TELK OMNIKA T elecommunication, Computing, Electr onics and Contr ol V ol. 24, No. 5, October 2026, pp. 1680 ∼ 1689 ISSN: 1693-6930, DOI: 10.12928/TELK OMNIKA.v24i5.27862 ❒ 1680 A unied articial intelligence-enabled intelligent home assistant r obot f or smart automation and safety Md. T aslim Arif 1,2 , Ab u Salyh Muhammad Mussa 1 , J ul J alal Al-Mamur Say or 3 , Md. Ashiqussalehin 1 , Md. Mehedi Hasan 4 , Mst Deloara Khushi 5 , Md. Moshiur Rahman 6 1 Department of Internet of Things and Robotics Engineering (IRE), Uni v ersity of Frontier T echnology Bangladesh (UFTB), Gazipur , Bangladesh 2 Department of Computer Science and Engineering (CSE), Daf fodil International Uni v ersity (DIU), Dhaka, Bangladesh 3 Department of Softw are Engineering (SWE), Daf fodil International Uni v ersity (DIU) Dhaka, Bangladesh 4 Department of General Education (GED), Uni v ersity of Frontier T echnology Bangladesh (UFTB), Gazipur , Bangladesh 5 Department of Cyber Security Engineering (CySE), Uni v ersity of Frontier T echnology Bangladesh (UFTB), Gazipur , Bangladesh 6 Department of Softw are Engineering (SE), Uni v ersity of Frontier T echnology Bangladesh (UFTB), Gazipur , Bangladesh Article Inf o Article history: Recei v ed Feb 12, 2026 Re vised Apr 28, 2026 Accepted May 25, 2026 K eyw ords: Articial intelligence Assisti v e robotics Home safety Internet of things Smart home object detection ABSTRA CT Intelligent home assistant systems are increasingly required to enhance safety , ef cienc y , and user con v enience in smart en vironments. This paper presents LUN A, a unied articial int elligence (AI)-enabled home assistant robot de- signed for smart automation and safety in indoor settings. The system inte grates autonomous na vig ation, Y ou Only Look Once v ersion 8 (Y OLOv8)-based object detection, gesture-based interaction, rea l-time g as and re hazard detection, mo- bile application control, and automated oor cleaning within a single lo w-cost embedded platform. Unlik e e xisting systems that focus on isolated functional- ities, the proposed approach combines perception, mobility , safety monitoring, and user interaction in a coordinated frame w ork suitable for real-w orld appli- cations, particularly for elderly and disabled users. The system is implemented using Arduino, Raspberry Pi, and ESP32 to enable ef cient sensing, processing, communication, and actuation. Experimental results demonstrate rob ust perfor - mance, achie ving 93.5% na vig ation accurac y , 91.2% object detection accurac y , 2.5 s emer genc y response time, 1.2 s control latenc y , and 88.7% cleaning ef - cienc y . These re sults indicate that LUN A is an ef fecti v e and practical solution for ne xt-generation smart home assistance. This is an open access article under the CC BY -SA license . Corresponding A uthor: Md. Ashiqussalehin Department of Internet of Things and Robotics Engineering (IRE) Uni v ersity of Frontier T echnology Bangladesh (UFTB) Kaliakair , Gazipur -1750, Bangladesh Email: ashiqussalehin0001@uftb .ac.bd 1. INTR ODUCTION The rapid adv ancement of articial intelligence (AI), robotics, and internet of things (IoT) technolo- gies has accelerated the de v elopment of intelligent syst ems capable of enhancing comfort, safety , and ef cienc y in smart homes [1]. In this scenario, home assistant robots are promising solutions to support daily acti vities, especially for elderly indi viduals and people with special needs [2]. AI robotics has rapidly e v olv ed to trans- form home automation and assisti v e technologies. The estimate for the year 2050 is that there will be 2.1 J ournal homepage: http://telk omnika.uad.ac.id Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA T elecommun Comput El Control ❒ 1681 billion elderly people aged 60 years and abo v e, while those o v er 80 years of age will reach 426 million [3]. This gro wing demographic identied the ur gent need to design af fordable intelligent systems that ensure re- liable assistance to the elderly and disabled while perpetuating their safety along with their independence in daily life and preser ving their quality of life. Research sho ws the healthcare industry and domestic en vironment require more assisti v e robots because of w orkforce shortages and the spik e in chronic diseases [4]. Standalone robotic solutions ha v e become more important than e v er because the corona virus disease 2019 (CO VID-19) pandemic demands equipment for limiting human contact during essential service deli v ery [5]. The mark et has seen substantial adv ancement in robotic home automation with the de v elopment of robotic v acuum cleaners and smart home de vices. These robotic systems w ork on indi vidual capabilities which hampers their v alue for multiple purposes. The e xtensi v e deplo yment of robotic systems f aces dif culties because the y are e xpensi v e to produce and cannot easily interact with other systems while also generating pri v ac y-related issues [6]. LUN A is a mul- tipurpose robot system since it is capable of e x ecuting multiple operations simultaneously , thereby deli v ering a whole solut ion to a v ariety of household needs. LUN A is able to e x ecute the detection of g as and res, as well as object recognition, to deli v er essential solutions for con v enience ass essment. Currently , robotics research is centered on the de v elopment of intelligent robots at a lo wer price. Literature has sho wn that cheaper robotics is able to f acilitate assistance for elderly patients, especially when implementing rehabilitation plans for areas with poor medical f acilities [7]. The usage of machine learning algorithms, such as Y ou Only Look Once (Y OLO), enables robots to e x ecute na vig ation with great precision, according to this technology’ s inte gration [8]. LUN A further impro v es other robotic de v elopments, bearing in mind its reduced-cost design for robotic technologies. This paper discusses LUN A ’ s de v elopment and implementation, focusing on its adv anced func- tionalities and af fordable design. This paper mak es the follo wing contrib utions: (i) a unied assisti v e robotic system inte grating na vig ation, intelligent perception, safety monitoring, human-robot interaction, and smart home control within a single platform; (ii) design and implementation of a multi-controller embedded architec- ture using Arduino, Raspberry Pi, and ESP32 for ef cient sensing, processing, communication, and actuation; and (iii) e xperimental e v aluation of system performance in terms of na vig ation accurac y , obje ct detection ac- curac y , emer genc y response time, smart home control latenc y , and cleaning ef cienc y in a real-w orld indoor en vironment. The proposed LUN A system inte grates i ntelligent home assistance functionalities, including auto- nomous na vig ation, AI-based object recognition, gesture interaction, real-time hazard detection, mobile ap- plication control, and automated cleaning within a lo w-cost platform. Unlik e e xisting system s that focus on isolated functionalities, the proposed frame w ork introduces a coordinated multi-controller architecture that en- ables real-time interaction between perception, mobility , and safety subsystems, ensuring synchronized and ef cient operat ion in dynamic indoor en vironments. This unied design allo ws seamless coordi n a tion among sensing, processing, and actuation modules, impro ving system responsi v eness and operational ef cienc y . Fur - thermore, the system i s v alidated through e xperimental e v aluation in real-w orl d indoor en vironments, demon- strating its practical applicability for elderly and disabled users. The remainder of this paper is structured as: recent related w ork is discussed in section 2. Section 3 comprehensi v ely analyzes the proposed method and implementation details of the system. Section 4 e v aluates the system’ s outcomes and performance, and section 5 presents the conclusion and future direction of this w ork. 2. LITERA TURE REVIEW Assisti v e robotics and smart hom e automation systems ha v e e v olv ed signicantly to support elderly and disabled indi viduals through intelligent monitoring [9], gesture-based control, en vironmental sensing [10], and AI-dri v en interaction. Prior studies ha v e e xplored v arious aspects of assisti v e technologies, including smart path follo wing for restaurants and hospitality [11], gesture recognition for home automation [12]-[14], socially assisti v e robotic frame w orks for elderly care [15]-[17], nursing-support robotic systems [18]-[20], healthcare- oriented