Multi-user conflict resolution mechanisms for smart home environments

A brief overview of pervasive computing (context-aware computing)

In 1991, Mark Weiser first proposed the concept of ubiquitous computing (Weiser, 1991), now often referred to as pervasive computing. According to his vision (which leads to smart environments), in future computing would move beyond the desktop and become widespread and invisible to individuals, in a sense that users interacts with the environment in a subconscious state. Context-awareness is the main pillar of pervasive computing that uses users’ contextual information (i.e., occupancy, activities, weather, location, etc.) and provide them with the service(s) of their interest (Abowd et al., 1999; Emmanouilidis, Koutsiamanis & Tasidou, 2013). Although context-awareness is a key component of managing daily activities in smart environments, there are a number of challenges associated with it. User control, inconsistent contexts, energy consumption, data privacy, security, and conflicts among users are a few examples (Dhyani et al., 2022; Rao & Prema, 2021; Alsamhi et al., 2022). While the research community is already investigating these challenges, more in-depth and focused research is required to broaden the scope of this fascinating area.

Smart environments and multi-user conflicts

A smart environment is made up of interconnected sensors, actuators, appliances, and applications. The connected gadgets adapt themselves according to the contexts to increase comfort and safety for their users. It is difficult to handle several users in a smart environment because it is ideally designed for a single user. Many users share a range of resources, and the environment is in charge of maintaining and structuring its resources to make them more accessible to its users. Multi-user conflicts occur when multiple users attempt to use context-aware applications at the same time or they have different priorities and preferences. For example, when user A enters a living room, the temperature and lighting control applications adjust to her preferences. But what if user B enters the same living room with a different set of lighting and temperature preferences?

Need to resolve multi-user conflicts

Devices in smart environments (specifically smart home environments) are typically controlled by a specific application embedded in the system device. This device should ideally be controlled by a single trusted person. A multi-user, multi-device smart home environment always pose different challenges like security, privacy, and manageability (Hua et al., 2022; Zeng & Roesner, 2019). Users have contradictory, complicated, and continuously changing demands on various devices, resulting in difficult-to-manage conflicts. So, in order to carry out daily tasks without difficulty in such circumstances, a thorough conflict resolution mechanism is needed (Ospan et al., 2018). Although multi-user conflict resolution has been a focus of study for decades, with an ever-growing Internet of Things (IoT) network and applications, it is necessary to focus more on meeting users’ requirements continuously.

Challenges of the traditional approaches to resolve multi-user conflicts

As discussed in “Literature review”, multiple strategies have been employed to handle multi-user conflicts. Most of these methods automatically detect and resolve user conflicts without the users’ active involvement. Nonetheless, there may occasionally be circumstances where users’ participation and discussion are needed to resolve the conflicts. User participation in conflict resolution is crucial since it improves the harmony of the home’s residents, as advocated in the literature. (Del Rio, 2022; Shin, Dey & Woo, 2010; Shin & Woo, 2009b). This calls for the environment to suggest the users with the best resolution candidates to resolve conflicts. Although these solutions successfully manage user conflicts, they fall short of dealing with some of the special cases. To the best of our knowledge:

• During conflict resolution, existing solutions do not take into account users’ special cases such as illness, examinations, visits, etc.

• To better resolve conflicts through discussion among users (mediation), the users’ involvement should be as little as feasible; nevertheless, the present literature is deficient in addressing this issue.

• If the user is not interested in using a certain application, she may be excluded from the conflict resolution discussion. This issue may lengthen the conflict resolution process; yet, the existing literature ignores this critical issue.

Research questions

Based on the research gap presented above, this research aims to address the following research questions:

• By factoring in user special cases into standard conflict resolution systems (such as automatic, mediated, and hybrid, see “Literature review” for more detail on these approaches), can multi-user conflicts be managed more effectively?

• How is user involvement reduced during a mediation-based approach to conflict resolution?

• If the user does not want to utilise a certain application, how is the conflict resolved?

