INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
Latency on Motion Synchronization in Game Engine-Driven Digital  
Twin Robotic Arms: Challenges and Techniques  
Mohamad Lutfi Dolhalit., Nazreen Abdullasim*., Aiman Hakim Azahari., Mohd Khalid Mokhtar  
Fakulti Teknologi Maklumat Komunikasi, Universiti Teknikal Malaysia Melaka, Melaka, Malaysia  
DOI: https://dx.doi.org/10.47772/IJRISS.2025.91200037  
Received: 11 December 2025; Accepted: 18 December 2025; Published: 31 December 2025  
ABSTRACT  
Ensuring precise and stable motion synchronization in game engine-driven digital twin robotic arms is  
challenging due to real-time data transmission delays. Various strategies—such as predictive modeling,  
network optimization, and robotic arm control techniques—have been proposed to address latency-aware  
motion synchronization. However, the absence of recent review papers in this area limits researchers’ ability to  
identify the most effective solutions for minimizing latency and enhancing system reliability. To fill this gap,  
we conducted a systematic literature review (SLR) to identify, analyze, classify, and summarize existing  
latency reduction techniques. A total of 125 research studies from reputable sources were reviewed to uncover  
recent trends in the field. We developed a taxonomy to group the identified methods based on common  
characteristics and provided concise summaries of each approach. Furthermore, this study outlines key  
research challenges and suggests future directions for improving synchronization accuracy. The findings offer  
a comprehensive and structured overview of existing solutions and serve as a valuable reference for  
researchers and practitioners aiming to advance real-time digital twin applications through game engine-based  
visualization.  
Keywords: Latency, Motion Synchronization, Digital Twin, Game Engine, Data Transmission, Robotic  
INTRODUCTION  
Digital Twin (DT) technology has become a key enabler in various industries by providing real-time  
monitoring, prediction, and control through virtual replicas of physical systems (Parashar & Gameti, 2024; Tao  
et al., 2019). Game engines, such as Unity and Unreal Engine, are increasingly adopted for digital twin  
development due to their advanced simulation and visualization capabilities, especially in robotic applications  
(Lin et al., 2022; Haque et al., 2023).  
However, a major challenge in this domain is ensuring accurate and real-time synchronization between the  
virtual model and its physical counterpart, particularly in robotic systems where latency can lead to motion  
discrepancies and reduced reliability in tasks like teleoperation and human-robot collaboration (Zheng et al.,  
2021; Zhang et al., 2022). In recent years, researchers have proposed various solutions to mitigate latency,  
including predictive algorithms (Wang et al., 2022), interpolation techniques (Ali et al., 2023), and network-  
level optimizations, such as edge computing and communication protocol enhancements (Cheng et al., 2023).  
While these methods have improved system responsiveness, they still face limitations, such as reduced  
accuracy under variable conditions, dependency on network infrastructure, and a lack of adaptability to  
complex real-time interactions. Moreover, there is a scarcity of comprehensive reviews that critically analyse  
and compare these techniques within the specific context of game engine-driven digital twin robotics (Lee et  
al., 2023; Hussain et al., 2022). To address this gap, this study conducts a systematic literature review (SLR)  
with three main contributions: (1) identifying and classifying existing latency mitigation techniques, (2)  
evaluating their effectiveness in real-time robotic applications involving game engines, and (3) proposing a  
taxonomy that organizes these methods based on shared characteristics. Rather than introducing a new  
technique, this work enhances current understanding by synthesizing and summarizing recent studies, offering  
a structured reference for researchers and developers. The rest of the paper is organized as follows: Section 2  
details the SLR methodology, Section 3 presents the taxonomy and classification of latency mitigation  
Page 405  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
methods, Section 4 discusses findings and ongoing challenges, and Section 5 concludes with future research  
directions.  
Related Work  
Method  
To ensure a high-quality and substantive literature review, a systematic literature search was conducted,  
dividing the identification of relevant literature into three key steps: Literature Search, Literature Selection,  
and Qualitative Analysis.  
LITERATURE SEARCH  
Following the systematic literature review (SLR) method proposed by Webster and Watson (2002), we began  
by identifying relevant studies on digital twin technology, with a specific focus on latency-aware  
synchronization in game engine-driven robotic applications. The initial search was conducted on Google  
Scholar using broad keywords, such as "digital twin", "game engine", "robotics", and "latency", combined with  
Boolean operators (e.g., "digital twin" AND "latency", "game engine" AND "robotic arm", "digital twin" AND  
"predictive synchronization"). We then refined the search using more focused terms including "motion  
synchronization", "interpolation", "latency compensation", "predictive algorithm", "edge computing", and  
"real-time control", to capture specific techniques and challenges addressed in recent studies. Academic  
databases, such as IEEE Xplore, SpringerLink, ScienceDirect, and EBSCOhost were queried, and only peer-  
reviewed journal articles or conference proceedings published in English between 2018 and 2024 were  
considered. Inclusion criteria required the presence of relevant keywords in the title, abstract, or keywords  
section. After screening titles and abstracts, we shortlisted 125 papers and documented metadata, such as DOI,  
citation count, publication venue, and research domain. Additionally, the "Connected Papers" tool was used to  
trace influential and closely related studies based on co-citation networks, ensuring the coverage of both  
foundational and recent contributions in the field.  
