A fuzzy logic control in adjustable autonomy of a multi-agent system for an automated elderly movement monitoring application

doi:10.1016/j.ijmedinf.2018.02.001

International Journal of Medical Informatics

Volume 112, April 2018, Pages 173-184

https://doi.org/10.1016/j.ijmedinf.2018.02.001 Get rights and content

Highlights

•
A Fuzzy Logic-based Adjustable Autonomy (FLAA) model to manage the autonomy of multi-agent systems.
•
The FLAA model is implemented in an Automated Elderly Movements Monitoring (AEMM-Care) system.
•
The AEMM-Care system monitors elderly people daily activities and implements fall detection and prevention tasks in a complex environment.
•
The results show that the model improves the performance of the agents and the accuracy of the fall detection and prevention tasks.

Abstract

Autonomous agents are being widely used in many systems, such as ambient assisted-living systems, to perform tasks on behalf of humans. However, these systems usually operate in complex environments that entail uncertain, highly dynamic, or irregular workload. In such environments, autonomous agents tend to make decisions that lead to undesirable outcomes. In this paper, we propose a fuzzy-logic-based adjustable autonomy (FLAA) model to manage the autonomy of multi-agent systems that are operating in complex environments. This model aims to facilitate the autonomy management of agents and help them make competent autonomous decisions. The FLAA model employs fuzzy logic to quantitatively measure and distribute autonomy among several agents based on their performance. We implement and test this model in the Automated Elderly Movements Monitoring (AEMM-Care) system, which uses agents to monitor the daily movement activities of elderly users and perform fall detection and prevention tasks in a complex environment. The test results show that the FLAA model improves the accuracy and performance of these agents in detecting and preventing falls.

Introduction

Many studies have reported a great increase in the median age of the humans, especially those who are living in developed economies [[1], [2]]. For instance, 17.4% of the population in the European Union have been classified as elderly (aged 65 years and above) in 2010, and this ratio is expected to reach 28.8% in 2050 [3]. As the world population continues to age, the number of elderly who are living independently or are left on their own at daytime has also been rising [4]. Some elderly people are facing health problems and require medical attention. To address the special needs of this population, many elderly remote care systems have been proposed, and some examples of these systems are presented in [5]. The elderly tend to show trembling, rigidity, and sluggishness in their movement which expose them to the risks of falling. This problem affects their ability to live independently, reduces their quality of life and can be hazardous and fatal [[1], [6]]. Several elderly remote care systems have been proposed in the attempt to solve or mitigate this problem, and some examples of these systems are presented in [2]. These systems monitor the elderly’s movement activities and daily routine patterns to prevent and detect fall situations.

Autonomous agents and multi-agent technologies have significant roles and contributions in many healthcare and elderly remote care systems [7]. For example, Kaluža et al. [3] propose an agent-based elderly remote care system that supports the independent living of the elderly. The system monitors their movement activities and automatically notifies medical personnel in case of a fall. Typically, agents in discrete and deterministic environments autonomously complete a substantial amount of tasks due to their prior knowledge about their surroundings. However, agents in complex environments that have the characteristics of uncertain, highly dynamic, or irregular workload tend to make decisions that lead to unwanted consequences [8]. These agents are deployed to handle some primitive, deducible, or critical tasks [9]. To make appropriate decisions amid such problems, agents must operate at different autonomy levels and with different autonomy properties [[10], [11], [12]]. In response to this need, several studies have attempted to model adjustable autonomy, which enables agents to operate at different autonomy levels. Some of these works have been reviewed in [8]. The adjustable autonomy in a multi-agent system is managed by grading the modifiable autonomy of agents within a specified range [11]. The grading process involves measuring the boundaries of autonomy parameters and estimating the extent to which autonomy is distributed and adjusted among the agents [[13], [14]].

In this paper, we propose the fuzzy-logic-based adjustable autonomy (FLAA) model to manage the autonomy of agents in multi-agent systems that operate in complex environments. This model employs fuzzy logic to linearly measure the autonomy of the agents. We implement and test this model in the automated elderly movements monitoring (AEMM-Care) system. The system observes the movement activities of the elderly to detect or prevent a fall. Given the intricate nature of fall detection and prevention tasks, the ambiguity of sensory data, and other challenges related to the application domain, we prove that the FLAA model improves the performance of the agents and the accuracy of the AEMM-Care system.