automation platforms [21], [22], and IoT -based multipurpose home assistants with object detection capabilities [23]-[25]. These studies ha v e sho wn signicant de v elopments in the inte gration of v arious sensors, AI, and distant communication technology systems. Ho we v er , some of these systems may ha v e limitations when it comes to their multifunctionality , cost-ef fecti v eness, scalability , autonomous mobility , and thorough hazard detection. Some of the robots ha v e been more focused on addressing social interactions [15], health monitoring [21], and gesture-based controls [12], others ha v e addressed limitations such as system comple xity A unied articial intellig ence-enabled intellig ent home assistant r obot for ... (Md. T aslim Arif) Evaluation Warning : The document was created with Spire.PDF for Python.
1682 ❒ ISSN: 1693-6930 [18], [23]. Ne v ertheless, the use of multi-robot systems for emotional and household support w as also e xplored [26]. Y et, the necessity of a cost-ef fecti v e, inte grated, and safe intelligent assistant remains a requirement. Henceforth, the LUN A system proposed in the paper bridges the g aps, inte grating a cost-ef fecti v e and safe frame w ork that implements object recognition using AI, autonomous na vig ation, human follo wing, en viron- mental hazard detection, a cleaning feature, and mobile-based remote control. T o well illustrat e the g ap in the e xisting system and the proposed system, T able 1 summarizes the comparison of the current w ork and proposed LUN A robot system. As sho wn in the table comparing the LUN A proposed robot and the e xisting w ork, the intelligent robot proposed is observ ed to inte grate object recognition using AI, autonomous na vig ation, human follo wing, g as and re hazard detection, a cleaning feature, and mobile remote control within a unied and cost-ef fecti v e frame w ork. T able 1. Summary of related w orks compared with the proposed LUN A system K e y aspect [12] [15] [18] [21] [23] LUN A Gesture recognition Y es Limited No Y es Y es Y es AI-based object detection No P artial No P artial Y es (Y OLOv8) Custom Y OLOv8 Autonomous na vig ation No No Y es Limited No Y es Health/social assistance No Y es Y es Y es P artial Y es Gas/re detection No No No Emer genc y only Y es Y es Remote app control No Y es Limited Y es Y es Y es Cleaning functionality No No No No No Y es Cost-ef fecti v e inte grated design Limited Costly Prototype Needs v alidation Moderate Y es 3. METHODOLOGY The proposed approach is an intelligent embedded platform using Arduino, Raspberry Pi, and ESP32 to pro vide sensing, automation, and remote communication. Arduino is used as the main contr o l ler , which controls g as and ame detection using sensors, turns on the b uzze r , and sends emer genc y calls with a global system for mobile communications (GSM) module. It controls v acuum cleaning, supports human-follo wing mobility with motor dri v ers, infrared (IR), and ultrasonic sensors, and pro vides Bluetooth-based control. The Raspberry P i is us ed to enhance the syst em with a camera-based image and gesture recognition using Y OLOv8 and pro vides serv o-based actuations. The ESP32 is responsible for inb uilt mobile app inte gration, en viron- mental monitoring through the digital humidity a n d temperature (DHT) sensor , and remote control of output de vices. These all together support a coherent multifunctional smart automation system. 3.1. System achitectur e LUN A is intended to be a multifunctional IoT -based and AI-assisted assistant robot. In this de vice, v arious sensing, processing, and e x ecuting modules are inte grated. The system includes an embedded con- troller module, en vironment sensing modules , a na vig ation module, and communication modules. The system architecture of the proposed system is sho wn in Figure 1. Figure 1. System architecture of the proposed intelligent home assistant robot sho wing sensing, processing, and communication modules TELK OMNIKA T elecommun Comput El Control, V ol. 24, No. 5, October 2026: 1680–1689 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA T elecommun Comput El Control ❒ 1683 3.2. Hard war e and softwar e components A comprehensi v e list of t he essential hardw are and softw are components, as well