments and (3) to take into account the abilities of individuals that can vary over time. It is therefore necessary to collect information, organize it, and use it to think about the possible support of each individual person. This approach, together with the adoption of the most innovative technologies of data processing, such as Artificial Intelligence, requires an accurate study of the context and information on it. This step often represents the real application difficulty.
In the present work, the previous general statements are first developed in a conceptual way and then the design of a concrete application under development at the IFAC laboratory is described, which aims to provide pedestrians with accessibility to the surrounding environment, including both knowledge of the available services (e.g. pharmacies or post offices) and information on the possibility of reaching them.
The main contributions of the paper at the conceptual level are:
Adaptation of environments must start not from the dis/abilities of people, but from the activity that they have to perform. The service created should be adaptable to each person who uses it and therefore it must know, even in real time, the state of the user.
Support of people by technology
Since the development of ICT technology, activities have been undertaken to facilitate its use by all possible users. Examples are the use of synthetic voice (Emiliani et al., 1984) to allow blind people to work with computers or the use of voice recognition and presentation as text to facilitate communication for deaf people.
This section summarizes the increased needs of people, important new technological possibilities and how technology can and should be used to support people to enable an independent and enjoyable life.
Well-being—definition and necessary abilities
The attitude towards supporting people in their social integration has changed after the publication of important documents by the WHO and the UN, where the main emphasis is on people's well-being. For a long time, well-being has been mainly connected to the health situation, even if this was clear since 1948 with the WHO health definition, which states health as a complex concept depending on several components in addition to the absence of disease or infirmity (WHO, 1948). It is recognized that the functioning of an individual in a specific domain reflects an interaction between her health condition and the contextual environmental and personal factors. It is commonly accepted that one component of well-being is the possibility of conducting an independent, active, and fulfilling life.
The starting point for deciding how to support people in their social integration is to identify what they need to be able to do to live independently, considered as the first step toward well-being. Fortunately, WHO has published a document (WHO, 2001) “ICF International Classification of Functioning, Disability and Health,” where activities for independent living and necessary abilities of people for carrying them out are classified. ICF is fundamental from the perspective of thinking about usefulness of any support to people, taking into account factors, such as the body structure and corresponding functions, environmental factors, personal factors, and the possibility of participation of other people to offer support (participation). In ICF, the term functioning (i.e. being able to carry out an activity) refers to all body functions, activities, and participation. Body functions are defined as the physiological functions of the body system (including psychological functions). Activities are defined as the execution of a task or action by an individual. Activities may relate to the interplay of multiple functions and structures.
ICF definition of activities and the classification of health and health-related domains allow the identification of what people can do in a standard environment (their level of capacity), as well as what they do in their usual environment (their level of performance).
ICF acknowledges that every human being, and not a minority of humanity, can experience a decrement in health or live in an unfriendly environment, thereby experiencing integration problems. This is particularly evident for people who are ageing (Burzagli, 2023; UN, 1982; WHO, 2014), because they constitute a continuum from people who are still able to be completely independent to others who are not able to perform the simplest operations. Their situation also changes depending on the environment in which they are living and on time. Moreover, it must be recognized that support by technology may be useful, effective, and efficient, but in many situations, support by humans is needed. This means that applications should be based on or supported by social interactions to allow for human intervention in the support loop.
A fundamental shift in the conceptual attitude about the use of technology to support people is the reversal of the formulation of problems in this area. The starting point is not the lack of abilities of a person or a group of them, but the activities to be performed to live independently. The solution is to try to set up applications, to enable as many people as possible to carry out these activities in a satisfactory way. It is therefore necessary for living environments to be able to adapt to the needs of individual people.
A structured iterative process is necessary to identify user requirements and to design support applications. The one proposed in this paper is schematically shown in Figure 1. The starting point is provided by the knowledge available in the ICF classification that makes possible an approach based on design for all. The entire environment should be organized in an application (or support service) able to satisfy the identified needs of target users.
Design process.
Information and Communication Technologies (ICT), with their enormous developments in recent years, show ample possibilities for improving people's well-being, since they allow the creation of information applications and cyber physical (social) systems (CPS, CPSS) to support human daily life activities (Cassandras, 2016). In all environments, such as home, office, external environment, walking or using means of transport, a significant advantage can be represented by the availability of ICT applications, structured according to one's profile, and able to advise, support or guide in carrying out necessary activities.