The rest of this paper is organized in the following order. Section 2 reviews the literature on agent-based elderly remote care systems, cites some previous attempts in applying fuzzy logic to manage the autonomy of agents and describes the needs and necessities of using adjustable autonomy. Section 3 illustrates the proposed FLAA model for multi-agent systems which includes the representation, measurement, distribution, and adjustment of autonomy. Section 4 discusses the prototype design, implementation, and application of the adjustable FLAA model in the AEMM-Care system. Section 5 presents the AEMM-care system test results and discusses the research outcomes. Section 6 concludes the paper and offers recommendations for future work.

Section snippets

Literature review

Our literature review reveals that the agent-based healthcare systems being used today are not using fuzzy logic to control the autonomous behavior of agents. Previous studies have also largely ignored the possible application of this technique in managing adjustable autonomy in multi-agent systems. To address these gaps, our literature review focuses on three topics of multi-agent elderly remote care systems, the use of fuzzy logic to control agents' autonomy, and a brief study of adjustable

The fuzzy logic-based adjustable autonomy model

This section explains how autonomy is represented, measured, distributed, and adjusted in the FLAA model. It also explains the architecture and operational behaviors of the agents that work according to the FLAA model.

The AEMM-Care system

The proposed FLAA model is implemented and tested in the AEMM-Care system, which detects and prevents fall incidents among the elderly by tracking their daily activities [25]. The system architecture, the testing scenario settings, and the falling situations that are investigated in the test are all adapted from [[1], [3]] and [6]. The AEMM-Care system is situated in an environment filled with uncertainty, which can be attributed to several factors. First, both the sensory data and the data

Results and discussion

The AEMM-Care system achieves irregular success rates in the fully autonomous mode (i), and such irregularity can be ascribed to the variances in the random forest agent’s learning of different cross-validation folds. t₁ and its actions have much lower success rates than t₂ and its actions because predicting those movement activity patterns that lead to fall situations is a complex process. The success rates of both t₁ and t₂ also do not change after multiple runs. The success rate of t₁

Conclusion and future work

We perform this research with an aim to develop a model for managing the adjustable autonomy of multi-agent systems that are operating in complex environments. The FLAA model quantifies the autonomy of agents based on knowledge and authority criteria, sets proper autonomy levels for these agents, and manages adjustable autonomy of the agents based on their performance. The autonomy management helps the agents to make highly flexible and efficient decisions and guides them toward achieving

Authors contributions

The contributions of this work are represented by the following points:

•
The first contribution of this work lies in its development of the FLAA model, which uses fuzzy logic to formulate the adjustable autonomy of a multi-agent system. This model is applied in a remote elderly movement monitoring system.
•
The second contribution of this work lies in its proposal of random forest agents with three adjustable autonomous run phases.
•
The third contribution of this work lies in the recognition of

Conflict of interest and authorship confirmation

•
All authors have participated in (a) conception and design, or analysis and interpretation of the data; (b) drafting the article or revising it critically for important intellectual content; and (c) approval of the final version.
•
The Article I have submitted to the journal for review is original, has been written by the stated authors and has not been published elsewhere.
•
The Images that I have submitted to the journal for review are original, was taken by the stated authors, and has not been

Acknowledgements

This project is sponsored by the postdoctoral grant of Universiti Tun Hussein Onn Malaysia (UTHM) under Vot D004 and partially supported by the Tier 1 research grant scheme of UTHM under Vot U893.

References (27)

E. Marcon et al.
A multi-agent system based on reactive decision rules for solving the caregiver routing problem in home health care
Simul. Modell. Pract. Theory
(2017)
A. Armentia et al.
A multi-agent based approach to support adaptability in home care applications
IFAC-PapersOnLine
(2015)
M.S. Couceiro et al.
A fuzzified systematic adjustment of the robotic Darwinian PSO
Rob. Auton. Syst.
(2012)
M. Lustrek et al.
Fall detection using location sensors and accelerometers
IEEE Pervasive Comput.
(2015)
I. Azimi et al.
Internet of things for remote elderly monitoring: a study from user-centered perspective
J. Ambient Intell. Hum. Comput.
(2017)
B. Kaluža et al.
A multi-agent care system to support independent living
Int. J. Artif. Intell. Tools
(2014)
L. Almazaydeh et al.
A panoramic study of fall detection technologies
Int. J. Comp. Sci. Issues (IJCSI)
(2016)
D. Isern et al.
A systematic literature review of agents applied in healthcare
J. Med. Syst.
(2016)
B. Cvetković et al.
Adapting activity recognition to a person with Multi-Classifier Adaptive Training
J. Ambient Intell. Smart Environ.
(2015)
S.A. Mostafa et al.
Adjustable autonomy: a systematic literature review
Artif. Intell. Rev.
(2017)

J. Jaafar et al.

Decision making method using fuzzy logic for autonomous agent navigation

Electr. J. Comp. Sci. Inf. Technol.