as the v arious cost s, is pro vided in T able 2. This table reintroduces and combines v arious microcontrollers, sensors, actuators, as well as communication modules, with a focus placed on softw are. Inte gration of embedded softw are and mobile applications enhances their usa b i lity , while AI-po wered object detection will further strengthen the capability of the system. This functional prototype cost can be esti mated to be around 177.49 United States dollar (USD), thus making it cost-ef fecti v e. Ho we v er , in the case of the operational prototype with similar congurations, it may increase. T able 2. Detailed list of hardw are components used in the proposed system with corresponding costs (USD) Component cate gory (Items included) Cost (USD) Microcontrollers (Raspberry Pi 5, ESP32, Arduino Uno) 130.34 Sensors (ultrasonic, IR, MQ2, ame) 5.50 Actuators (direct current (DC) motors, 2 serv o motors) 5.00 Communication modules (GSM, wireless delity (W i-Fi)) 16.40 Other components (Pi camera, cables, chassis) 20.25 T otal hardw are cost 177.49 4. EXPERIMENT AL SETUP AND RESUL TS LUN A w as tested in a c o nt rolled indoor e n vir on m ent designed t o replicate real-w orld home condi- tions. The e xperimental setup consisted of a 3 m × 3 m area furnished with household objects to e v aluate na vig ation and object detection performance. As sho wn in Figure 2, a human subject w as in v olv ed in trials to assess human-follo wing capability . Simulated g as leaks and re hazards were introduced to test the emer genc y alert system, and W i-Fi-enabled home appliances were used to v erify remote control functionalities. Each feature w as tested under multiple scenarios to v alidate performance consistenc y and ef cienc y . Figure 2. P ath na vig ation and human-follo wing performance of the robot in a controlled indoor en vironment 4.1. Human f ollo wing and na vigation LUN A uses ultrasonic and IR sensors to detect human presence and track the user . The process is described in (1): Distance ultrasonic = Speed of sound × T ime of ight (1) This equation calculates the distance between the robot and the user based on the time it tak es for the ultrasonic signal to tra v el. The system uses this distance to detect human presence. Angle IR = arctan ∆ y ∆ x (2) The IR sensor detects the user’ s posi tion by calculating the angle based on the relati v e change in position. The robot then tracks and follo ws the user based on this angle and the distance from the sensors. A unied articial intellig ence-enabled intellig ent home assistant r obot for ... (Md. T aslim Arif) Evaluation Warning : The document was created with Spire.PDF for Python.
1684 ❒ ISSN: 1693-6930 4.2. Object detection and gestur e-based assistance LUN A inte grates both object detection and gesture-based assistance to enhance its functionality for smart home automation, the processes are as follo ws. 4.2.1. Object detection using Y OLOv8 model LUN A captures an image using its onboard camera and processes the image with the Y OLOv8 model to detect household objects. The detection process is represented as (3). Object detected = Y OLOv8 ( I input ) (3) Here, I input represents the captured image, and the model processes the image to detect objects, resulting in Object detected , which are the identied household objects. After processing the image, LUN A retrie v es and recognizes the detected objects, assisting in locating them within the en vironment. Figure 3(a) sho ws the object detection outcome. 4.2.2. Gestur e-based assistance using Raspberry Pi and Pi camera LUN A emplo ys a R aspberry Pi and Pi Camera to capture and recognize hand gestures for hands-free home control. The captured image I gesture is processed using a gesture recognition model, e xpressed as (4). Gesture detected = Gesture Model ( I gesture ) (4) where the identied gesture triggers the corresponding action, such as acti v ating alarms or controlling appli- ances. This approach enables intuiti v e and touchles s interaction. Object detection and gesture recognition performance is illustrated in Figures 3(a) and (b), respecti v ely . (a) (b) Figure 3. Object detection and gesture recognition results: (a) object detection results using the Y OLOv8 model for real-time identication of household