This is due first to the fact that it is possible to see the environment not as a standard implementation where people are supposed to learn to live as comfortably as possible, but as a continuously developing system where intelligent objects and communication channels can be organized to help people in their activities and to be adaptive to their needs and preferences. This is the AmI concept, which is relatively old (Antona et al., 2007; European Commission, 2001) and has been presented as the emergence of an environment in which “people are surrounded by intelligent environments that are capable of recognizing and responding to the presence of different individuals in a seamless, unobtrusive and often invisible way.” Interaction is intended as taking place through “natural” interfaces in the context of an environment which meets the requirements of being unobtrusive (i.e. it impinges on people's consciousness only when needed), customable, adaptive to different user needs, and anticipatory (i.e. it tries to anticipate user needs). However, the main emphasis is not on making possible the interaction with the environment, assumed to be made available by the intelligent use of multimodality in it, but on the useful functionalities that applications in the environment are able to make available. These applications are characterized by increasing ubiquity, nomadism, and personalization, and are likely to pervade all daily human activities. They have the potential to enhance security in the physical environment, save human time, augment human memory, and support people in daily routines and simple activities, as well as in complex tasks (Emiliani and Stephanidis, 2005).
The technical implementation of such concepts requires the design of complex infrastructures, upon which the level of interconnected support service can be hosted. Anyway, important considerations are also necessary at the level of individual services.
From a design point of view (see Figure 2), the activity must start from the selection of one human activity and the structuring of the functionalities necessary to make the user able to carry it out, with the production of a generic application (let's say for the average user, as is the industrial approach now). If the functionalities offered are not satisfactory for a particular user, an additional component (based on artificial intelligence techniques) must start from the user's profile to adapt the available functionalities. Of course, in particular cases, it may be necessary to incorporate adaptations based on assistive technology into the system. It should be noted, however, that many adaptations of interfaces for people with reduced sensory and motor skills that have so far been solved with assistive technology products (e.g. speech or written text synthesis and recognition) are currently available in operating systems of commercial products, including telephones.
An AmI application.
Two main aspects are relevant from the perspective of actions aimed to support people in independent living and well-being. The first is the different meaning of interaction and therefore of accessibility. Most interactions with available systems were (and often are) based on the performance of tasks decided by the used application. Instead, in the smart environment the starting points are the goals of the user. They must be inferred by the system and decomposed into tasks that are adapted to the preferences of the individual.
Moreover, from the perspective of the industrial development of support systems, there is a huge difference if compared with the past. In the past the prototype product could go directly to industry (such as, e.g. the voice synthesizer for access to information by blind people). Now the prototype contributes only a small part of the creation of a necessary much more complex support ecosystem.
Due to the developments summarized in the previous sections, new perspectives are arising, which emphasize the role of intelligent environments towards providing useful means to support all people in their daily life activities, including older people and people with limitations of activities (Burzagli et al., 2022). In this context, accessibility, and usability, although necessary, are not sufficient. In order to achieve the above objective, it is necessary that the design of intelligent environments emphasizes usefulness in addition to usability.
So, to be able to structure any application to support people it is necessary to know in which activity it is necessary to support them, the environment in which they are and the available skills of the considered persons (of each individual person). To make the exposition clear, we refer to a laboratory activity that is being carried out at an experimental level, that is, the structuring of an application to support access to the external environment, in particular, the part that can be reached by walking. It is important to note that the approach involves important problems of information availability. It is necessary to know in detail the characteristics of the user and of the environment in which she needs to move and of their eventual evolution over time. Then, it is necessary to process this data to adapt each suggestion to the person and the present. So, it is necessary to have a profile of any person of interest, to be adapted to any change in the user's abilities and preferences, and a detailed description of the environment in which the person is located.