Hypothesis

The null hypothesis ( $H_0$) and the alternative hypothesis ( $H_1$) for the respective research questions are the following:

• $H_0(1)$: Multi-user conflicts resolution in smart home environments is improved by including users’ spacial cases into existing automatic, mediated, and hybrid conflict resolution approaches.

• $H_1(1)$: Conflict of at least one of the existing conflict resolution approaches (automatic, mediated, and hybrid) remained the same by including users’ spacial cases.

• $H_0(2)$: During mediation conflict resolution approach, users’ involvement is decreased by automatically adapting to the preferences of the special case users?

• $H_1(2)$: During mediation conflict resolution approach, users’ involvement of at least a single user remained the same by automatically adapting to the preferences of the special case users?

• $H_0(3)$: Users’ conflicts in smart home environments is detected and resolved even if the user herself is not interested in using a particular application/service.

• $H_1(3)$: Users’ conflicts in smart home environments is not detected and resolved even if the user herself is not interested in using a particular application/service.

Research goal

The purpose of this study is to suggest a method for resolving conflicts between several users in environments like smart homes by actively involving the users and taking into account their special cases, such as illness, exams, visitors, etc.

Research contributions

Following are the contributions of this research:

• The proposed conflict manager takes into account users’ special cases (as stated above) for determining a resolution approach to be applied to detect and resolve users’ conflicts.

• To the best of our knowledge, no current literature has taken user special cases into account when choosing the resolution strategy to settle users’ conflicts, making this a novel method for resolving users’ conflicts.

• Users’ conflicts are resolved utilising the automatic, mediated, and hybrid resolution mechanisms that are already in place, but now users’ special cases are incorporated.

• Users’ conflicts for the special case users are resolved using mediated technique with less user engagement, and the user involvement during mediation has been reduced by letting the applications automatically adjust to the special case users’ preferences.

• The situation where the user herself is not interested in the application that is offered is regarded significant and is included in the proposed system to detect and resolve multi-user conflicts. This is an important but neglected issue that is also taken into account.

The proposed system is tested and evaluated utilising usability testing, which is implemented in Java. According to the usability research, the proposed system can be used to identify many conflicting circumstances with special cases and can soothe people by quickly resolving their conflicts.

The reminder of the article is organized as follow: “Literature review” presents the review of the related works that address the multi-user conflict issues in smart environments. “Methodology” presents the methodology of the proposed study that includes research design and complete experimentation of the proposed system. Results and discussion is presented in “Results and discussion”. Finally, conclusion remarks and future directions are presented in “Conclusion and future directions”.

Automatic conflict resolution approach (ACRA)

This strategy focuses on automatically resolving conflicts based on user priorities and/or preferences without users’ active participation. Various approaches have been used, using preferences in some cases, priorities in some other cases or a combination of both priorities and preferences. The method proposed in (Haya et al., 2006) suggested three algorithmic conflict-resolution techniques. The three methods are: (1) the fair principle, which is based on user preferences; (2) use first, which assigns priorities to the user who enters the environment first; and (3) preference priority, which gives priorities to preferences and settles disputes using the preference with the highest priority. The authors also advocated for giving the sickest user the highest priority during a quarrel and adjusting priorities or preferences accordingly.

Carreira & Sílvia Resendes (2014) considered conflicts resolution as the constraint satisfaction problem (CSP). Their system automatically resolved conflicts based on users’ preferences using some constraints. Constraints are valid range of values for users preferences and services that enable performing activities in environment. In case of non-satisfiable constraints, the system assists users in resolving conflicts by mediating of resolution candidates. Research work presented Camacho et al. (2014) also used CSP for conflicts resolution. The difference is that the latter used ontologies to detect and satisfy constraints imposed on the environment. Ontologies, is one of the concepts used in smart environments. It allows grouping of devices based on similarities e.g., device kind, device location, etc. (Elenius & Ingmarsson, 2004). Chaki, Bouguettaya & Mistry (2020) also formulated multi-user conflicts as ontology conflicts. Their system was able to detect whether a conflict happened in a single application or in a number of them, as well as whether it was a functional or a non functional conflict.