objects and (b) gesture recognition for user interaction using camera-based detection TELK OMNIKA T elecommun Comput El Control, V ol. 24, No. 5, October 2026: 1680–1689 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA T elecommun Comput El Control ❒ 1685 4.3. Gas and r e detection LUN A monitors en vironmental sensors to detect g as and re hazards. The detection process can be described as follo ws. 4.3.1. Gas detection using MQ2 sensor The MQ2 g as sensor detects the presence of g ases by measuring the resistance change in the sensor . The g as concentration ( C g as ) can be calculated based on the ratio of the resistance of the load resistor ( R load ) to the resistance of the sensing material ( R sensing ): C g as = R load R sensing (5) Here, R load is the resistance of the load resistor , R sensing is the resistance of the sensing material. If the g as concentration e xceeds a predened threshold, it indicates the presence of a g as leak. 4.3.2. Fir e detection using ame sensor The ame sensor detects re by measuring the intensity of the light emitted by the ame. The output v oltage ( V ame ) from the sensor can be calculated as (6): V ame = I ame × R sensor (6) Here V ame is the v oltage output from the ame sens or , I ame is the current generated by the ame’ s light intensity , R sensor is the resistance of the ame sensor . If the output v oltage e xceeds the threshold, it indicates the presence of re. The re detection process is demonstrated in Figure 4(a), whe re the ame sensor detects the presence of a ame on the robot. When either g as or re is detected, the system immediately triggers an alert to notify the user , as sho wn in Figure 4(b). In addition, the system sends notications through SMS and phone calls, as illustrated in Figure 4(c). This ensures that the user is promptly notied of the detected hazard. Figure 4. Emer genc y detection system for g as and re hazards with real-time alert response: (a) ame detection by the robot, (b) alert notication displayed on the smartphone, and (c) SMS notications for detected g as and re hazards 4.4. Smart home contr ol via mobile app LUN A helps the user control and interact with their respecti v e smart objects such as lights, f ans, and appliances via a bespok e mobile application, and at the same time, users will be able to enjo y the ability to observ e their respecti v e en vironments in real-time. As depicted in Figure 5, the maj o r pages of the de v eloped mobile application will be able to of fer easy interaction with the robot’ s status, location, and conditions. A unied articial intellig ence-enabled intellig ent home assistant r obot for ... (Md. T aslim Arif) Evaluation Warning : The document was created with Spire.PDF for Python.
1686 ❒ ISSN: 1693-6930 Figure 5. Mobile application interf ace for monitoring, control, and interaction with the robot 4.5. A utomated oor cleaning LUN A ’ s v acuum cleaning system is po wered by an Arduino Uno with a DC motor coupled with a f an to pro vide automated oor cleaning. The suction created from the DC motor sucks the dust and deposits it into a compartment. W i th the inclusion of a camera, the t rolle y can e x ecute na vig ation with high precision. Moreo v er , this trolle y can also be controlled using an y mobile application. The system resulted in an a v erage cleaning ef cienc y of 88.7%. 4.6. Comparati v e analysis of r esult T o highlight its ef fecti v eness, LUN A is e v aluated ag ainst e xisting smart assistant and static IoT -based home automation systems. Unlik e x ed sensor -based solutions, LUN A of fers dynamic en vironmental inter - action through mobilit y . It achie v es 91.2% object detection accurac y using Y OLOv8 and 93.5% na vig ation accurac y , outperforming comparable lo w-cost robotic systems reported in prior studies. The system records a f ast emer genc y response time of 2.5 seconds and a smart home control latenc y of 1.2 seconds, demonstrating superior responsi v eness compared to con v entional static detection systems. T able 3 represents performance e v aluation of our de v eloped intelligent home assistant robot LUN A. In general, the results conrm that LUN A pro vides ef cient and intelligent home automation by inte grating AI and IoT technologies. T able 3. Performance e v aluation