The user profile
The ageing of population is a particularly influential factor in this respect, resulting in a considerable proportion of people to be supported. (Old) people have a different perception of technology, even for a quite common decrease of mental processes (Charness and Boot, 2009; Czaja and Lee, 2007). However, smart environments have the potential to support independence and quality of life of older adults (Blackman et al., 2016). Intelligent living systems can also contribute to addressing emergency situations, which are expected to have an increase with age. Moreover, ambient assisted working can foster adaptations of the workplace, thus ensuring that all people are active and take part for the longest possible in the workforce (Bühler 2009). Then, it is necessary that public environments, for example transportation systems are redesigned to be usable by all (Coughlin and Brady, 2019). Finally, social networks can be fruitfully embedded in the very fabric of an intelligent environment. The information coming from social networks and any other application such as a forum, if conveniently processed, may contribute to the available knowledge, and contribute to limit possible social segregation. The system must be able to follow the persons over time, adjourning, their profile with information about their sensory characteristics (physical, motor, cognitive), interests, and behavior.
To clarify the necessary functions of the support systems, let's start with an example: the production of information to find a service in the environment and to go from a site to this service. A traveler is in a hotel. She needs a pharmacy and to be informed if the pharmacy is reachable for a pedestrian. Therefore, the system should know the environment at two levels. First, it needs to know the services that are around the hotel. Is a pharmacy available at a pedestrian distance? Then it must be able to find a path to it that is suitable for her abilities to walk. The problem seems trivial because navigation systems are available. However, not all the available services are reported in these systems and, normally, the information about accessibility is limited to the lack of architectural obstacles. For example, there may be no information about the slope of a road or the bad conditions of the sidewalks. Then, the system should consider the habits and preferences of the person. It should consider if she normally prefers to be informed about the shortest walk to the service or likes to use a longer route if it passes through a garden or near a shop of interest, like a bookshop. Her reaction to possible unexpected difficulties, due to the fact that the environment does not correspond to the available information, or her changed physical or psychological situation should be used to give her additional help. Any change from normal habits should be considered as a possible integration in the profile and controlled in time. Finally, from the perspective of interaction any suggestion should be organized to be presented on different devices, such as smart phones, tablets, computers, according to her present abilities.
Profile attributes are not equally important at all levels: they can be divided into two separate areas:
Personal characteristics: all data relating to the user's psychophysical health:
Motor deficit: Difficulty walking or even just stretches of bumpy road. Visual impairment: Trouble identifying objects or road signs, or oncoming cars. Health deficit: Health problems that can affect walking (heart problems, blood pressure …) Cognitive impairment: Difficulty interacting with the surrounding world. Personal preferences of individual users: these are personal characteristics that do not directly depend on the person's health, but rather on their mood or their concepts of quality of life (well-being). In this category of characteristics, the following can be considered:
Nature: Preference to pass through parks or near public green areas. Sociability: Preference to spend time with people, in meeting places or places of aggregation. Shopping: In a broad sense, preference to visit shops or markets. Public Services: Intended as services such as banks, post offices, municipal offices.
The application—accessibility to the environment
In planning living spaces (urban planning) accessibility is defined as “the potential for interaction” (Hansen, 1959), that is, the potential of the environment to provide services to people, while mobility is defined as “the potential for movement,” that is, the potential of people to move to the places where the services are available. Then, from the perspective of mobility, it is necessary to guarantee physical accessibility, that is, absence of physical obstacles (e.g. stairs). For example, paths on uneven ground are particularly dangerous for a lame person, while paths not known in advance or not signaled with tactile or sound aids are dangerous for a person with vision problems.
In the description of human activities, as pointed out previously, the ICF classification dedicates a specific chapter of the ontology to “Activities and Participation.” The specific section related to mobility exhibits an accurate and extensive description ranging from the ability to change the position of the body (d410-d429), to that of walking to move from one place to another, up to the use of means of transport (d470) or the ability to move objects from one place to another. Mobility is one of the person's abilities, that is, the ability to move the various parts of the body, or to move it from one place to another. From a broad perspective, mobility also is the support for conducting daily activities, such as reaching healthcare places, purchasing goods, cultivating social relationships, or even dedicating oneself to leisure activities. The possibility of conducting them substantially affects the person's level of autonomy, which depends on several factors: the person's physical and mental condition, the characteristics of the physical environment where she lives and last, but not least, the knowledge of the environment.
Pedestrian mobility plays a key role because it addresses a wide pool of people, including those who see the possibility of using means of transport diminishing over time. Therefore, a suitable management of this activity assumes even greater importance if the reference group are elderly people.