(2011)

S. Mostafa et al.

Defining tasks and actions complexity-levels via their deliberation intensity measures in the layered adjustable autonomy model

J.M. Bradshaw et al.

Dimensions of adjustable autonomy and mixed-initiative interaction

International Workshop on Computational Autonomy

(2003)

Cited by (66)

Resolution strategies for cooperative vehicle fleets for reducing rear-end collision risks near recurrent freeway bottlenecks
2023, Journal of Intelligent Transportation Systems: Technology, Planning, and Operations
Advanced cooperative vehicle fleet technology has shown potential for improving safety and efficiency near freeway bottlenecks. This study focuses on resolution strategies for cooperative vehicle fleets for reducing rear-end collision risks near recurrent bottlenecks. Based on different vehicle types and time requirements and intentions, different resolution strategies are proposed to determine the travel rules and sequence of vehicle flows when a cooperative vehicle fleet needs to change its original fixed formation. Then, a series of experimental scenarios is designed by setting different vehicle types (conventional vehicles and connected and automated vehicles), requirements and intentions in different areas of multilane freeways. The experimental scenarios are simulated based on the MATLAB platform to test the proposed strategy. The results show that the proposed strategy can effectively reduce vehicle rear-end collisions, and the average cost savings of the conflict resolution process is 25.71%. The proposed strategies in this research provide autonomous resolutions for vehicle conflicts in different scenarios.
A fuzzy rule-based approach via MATLAB for the CDR instrument for staging the severity of dementia
2022, Computer Methods and Programs in Biomedicine Update
Background: The CDR scale is a standard qualitative staging instrument that has been widely applied for assessing the severity of dementia which is based on information elicited through a semi-structured interview standardized in an assessment protocol. Despite clinical skills to elicit appropriate information are required, subjectivity still lies in the administration of the protocol and scoring process of the CDR. In this paper we propose a fuzzy rule-based CDR instrument to stage dementia based on the usual CDR, aiming to cover the subjectivities of the scoring process in the usual CDR which are directly related to the scoring system. This is effectively achieved by the F-CDR, our proposed expert system, which allows assigning scores continuously throughout the interval [0,3].
Results: In order to test the performance of our fuzzy model, we compare the outputs FCDR obtained from of F-CDR approach to the outputs U-CDR obtained by a usual application of the CDR via the same inputs for both. The dataset provided by ADNI, composed of more than eleven thousand CDR tests, including the inputs and outputs (U-CDR), is the source for comparisons.
Methods: The fuzzy rule-based model for the CDR that we propose in this paper is a fuzzy inference system (FIS) constructed in MATLAB with the aid of the Fuzzy Logic Designer app. The FIS was constructed based on the CDR and the specialist’s indications and tested on real data provided by ADNI.
Conclusion: The high accuracy of matches between U-CDR and F-CDR via the same inputs over random samples selected from the ADNI dataset suggests that the fuzzy approach to the CDR instrument here proposed is suitable to extend the scoring process of the usual CDR since the fuzzy approach allows the possibility of scoring continuously in the interval [0,3].
Leader-following bipartite consensus with disturbance rejection for uncertain multiple Euler–Lagrange systems over signed networks
2021, Journal of the Franklin Institute
Citation Excerpt :
In recent years, multi-agent systems(MASs) have always been a hot issue in the control fields and have aroused a storm of investigation because of their extensive applications [1–4].
In this paper, the leader-following bipartite consensus is investigated for a group of uncertain multiple Euler–Lagrange systems with disturbances. An innovative adaptive distributed observer is developed without requiring that followers surely acquire the leader’s auxiliary state and system matrix. A directed signed network satisfying the principle of structural balance is exploited to describe the interaction among agents. Then a novel bipartite consensus control protocol is proposed to solve the bipartite consensus problem of multiple Euler–Lagrange systems. The theoretical proof is provided via constructing a Lyapunov function and applying Barbalat lemma to analyze the convergence problem. Finally, a numerical simulation is utilized to demonstrate the effectiveness of proposed method.
Robustness analysis of classical and fuzzy decision trees under adversarial evasion attack
2021, Applied Soft Computing
Citation Excerpt :
This study attempts to investigate whether and how fuzzifying a system affects its robustness against evasion attack in an adversarial environment. Classical and Fuzzy Decision Trees (DT and FDT) are the focus of this paper due to their popularity in broad applications, e.g.multi-agent systems [25], natural language process processing [26], diagnosis systems [27], malware detection [28,29], network intrusion detection [30] and spam filtering [31]. A fair comparison on the robustness of DT and FDT requires a general evasion attack which is able to attack both DT and FDT.
Although decision trees have been widely applied to different security related applications, their security has not been investigated extensively in an adversarial environment. This work aims to study the robustness of classical decision tree (DT) and Fuzzy decision tree (FDT) under evasion attack that manipulate the features in order to mislead the decision of a classifier. To the best of our knowledge, existing attack methods cannot be applied to DT due to non-differentiation of its decision function. This is the first attack model designed for both DT and FDT. Our model quantifies the influence of changing a feature on the decision. The effectiveness of our method is compared with Papernot (PPNT) and Robustness Verification of Tree-based Models (RVTM), which are state-of-the-art attack methods for DT, and the attack methods employing surrogate and Generative Adversarial Network (GAN) methods. The experimental results suggest that the fuzzifying process increases the robustness of DT. Moreover, FDT with more membership functions is more vulnerable since a smaller number of features is usually used. This study fills the gap of examining the security issue of fuzzy systems in an adversarial environment.
Agent architecture of an intelligent medical system based on federated learning and blockchain technology
2021, Journal of Information Security and Applications
Citation Excerpt :
Solutions based on the operation of multiple agents boils down to modeling communication and information exchange mechanisms depending on the current goals. One such solution is to use a fuzzy logic controller for the adjustable autonomy of a multi-agent system (MAS) [8], where the main application of the described proposal was automated elderly movement monitoring application. This kind of research shows that an agent-based approach can be valuable due to analyzing data.
Multi-agent systems enable the division of complicated tasks into individual objects that can cooperate. Such architecture can be useful in building solutions in the Internet of Medical Things (IoMT). In this paper, we propose an architecture of such a system that ensures the security of private data, as well as allows the addition and/or modification of the used classification methods. The main advantages of the proposed system are based on the implementation of blockchain technology elements and threaded federated learning. The individual elements are located on the agents who exchange information. Additionally, we propose building an agent with a consortium mechanism for classification results from many machine learning solutions. This proposal offers a new model of agents that can be implemented as a system for processing medical data in real-time. Our proposition was described and tested to present advantages over other, existing state-of-the-art methods. We show, that this proposition can improve the Internet of Medical Thing solutions by presenting a new idea of a multi-agent system that can separate different tasks like security, or classification and as a result minimize operation time and increase accuracy.
Artificial Intelligence and IoT in Elderly Fall Prevention: A Review
2024, IEEE Sensors Journal