of our de v eloped intelligent home assistant robot LUN A Metric V alue Na vig ation accurac y 93.5% Object detection (Y OLOv8) 91.2% Emer genc y response time 2.5 s Home control latenc y 1.2 s Cleaning ef cienc y 88.7% 5. CONCLUSION The proposed LUN A system presents a unied home assistant robot that inte grates na vig ation, per - ception, safety monitoring, and user interaction within a single lo w-cost platform. The system successfully combines multiple assisti v e functionalities, including object detection, human-follo wing, hazard detection, and smart home control, demonstrating its applicability for real-w orld indoor en vironments. The e xperimental re- sults conrm that the proposed approach pro vides reliable performance and practical applicability , particularly TELK OMNIKA T elecommun Comput El Control, V ol. 24, No. 5, October 2026: 1680–1689 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA T elecommun Comput El Control ❒ 1687 for elder ly and disabled users. Despite its ef fecti v eness, the system has certain limitations related to process- ing latenc y , po wer consumption, and multi-de vice comple xity . Future w ork will focus on optimizing system architecture, impro ving real-time performance using adv anced AI techniques, and incorporating cloud-based inte gration for enhanced scalability and data management. REFERENCES [1] A. Anubha, D. Narang, and A. Raj, “ AIoT -Enabled Smart Homes Solution to Security , Comfort, and Ener gy Ef cienc y , ” in Advances in Computational Intellig ence and Robotics. IGI Global Scientic Publishing , pp. 239–268, May 08, 2025, doi: 10.4018/979-8- 3693-8547-0.ch008. [2] P . Harmo, T . T aipalus, J. Knuuttila, J. V allet, and A. Halme, “Needs and solutions - Home automation and s ervice robots for the elderly and disabled, ” in 2005 IEEE/RSJ International Confer ence on Intellig ent Robots and Systems, IEEE , 2005, pp. 3201–3206, doi: 10.1109/IR OS.2005.1545387. [3] WHO, “ Agei ng and health, ” W orld Health Or ganization . Accessed: Jan. 23, 2026. [Online]. A v ai lable: https://www .who.int/ne ws- room/f act-sheets/detail/ageing-and-health [4] D. Silv era-T a wil, “Robotics in Healthcare: A Surv e y , ” SN Computer Science , v ol. 5, no. 1, p. 189, 2024, doi: 10.1007/s42979-023- 02551-0. [5] Y . Shen et al. , “Robots under CO VID-19 P andemic: A Comprehensi v e Surv e y , ” IEEE Access , v ol. 9, pp. 1590–1615, 2021, doi: 10.1109/A CCESS.2020.3045792. [6] O. Aouedi et al. , ”A Surv e y on Intelligent Internet of Things: Applications, Security , Pri v ac y , and Future Directions, ” in IEEE Communications Surve ys & T utorials , v ol. 27, no. 2, pp. 1238-1292, April 2025, doi: 10.1109/COMST .2024.3430368. [7] M. J. Johnson, S. K e yv anian, and R. J. Mendonca, “T o w ard inclusi v e rehabilitation robots, ” in Rehabilitation Robots for Neur or e- habilitation in High-, Low-, and Middle-Income Countries , 2024, pp. 471–498. doi: 10.1016/B978-0-323-91931-9.00032-3. [8] M. L. Ali and Z. Zhang, “The Y OLO Frame w ork: A Comprehens i v e Re vie w of Ev olution, Applications, and Benchmarks in Object Detection, ” Computer s , v ol. 13, no. 12, p. 336, Dec. 2024, doi: 10.3390/computers13120336. [9] A. Logeshw ar , R. M. Manikandan, R. P ar v es h, A. R. Sol aiappan, and L. Anju, “Smart Home Robotic Companion with AI-Dri v en Personalized Care for Elderly Assistance, ” in Pr oceedings of the International Confer ence on Advanced Resear c h in Electr onics and Communication Systems (ICARECS 2025) , 2025, pp. 322–332, doi: 10.2991/978-94-6463-754-0 29 . [10] C. Ok onkw o and I. A w olusi, “En vironmental sensing in autonomous construction robots: Applicable technologies and systems, ” A utomation in Construction , v ol. 172, p. 106075, Apr . 2025, doi: 10.1016/j.autcon.2025.106075. [11] J. J. A.