Existing systems to support mobility
The design of a support application for people's mobility, based on Information and Communication Technologies (ICT) could appear at first glance to be scarcely innovative, due to the existence of numerous navigation applications particularly efficient and effective, such as, just by way of example, Google Map (GOOGLE MAPS Home Page, 2005) and Open Street Map (OPEN STREET MAP Home Page, 2004). The creation of routes between two or more points, their measurement, and the indication of the characteristics, such as the signaling of points of interest or traffic flows, are current and widely used functions. Furthermore, specific aspects related to disability are also addressed by specific projects such as Access map (ACCESSMAP, n.d.) or Open Sidewalks (Open Sidewalk Project, 2012), which directly concern physical pedestrian accessibility, such as the mapping of sidewalks.
However, the selection of the path made by the system does not normally consider the user profile, but rather the choice of some parameters made by the user among those available. An added problem arises because the user profile information related to the user should not be static, but dynamic: it includes physical conditions, which can change over time, for example, if the user is tired. Additionally, the characteristics of the user profile are not limited to personal skills, but also include several preferences concerning social services, shops, and socialization activities.
One of the important problems for the implementation of an application to support mobility is the collection of geographical data. There are institutional sources, such as Open Data of Municipalities or Metropolitan Cities, which provide users with the possibility of accessing data directly from an Internet site. The data provided are often in Excel/CSV format or in GIS format, which therefore must be converted before being used by an application. This data is subject to variable updates, based on the availability of the source: in some cases, the data is updated on an annual basis, in others, there is not a clear updating plan. Another aspect is that not all types of Point of Interest (PoI) are registered and made available by Open Data.
Another source of data are the online services that provide maps and manage cartographic data, such as OpenStreetMap, Bing Maps, Google Maps, MapQuest, Here WeGo, Apple Maps, and Yandex Maps. These online services mainly assist with “navigation” and the elaboration of routes, and also allow access to PoIs data that populate the maps as an accessory. However, such availability, which is technologically possible through REST APIs (REST API, 2000) to be queried through personal tokens, is usually limited to some calls per minute or based on the obtained results. The paid API type of release format is the one adopted by Google, which makes the data available, but only after registering and selecting a commercial plan to use the APIs. Although there is a free commercial plan, the traffic limits and current requests are decidedly too low to be able to think about a consistent collection of data automatically. The information is provided in JSON-encoded format (JSON, 2006), at once readable by applications. The dynamic nature of online platforms guarantees an almost continuous updating of the data, but not its completeness: often the data is limited to the geographical position of a PoI without other details (timetable, price, goods sold, etc.).
Following an analysis of the various options, it appears that it may be convenient to use the data made available by OpenStreetMap (OSM), for two main reasons: OSM has the advantage of being an “open source” project and provides a detailed classification of PoIs, capable of reaching a deep level of detail. This represents an important advantage for taking into consideration the abilities and preferences of the user. Moreover, unlike Google, OSM freely releases geographic data (both at the level of “streets” and the level of points of interest).
Need of artificial intelligence
The approach previously described has two main consequences. First, a richer architectural approach for the support systems must be considered, where a control component is introduced (see Figure 3). Devices and smart objects need to be integrated to obtain a system able to support people in their activities. This control is not only seen from the perspective of the communication and interoperability of the technologies but has the specific objective of embedding the knowledge about technology and users and being able to match the implementable functionalities with the different and often continuously varying needs of users (Margetis et al., 2012).
Block diagram of an intelligent application.
This can be obtained by introducing Artificial Intelligence in the system, with the evolution of the control component to include two main blocks: a knowledge base and a reasoning system (Burzagli and Emiliani, 2014). The knowledge base needs to contain information about the activities to be carried out in the environment, the necessary functionalities, the available technologies, the interoperability issues (i.e. interfaces and communication protocols), the user profiles (abilities of individual users and their requirements and preferences), and the interaction issues (available interaction devices and modalities).
In this section, the conceptual structure of a system able to support people in their access to the environment is outlined. The system is composed of three building blocks: (i) a block capable of interacting with the user and to collect information about what she wants to do (e.g. going to the pharmacy) or where she wants to go, about her abilities and any variation over time, and about her preferences; (ii) a block that, connected with the artificial intelligence system, generates the paths to be proposed; (iii) the artificial intelligence block that is delegated to store and update the information on the environment and the user and to refine, if necessary, the tasks proposed by the system on the basis the reactions of the user and of the eventual community (data collection and reasoning).