Chaki & Bouguettaya (2020) gathered information about users’ device and service usage patterns using the principles of entropy and information gain (IG), and then created an algorithm based on temporal proximity to identify and resolve conflicts. “Kratos” is a multi-user and multi-device aware access control mechanism that was developed by Sikder et al. (2020, 2022). The system consists of three parts: (a) a user interaction component that enables users to specify their access control settings; these settings are then translated into policies in the system’s second component; (b) a backend server; and (c) a policy manager that examines these policies to determine how to negotiate conflicts between users and creates final policies that will be used to resolve conflicts.

Mediated conflict resolution approach (MeCRA)

In certain circumstances, such as public settings and gatherings, identifying preferences and priorities may be challenging. Researchers have concentrated on settling conflicts in these circumstances by taking into account the preferences of the majority of users. Jukola (O’Hara et al., 2004) is a music mediator system for a public place i.e., cafe. It allows customers to influence the selection of songs played in the cafe. The system provides the customers a device on every table to mediate the list of songs showed on a shared display screen. After selecting songs by the customers, it plays the most rated song. This approach requires active users’ participation to resolve the conflicts.

As compared to public places (like restaurants), private places (like homes) require different kind of mediation as the home members can easily resolve their conflicts through discussion. Shin, Yoon & Woo (2007) in their research, proposed a user-centric conflict management system that considers different contexts and recommendations of a personal companion. Their system allows users to select from the recommendations of their personal companions to resolve conflicts. Mediation process enables users to exchange their opinion regarding media content to be agreed upon an item that reflects all the users’ preferences. Shin & Woo (2009a) proposed a conflict management framework for smart home environments where they used ontology to detect conflicts. An approach determination tree was used to assign an appropriate resolution approach to the detected conflicts. Based on the selected approach, conflict was resolved either by automatic decision or through discussion among the users.

Garg & Cui (2022) analysed three types of the most occurring conflicts in real life setting environments i.e., activity conflict, value conflict, and preference conflict. The conflict resolution was left on users so that they could resolve it through their mutual discussions as happens in their real life where the conflicts are either resolved by compromises or through mutual negotiations. Based on their findings, a set of considerations were provided that enable designers to design future IoT that may better fit into people’s homes and everyday lives.

Hybrid conflict resolution approach (HyCRA)

In a household setting where several conflicting circumstances could arise, resolving conflict with one resolution strategy would not lead to good results. As a result, a system supporting several techniques and multiple schemes is required to handle various conflicting circumstances.

Otto et al. (2006) designed a system that resolves conflicts using an input control device through explicit user interaction. Their solution used different approaches to deal with multi-user conflicts such as (a) giving priority to that person only who has the input control device, (b) allowing the person who entered first into the environment to have the input control device, (c) giving a specific time to every person to have the input control device, (d) involving every person to actively participate in the conflicts through mediation using personal digital assistant (PDA). In the last case, the system involved all the users even if they were not part of the conflicts and it then adapted itself according to the input to which all the users were agreed. Shin & Woo (2009b) proposed a socially aware TeleVision (TV) through which conflicts were resolved in either of two methods (a) automatically based on users profiles and, (b) by recommending users a common group profile. Their system provided a remote control that allowed users to mediate the final decision. The system then provided a final decision based on users’ recommended common group profile. The work presented in Shin, Dey & Woo (2010) resolved conflicts by either of two methods (a) using profile based automatic approach, and (b) by the use of social mediation. In social mediation approach, the users engaged in negotiating for a proper resolution. The system had a balance model to evaluate a group feelings to reduce a discussion time. Bisicchia, Forti & Brogi (2021) proposed a a declarative framework along with its opens source Prolog prototype “Solomon”, to specify policies for mediating contrasting goals and actuator settings in smart environments. Their proposed prototype resolved user to user and user to admin conflicts into a target state for the smart environment and its actuators.

To acquire a more in-depth understanding, the above-mentioned literature is also provided in a table. Table 1 gives a tabular form of the above-mentioned literature review.