-M. Sayor , N. T . Shishir , B. B. Barmon, S. Ahemed, Nurjahan, and M. Whaiduzzaman, “RoboServ e: Design and Imple- mentation of a Robotic Agent for Serving F ood at Restaurants, ” in Pr oceedings of the 3r d International Confer ence on Computing Advancements , Ne w Y ork, NY , USA, Oct. 2024, pp. 588–597, doi: 10.1145/3723178.3723256. [12] B. I. Alabdullah et al. , “Smart Home Automation-Based Hand Gesture Recognition Using Feature Fusion and Recurrent Neural Netw ork, ” Sensor s , v ol. 23, no. 17, p. 7523, Aug. 2023, doi: 10.3390/s23177523. [13] D. J. Mohan, A. Je yapal, D. Chandrasekar , K . S. Babar , and D. K. Ra vi, “Home Automation Using Gesture Recognition, ” in AIP Confer ence Pr oceedings , 2025, v ol. 3279, p. 20028, doi: 10.1063/5.0263203. [14] N. M. Bhiri, S. Ameur , I. Je gham, I. Alouani, and A. Ben Khalif a, “2MLMD: Multi-moda l Leap Motion Dataset for Home Au- tomation Hand Gesture Recognition Systems, ” Ar abian J ournal for Science and Engineering , v ol. 50, no. 10, pp. 7511–7535, Aug. 2024, doi: 10.1007/s13369-024-09396-6. [15] M. Luperto et al. , “Inte grating Social Assisti v e Robots, IoT , V irtual Communities and Smart Objects to Assist at-Home Indepen- dently Li ving Elders: the Mo v eCare Project, ” Internat ional J ournal of Social Robotics , v ol. 15, no. 3, pp. 517–545, Feb . 2022, doi: 10.1007/s12369-021-00843-0. [16] A. Poulsen, O. K. Burmeister , and D. T ien, “ A ne w design approach and frame w ork for elderly care robots, ” in A CIS 2018 - 29th A ustr alasian Confer ence on Information Systems , 2018, doi: 10.5130/acis2018.cj. [17] B. R. K onk epudi and T . K. R. K. Rao, “Enhancing Elderly Care: Microservices-Enabled IoT Models for Ad- dressing and Monitoring Needs, ” F r ontier s in Health Informatics , v ol. 13, no. 3, 2024. [Online]. A v ailable: https://healthinformaticsjournal.com/inde x.php/IJMI/article/vie w/79/82 [18] C. Ohneber g et al. , “ Assisti v e robotic systems in nursing care: a scoping re vie w , ” BMC Nur sing , v ol. 22, no. 1, Mar . 2023, doi: 10.1186/s12912-023-01230-y . [19] K. Y asak and G. S ¸ ahb udak, “Digital T ools and Articial Intelligence in Postoperati v e Nursing Care for Older Adults: A Biblio- metric Analysis, ” Sakarya ¨ Univer sitesi Holistik Sa ˘ glık Der gisi, no. Advanced Online Publication , pp. 163–178, Dec. 2025, doi: 10.54803/sauhsd.1812217. [20] Y . Ito et al. , “ Articial intelligence in nursing support for patients: A rapid re vie w , ” International J ournal of Medical Informatics , v ol. 204, p. 106087, Dec. 2025, doi: 10.1016/j.ijmedinf.2025.106087. [21] G. Dilip et al. , “ Articial Intelligence-Based Smart Comrade Robot for Elders Healthcare with Strait Rescue System, ” J ournal of Healthcar e Engineering , v ol. 2022, pp. 1–12, Jan. 2022, doi: 10.1155/2022/9904870. [22] J. Banik, “De v elopment of an Automated Simulation-Based AI Inte grated Emotional-A w are Robotic Assistant for P atient Support and Emer genc y Interv ention in Hospitals, ” Thesis, Jyv ¨ ask yl ¨ an ammattik ork eak oulu, Jyv ¨ ask yl ¨ a, Finland, 2025. [Online]. A v ailable: https://urn./URN:NBN::amk-2025121536555 (accessed: Apr . 14, 2026) [23] A. Andriella, C. T orras, C. Abdelnour , and G. Alen y ` a, “Introducing CARESSER: A frame w ork for in situ learning robot social assistance from e xpert kno wledge and demonstrations, ” User Modeling and User -Adapted Inter action , v ol. 33, no. 2, pp. 441–496, Mar . 2023, doi: 10.1007/s11257-021-09316-5. [24] I. Y asri, A. Al-Al if, and F . Pi rw anhadi, “ A Security Home System Feature as a P art of Home Assistant through IoT Plat- form, ” J ournal of T elecommunication, Electr onic and Computer Engineering (JTEC) , v ol. 17, no. 2, pp. 15–21, Jun. 2025, doi: 10.54554/jtec.2025.17.02.002. [25] Y . Khomenk o and S. Babiche v , “Modular IoT Architecture for Monitoring and Control of Of ce En vironments Based on Home A unied articial intellig ence-enabled intellig ent home assistant r obot for ... (Md. T aslim Arif) Evaluation Warning : The document was created with Spire.PDF for Python.
1688 ❒ ISSN: 1693-6930 Assistant, ” Internet of Things , v ol. 6, no. 4, p. 69, No v . 2025, doi: 10.3390/iot6040069. [26] R. Barber et al. , “ A Multirobot System in an Assisted Home En vironment to Support the Elderl y in Their Daily Li v es, ” Sensor s , v ol. 22, no. 20, p. 7983, Oct. 2022, doi: 10.3390/s22207983. BIOGRAPHIES OF A UTHORS Md. T aslim Arif recei v ed his B.Sc. and is currentl y pursuing his M.Sc. in Internet of Things and Robotics Engineering from the Uni v ersity of Frontier T echnology , Bangladesh. He is currently a Lecturer in the Department of Computer Science and Engineering at Daf fodil International Uni v ersity , Bangladesh. His research interests include embedded systems, IoT , robotics, and AI- inte