The description, based on the first experiences of an experimental system under construction at our Institute, is purely functional without reference to particular technical solutions.
The interaction subsystem
The Interaction subsystem (Figure 4) must be able to perform three essential functions. The first is directly connected to the collection of the user's requests (PoI, point of departure and point of arrival—provided by the system if the user has chosen a PoI) and to the receipt of any comments on the proposals that are produced. The second is to visualize the proposed paths so that the user can have an impression of the environment in which she has to move and see how the paths are distributed in it. The third is to provide the system with the user's data (user profile). Through the interface, it must be possible to select one's own dis/abilities and to update them and to be able to effectively assess the reaction of the reasoning system to these variations. The AI component is also supposed to monitor the behavior of the user and to intervene if she has unexpected difficulties. In view of the previous discussion, this subsystem must therefore be able to collect information about the user in real time (taking privacy concerns into account).
Interaction system.
In Figure 5, a schematic representation of the system (Client) to which the user interface is connected is presented. It contains the user profile, that is, the data that describes the user, reporting the medical situation, any lack of ability, her needs, and preferences. There is also an agenda that checks the compatibility of requested movements with commitments already made. This information can be updated in real time by the user himself or by people in the community. All available data is sent to the reasoning system which produces a preliminary proposal suitable for the request, organized in a list of steps necessary to achieve the goal. The user can react by commenting on each proposed step, while the system monitors the behavior in real time to prevent user difficulties.
Client system.
When the system is started by a user for the first time, the interface is already set up to collect information on the personal features and preferences of the person using the application. Figure 6 shows a mock-up of a possible user interface. After entering the first and last name, the user can begin to report her skill deficiencies and preferences. For example, she can signal to the system that she may prefer longer routes, if they pass through a garden (nature) or shopping centers (shopping).
Mockup of the interface.
When the user thinks that the data entered is sufficient, they are sent to the data and reasoning system, which produces one or more solutions and transfers them to the user. Her comments or the observed behavior are transferred in real time to the system to decide whether adaptations to the suggestions made are necessary. Of course, these changes are stored in the system database.
The problem of generating routes can be tackled in a modular way, managing two separate criticalities: detecting the data of the road network and calculating the possible routes on them. To address these two critical issues, an ad hoc script can be created, for example, in Python to simplify data management and optimize processing times. Unfortunately, the data made available by OSM does not at once allow for the reconstruction of the “network” of nodes and connections that make up the urban fabric. To obtain this network another dedicated Python module, as osmnx (Boeing, 2024), may be used. This already has internal functions dedicated to detecting the OSM road networks of a given area.
To reduce processing times to a minimum, a solution was implemented in parallel with the Os. Path module, again written in Python, and which allows you to “stabilize” a network, saving the data in a dedicated file, which can be recalled from the internal functions of Python, and which therefore allows you to avoid having to download data at each iteration.
This represents the process of the creation of paths. The problem is not to obtain the shortest path between two points or nodes, but to generate several possible paths that can be used by a neural network in the reasoning subsystem. When the possible paths are made available, following the “human in the loop” approach the end user is placed at the center of the entire system, because all decisions and all proposals coming from the system must be adjusted according to the individual user who is requesting them. The strength of the whole approach lies in the possibility of implementing a dynamic system capable of customizing the proposals not only based on data from the outside world but also on the user's profile and behavior.
Let us now see in detail the functions that the block for finding possible solution (Figure 7) must be able to perform, first considering that the system for generating itinerary proposals (all the possible ones) can consider two possibilities: that the user is able to move on his own or that she needs support from the community. This is the reason why the requests that come from the user follow two paths, the one that concerns the user herself and the one that concerns the eventual support community.
Identification of practical solutions.
In both cases, the system considers all possible options, evaluating their characteristics and selecting those that appear possible according to the user's characteristics and preferences. If the community intervention is required, it accesses the information of the available people and selects the possible solutions (person to be involved) according to the characteristics of the user. In both cases, the different possible options are presented to the user and/or the community, who can make a preliminary assessment, discarding those that they do not consider clearly suitable or choosing one of them. Then, if a choice has not been made, the proposed paths are transferred to the reasoning system, each with the relevant data to characterize each user based on the information coming from the database. The system selects the most suitable, according to the available knowledge.