Summary of literature review

Literature review presented above suggests that a multitude of research has been conducted in the proposed research area, resulting in proposition of different resolution algorithms (based on priorities and/or preferences) and approaches (ACRA, MeCRA, and HyCRA) to detect and resolve multi-user conflicts. Some of the proposed algorithms are suitable for public places (e.g., restaurants), while others are suitable for the private places (e.g., homes). The importance of the MeCRA in private places is stressed in the reviewed literature but the context-aware home environment dictates the need of minimizing users involvement to reduce their distractions, especially in situations where users may have special cases (i.e., illness, preparation for examination and guests). In the existing systems, if ACRA is applied for such special cases, it may lead to unpleasant results for those special case users. Moreover, if the MeCRA is applied, this will lead to a discussion among home users and the result will most likely be the home users conceding their right to the special case users as a resolution for the conflicts to provide them comforts. Since, the users’ special cases are temporary situations that might occur for specific amount of time. Through mediation, the family members show care for each other by conceding everyone’s rights of using the applications especially the one who has special cases (i.e., illness). This allows the family to live in more harmonic situations by providing the special case users the feelings that the other home members are caring for them. However, to better resolve the conflicts, there is a need to decrease the users involvement during mediation.

In order to lessen the users involvement, mediation can be minimized by allowing the applications to automatically adapt to the preferences of the special case users. The same has been focused here in this research. Also, an important but neglected aspect is considered i.e., in case if user herself is not interested in the application available at the vicinity such as TV (maybe because of work overburden or some other reasons). This situation is considered important and embed in our proposed approach to multi-user conflict detection and resolution. While the HyCRA to multi-user conflict detection and resolution has been used in the literature in which some conflicts are resolved using ACRA and others using MeCRA, none of the existing systems have considered use of special cases in the decision making of the selection of the resolution approach. The proposed approach takes into account special cases in determining a resolution approach to be applied to detect and resolve the multi-user conflicts in the smart home environment.

High level architecture of the proposed system

Figure 2 depicts the proposed system’s high-level design. UbiREAL simulator, users’ profiles, and users’ conflicts manager make up its three primary parts. This is a brief explanation of how each of these components function.

Conflict manager component

The conflict manager component is responsible for detecting and resolving multi-user conflicts. Its working is assisted by three sub-components i.e., conflict detection component, determination approach component, and resolution component.

(1) Conflict detection component: Its goal is to identify conflicts, collect information about them (such as involved users’ identities, profiles, and the location of the conflict, etc.), and deliver that information to the determination approach component to select the best strategy for resolving them.

(2) Determination approach component: Based on the data from the conflict detection component, it is incharge of choosing an effective resolution strategy. Its operation is aided by an algorithm i.e., “approach determination structure” (see Fig. 3), which helps choose the best conflict resolution strategy from among ACRA, MeCRA, and HyCRA (see “Literature review” for more information on these strategies).

As shown in Fig. 3, the approach determination structure selects ACRA in the following four cases:

• 1. If there is no special case user involved in the conflicting situation.

• 2. If there is only one special case user from the involved users.

• 3. If there are multiple special case users and the deviation in their preferences is low.

• 4. If there are multiple special case users and the deviation in their preferences is high.

MeCRA is selected when the involved users have the same special case and the deviation in their preferences is high. Finally, HyCRA is selected when there are multiple special case users with the multiple applications present in the environment and ACRA is not applicable. Once the appropriate resolution approach is selected, it is then sent to the resolution component to resolve the conflict.

(3) Resolution component: This component is responsible for making resolution about the detected conflict based on the selected resolution approach. The conflict might be resolved automatically without active users’ involvement, or by mediating some resolution candidates based on the involved users’ preferences, letting the users discuss among themselves and selecting the appropriate resolution approach. The conflict manager, after resolving the conflict, passes the values to the application(s), which adapts itself/themselves according to these values.