View all citing articles on Scopus

View full text

A fuzzy logic control in adjustable autonomy of a multi-agent system for an automated elderly movement monitoring application

Highlights

Abstract

Introduction

Section snippets

Literature review

The fuzzy logic-based adjustable autonomy model

The AEMM-Care system

Results and discussion

Conclusion and future work

Authors contributions

Conflict of interest and authorship confirmation

Acknowledgements

Simul. Modell. Pract. Theory

IFAC-PapersOnLine

Rob. Auton. Syst.

Fall detection using location sensors and accelerometers

IEEE Pervasive Comput.

Internet of things for remote elderly monitoring: a study from user-centered perspective

J. Ambient Intell. Hum. Comput.

A multi-agent care system to support independent living

Int. J. Artif. Intell. Tools

A panoramic study of fall detection technologies

Int. J. Comp. Sci. Issues (IJCSI)

A systematic literature review of agents applied in healthcare

J. Med. Syst.

Adapting activity recognition to a person with Multi-Classifier Adaptive Training

J. Ambient Intell. Smart Environ.

Adjustable autonomy: a systematic literature review

Artif. Intell. Rev.

Decision making method using fuzzy logic for autonomous agent navigation

Electr. J. Comp. Sci. Inf. Technol.

Defining tasks and actions complexity-levels via their deliberation intensity measures in the layered adjustable autonomy model

Dimensions of adjustable autonomy and mixed-initiative interaction

International Workshop on Computational Autonomy