As previously suggested, new possibilities of proposing reasonable suggestions can be obtained introducing Artificial Intelligence in the system in a control component including two main blocks, as shown in Figure 8: a knowledge base and a reasoning system.
Data collection and reasoning.
The knowledge base contains information about:
The user profiles: Abilities of individual users and their requirements and preferences). The activities to be conducted in the environment (e.g. shopping or walking in a park). The abilities necessary to conduct these activities. The available technologies, which can be used to support the walking person. Their interoperability issues, that is, interfaces and communication protocols. The interaction issues: available interaction devices and modalities.
The implementation, maintenance, and run-time use of the system must be conducted under the control of a reasoning system (intelligence in the environment) capable of:
Enriching the knowledge base, being able to get, and integrate information already received in a formal representation (rule-based AI) and/or to extract it (machine learning) from informal information (e.g. from natural language text). Using the information in the knowledge base to adapt, the functionalities of the controlled systems and of their interfaces. Learning from usage to refine the knowledge base, introducing the necessary updates. Suggesting to the user alternative suitable means of interaction.
The system takes all the input data available (proposed tasks, as figured out by the block dealing with the selection of possible solutions) and supplies to the client the list of routes ordered according to the results of the reasoning system. It uses the information contained in the vector of user characteristics (W), containing the attributes associated with the profile of who is using the application. The vector W holds the values relating to lack of abilities that the subject has indicated in the input form and provides the data necessary for the system to personalize the results.
As a summary, the reasoning system selects the best path for each individual user, according to their characteristics, comments, and behaviors.
In the example, the user is an elderly tourist who lives in a bed and breakfast (green dot) and has told the system that she needs a bank. The nearest bank is in the square where a red dot is shown in Figure 9, along with three possible routes. To understand the map, it is necessary to start from the awareness of the fact that the system is not supposed to draw the shortest route between two of its points, but all walkable routes and, possibly, passing near services of interest for the user.
Path presentation.
There are three routes to get to the bank, all flat and, if necessary, passable with a wheelchair, a circumstance that is reported to the user. But they have distinct characteristics. Route 1 is the shortest, but the user must be warned to cross the street after the square, because the right-hand sidewalk, when passing under the second railway line, has an interruption that can be overcome with a route that is not immediately apparent. In addition, along the way, there are shops, but these can also be found in the arrival square (red dot) and on the other routes. Route 2 has no problems in crossing the roads (there are pedestrian crossings on both sides adapted for the passage of wheelchairs), and there is also a supermarket. The third route is longer, but it passes through a square where the user can rest on a bench and there is a post office along it. Therefore, it is convenient to propose the three paths to the user, describing their characteristics.
Finally, it is necessary for the user himself or for members of his community to be able to report problems that may not have been considered by the producer of the map or that may be present due to work in progress or carried out after the available version of the map. This can be achieved with an interface of the type shown in Figure 10.
Interface for input of road characteristics.
The aim of the paper is to show that technological developments in ICT can offer support to all people in conducting activities necessary to live independently, with an impact on their well-being.
However, in order to harmonize any support with the WHO attitude about people with ability limitations and the definition of well-being, it is first necessary that the study of support applications start not from the user's skills, as was the case with Assistive Technology, but with the activity they have to perform. People, all people, must be characterized by their abilities to be confronted with abilities necessary to conduct the necessary activities.
Then, from a technological perspective, it is necessary to be able to organize all computer-based objects in the environment so that it becomes not only a smart environment, but an instantiation of the general concept of ambient intelligent as discussed in Section 2.2. The spaces where people live must contain an interconnected system of computer-based objects, as a node of the Internet.
This needs the following components:
An in-depth knowledge of technology. Detailed knowledge of the environment and of the activities to be conducted in it. Detailed information about the users and their abilities in the activity they have to conduct, to be continuously adjourned. The possibility of reasoning about viable solutions, in cooperation with user as far as it is possible.
What is important is that the knowledge about what people need to do to live independently and the technology for supporting them is available. However, problems to be solved are complex, as evident in the support application under development in the IFAC laboratory, where it is shown that artificial intelligence techniques can be used to identify problems and suggest solutions.
Footnotes
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