Participants selection

A total of 84 people were chosen from the academic community to take part in the usability study. Table 2 displays the demographic data for the participants, who were a mix of students and teachers. The participants were divided into 21 groups, each with four participants. Furthermore, each participant of each group was assigned a role in the scenario that was similar to their actual role in the family.

All methods/experiments were carried out in accordance with relevant guidelines and regulations as well as all experimental protocols were approved by Ethics Committee of University of Sindh, Jamshoro. Ethics approval and participant consent was taken as per policy of the University of Sindh, Jamshoro. All subjects in the database were enrolled at the university and have given informed consent, and if under 18, consent was taken from parent and/or legal guardian. Additionally, all the subjects have given the right to withdraw form the study at any time. Furthermore, an informed consent was taken from all subjects and/or their legal guardian(s) for publication of identifying information/images.

Environment selection

In order to implement the study a simulated virtual environment i.e., UbiREAL (Nishikawa et al., 2006; Alshammari et al., 2017) was selected. UbiREAL is a three dimensional (3D) virtual environment that provides a suitable environment to test the context-aware applications and allows to visualise the state change of devices through a 3D Graphical User Interface (GUI). UbiREAL simulator was made public with the source code in the 2012.

Implementation

The proposed method has been implemented on UbiREAL using Java. To detect and address conflicts among users, the proposed conflict manager with the UbiREAL simulator is built on top of UPnP (see Fig. 4). The conflict manager, which is also a UPnP client control point, subscribes to sensor events to learn when a device’s state changes. Using this information, along with user profiles (which were created using XML, as shown in Fig. 5), a resolution is then suggested, which ends up resolving the conflicts. To edit the XML files according to the users’ preferences, a GUI was created. The straightforward XML editor in Fig. 6 just comprises the Open and Quit buttons. When the open button is pressed, a new dialogue box allowing for the selection of an XML file for editing displays. After that, users may specify their names and preferences through a GUI without having to manually change an XML file. Another Java-based GUI was created to allow users to choose a resolution from the suggested resolution candidates when a disagreement arose. The Java-based GUI for choosing a resolution from among the suggested resolution candidates is displayed in Fig. 7.

System testing

The proposed approach was tested through the usability study of the implemented system. The test experiments were carried out in a room with a projector that projected the UbiREAL smart environment in front of the participants (see Fig. 8). The participants were told to interact with the environment from the front so that they might feel as though they were inside the environment. Before the usability study began, the participants were given a briefing on the users’ conflicts in smart environments and the methods employed to resolve them automatically. They were also briefed about the necessity of taking users’ special cases into account when resolving conflicts as well as the working of the proposed system in resolving users’ conflicts by taking users’ special cases into account.

The usability study’s participants were split up into 21 groups, each of which had four individuals. Each participant from each group received a role to play in accordance with their family roles (i.e., parents were given a parent role; children were given a child role etc.). Two distinct scenarios were created based on roles to be executed during the experiments (for more information on the constructed scenarios, see “Evaluation”). The scenarios were designed to test ACRA, MeCRA and HyCRA resolution approaches. Using these approaches and based on users’ preferences, a resolution candidate was recommended by the proposed system. Based on the recommended resolution candidate and discussion among users, an actor was appointed to apply the selected resolution candidate to resolve the conflicts. Using projected smart environment, the usability was performed as follows:

Initially, in case of recommendations, the recommendations were popped up on the screen of the projected smart environment. A GUI based utility was added to set the preferences of users (see Fig. 9). The settled users’ preferences were saved as a separate profile for every participant. After the users’ preferences have been saved, the experimental situation involving the applications and the degree of deviations in the users’ preferences were controlled. Three applications were selected and projected on the wall screen by the simulation i.e., (1) simulated air conditioner application (2) simulated television application, and (3) simulated light application (see Figs. 10A–10C). As the conflict occurred, the conflict manager detected the conflict and responded immediately to the system to employ the ACRA or MeCRA for the resolution.

Evaluation

To evaluate the performance of the proposed system with multiple applications, multiple users and different special cases, two different scenarios were executed. Considering multi-user conflicting home environment and different special cases, in these scenarios, every participant was given a role and a special case.