grated intelligent systems, with a focus on intelligent automation and c yber -ph ysical systems . He can be contacted at email: 1801030@iot.uftb .ac.bd. Ab u Salyh Muhammad Mussa recei v ed his B.Sc. in IoT and Robotics Engineering from the Uni v ersity of Frontier T echnology , Bangladesh, and is currently pursuing his M.Sc. in the same eld. He is w orking as an Associate Engineer (IoT and T elematics ) at Bongo IoT , Bangladesh, con- trib uting to AI-po wered telematics and smart automation solutions. Hi s research interests include em- bedded systems, IoT , robotics, AI, and intelligent c yber -ph ysical sys tems, with a focus on real-time automation and smart urban technologies. He can be contacted at email: ab usalyh43@gmail.com. J ul J alal Al-Mamur Say or is a Lecturer in the Department of Softw are Engineering at Daf fodil International Uni v ersity . He completed his B.Sc. (Engr .) in IoT and Robotic s Engineering from UFTB, securing the rst position i n his f aculty , and is currently pursuing an M.Sc. in Robotics and Mechatronics Engineering at the Uni v ersity of Dhaka. His research interests encompass deep neural netw orks, liquid neural netw orks, vision amba, and the IoT . Throughout his career , he has been recognized with se v eral teaching and academic a w ards for his contrib utions and achie v ements. He can be contacted at email: mamur .swe@diu.edu.bd. Md. Ashiqussal ehin obtained his M.Sc. in Computer Science and Information T echnol- ogy (CSIT) in 2021 and his B.Sc. in Computer Science and Engineering (CSE) in 2019 from P atu- akhali Science and T echnology Uni v ersity , Bangladesh. He is currently serving as a Lecturer in the Department of IoT and Robotics Engineering at the Uni v ersity of Frontier T echnology , Bangladesh. Pre viously , he w ork ed as a Lecturer in the Department of Computer Science and Engineering at Green Uni v ersity of Bangladesh and Uttara Uni v ersity . He also w ork ed as a Softw are Engineer at Mangro v e Systems Banglade sh. His research i nterests include IoT , robotics and automation, machine learning (ML), c yber -ph ysical systems, and unmanned aerial v ehicles (U A Vs). He can be contacted at email: ashiqussalehin0001@uftb .ac.bd. Md. Mehedi Hasan recei v ed his B.S. (Hons) de gree in Mathematics and M.S. in Pure Mathematics from the Uni v ersity of Dhaka. He is w orking as a Lecturer in the Department of General Education, Uni v ersity of Frontier T echnology , Bangladesh. His research interest is on mathematical biology , agrosciences, mathematical programming and dif ferent areas of operation research optimiza- tion and ML. He can be contacted at email: mehedi0001@uftb .ac.bd. TELK OMNIKA T elecommun Comput El Control, V ol. 24, No. 5, October 2026: 1680–1689 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA T elecommun Comput El Control ❒ 1689 Mst Deloara Khushi is current ly a Lecturer in the Department of Cyber Security Engi- neering, Uni v ersity of Frontier T echnology , Bangladesh. W ith an M.Sc. in Computer Science and Engineering from Jag annath Uni v ersity (JnU), Bangladesh, and a B.Sc. in the same eld from Hajee Mohammad Danesh Science and T echnology Uni v ersity . Her current focus lies at ML, particularly deep learning techniques, and the critical eld of c yber security . She can be contacted at email: khushi0001@uftb .ac.bd. Md. Moshiur Rahman obtained his M.Sc. in Softw are Engineering from the Institute of Information T echnology (IIT), Uni v ersity of Dhaka in 2022, and his B.Sc. in Computer Science and Engineering from P atuakhali Science and T echnology Uni v ersity (PSTU) in 2019. He w ork ed as a lecturer in the Department of Computer Science and Engineering at Green Uni v ersity of Bangladesh (GUB) from February 2, 2021, to August 30, 2023. Currently , he is a lecturer in the Department of Softw are Engineering at Uni v ersity of Frontier T echnology , Bangladesh (UFTB), starting from January 1, 2024. His research interests include softw are engineering, ML, data science, transportation modeling and pattern recognition. He can be contacted at email: moshiur0001@uftb .ac.bd. A unied articial intellig ence-enabled intellig ent home assistant r obot for ... (Md. T aslim Arif) Evaluation Warning : The document was created with Spire.PDF for Python.