Literature Selection  
To ensure a manageable and high-quality selection of studies for this review, a systematic filtering process was  
implemented. Initially, duplicate records retrieved from multiple databases were removed. Titles and abstracts  
were then screened to exclude publications that were unrelated to digital twin technologies or lacked relevance  
to latency-aware motion synchronization in robotic systems. However, papers featuring notable application  
examples were selectively retained. To ensure academic rigor, the quality of publications was assessed using  
the Verband der Hochschullehrer für Betriebswirtschaft (VHB) JOURQUAL ranking system, which is widely  
adopted in the information systems and business research communities. Preference was given to journal  
articles ranked B and C, while exceptions were made for highly cited works that demonstrated significant  
relevance regardless of their ranking. As a result, fifteen core papers were selected for in-depth analysis. These  
papers were then systematically classified into three thematic streams implementation frameworks,  
synchronization techniques, and predictive algorithms based on their primary contributions to latency  
mitigation in digital twin robotic systems. This categorization is presented in Table 1, which maps each study’s  
methodological focus, application domain, and observed trade-offs. The structured classification highlights  
both well-established solutions and underexplored areas, such as the absence of integrated hybrid AI-edge  
synchronization frameworks.  
Qualitative Analysis  
Before presenting the results, we first outline the approach used to systematically analyse the selected  
literature. To ensure a comprehensive and structured evaluation, we adopted the qualitative analysis method  
proposed by Wolfswinkel et al. (2013), which consists of three stages: Open Coding, Axial Coding, and  
Selective Coding. In the Open Coding phase, we extracted key concepts, methods, and application themes  
related to latency mitigation in digital twin robotic systems. During the Axial Coding stage, we grouped  
similar studies based on methodological focus and their contribution to latency reduction. Finally, in the  
Page 406  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
Selective Coding phase, these groupings were refined into three core thematic streams: implementation  
frameworks, synchronization techniques, and predictive algorithms. These categories form the basis for the  
analytical classification presented in Table 1, which maps the focus, techniques, and trade-offs across the  
selected studies to reveal current trends and research gaps in latency-aware digital twin applications.  
Table 1. Analytical classification of latency mitigation studies in digital twin robotic systems across  
implementation frameworks, synchronization techniques, and predictive algorithms  
Stream  
Study  
Focus  
Method  
Trade-offs / Gaps  
Implementati Garrido-Hidalgo Predictive  
Data-driven models Limited  
scalability  
in  
on  
et al., 2022  
Maintenance via DTs  
heterogeneous systems  
Frameworks  
Mrzyk  
2023  
et  
al., Modular  
architecture  
DT Flexible  
IT- No real-time  
loop  
feedback  
infrastructure  
Synchronizat  
ion  
Tan et al., 2023  
DT  
formulation  
sync  
problem Sync protocols + Lack  
of  
performance  
framework  
benchmarks  
Techniques  
Shen et al., 2025 Real-time update for Update algorithms  
DTs  
Not evaluated under high-  
load scenarios  
Predictive  
Algorithms  
Liu et al., 2022  
ML-based Predictive Convolutional  
Maintenance Autoencoders  
High  
demand  
computational  
Polese  
2018  
et  
al., Latency prediction in ML  
+
Traffic Needs  
adaptability  
real-time  
5G networks Routing  
METHODOLOGY  
Digital Twin Technology Overview  
Digital Twin (DT) technology enables real-time monitoring, analysis, and optimization by creating digital  
replicas of physical systems (Tao et al., 2019; Fuller et al., 2020). A digital twin is a dynamic model that  
continuously interacts with its physical counterpart through three key components: the digital model, the  
physical entity, and the data connection (Boschert & Rosen, 2016). Efficient data transmission and  
synchronization within this system are essential for informed decision-making and operational accuracy (Yin  
et al., 2022; Kaur & Mishra, 2022). The concept of the digital twin was originally introduced by Michael  
Grieves in 2002 within the Product Lifecycle Management (PLM) paradigm (Grieves & Vickers, 2017) and  
has since evolved into a transformative tool across multiple domains. Digital twins have been widely adopted  
in industries, such as urban planning, manufacturing, healthcare, and aerospace, enabling applications like real-  
time asset monitoring, predictive maintenance, and virtual testing (Rehman et al., 2022; Liu et al., 2020; Xu et  
al., 2021; Jones et al., 2020). In robotics, digital twins facilitate state estimation, behavior prediction, and  
lifecycle analysis, contributing to increased system autonomy and reduced downtime (Zhang et al., 2021; Lee  
et al., 2014). This capability supports complex simulations and scenario testing without disrupting real-world  
operations, making digital twins increasingly valuable in the development of smart, interconnected systems  
(Alam, 2023). Table 2 provides an overview of Digital Twin Technology and its key components.  
Table 2. Overview of digital twin technology and its key components  
Aspect  
Details  
Definition  
A digital twin is a dynamic and continuously evolving virtual model that precisely  
mirrors a physical entity, enabling real-time monitoring, simulation, and optimization  
of its performance across its entire lifecycle (Modoni et al., 2023; Ünal et al., 2023).  
Page 407  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
It consists of three main components: the physical entity, the virtual model, and the  
data connections that allow for seamless interaction between them (Boschert &  
Rosen, 2016; Fuller et al., 2020). By utilizing real-time data, digital twins not only  
reflect the current state of their physical counterparts but also predict their future  
behaviour, ultimately improving decision-making and operational efficiency (Tao et  
al., 2019; Lee et al., 2014; Grieves & Vickers, 2017). This capability has led to their  
increasing adoption across sectors, such as manufacturing, aerospace, smart cities,  
and healthcare, due to their ability to enhance operational insight, enable predictive  
maintenance, and support complex simulations without disrupting real-world  
operations (Liu et al., 2020; Jones et al., 2020).  
Origins  
The concept of digital twins was first introduced by Michael Grieves in 2002 during a  
presentation at the University of Michigan, where he discussed a virtual model that  
mirrors a physical counterpart within Product Lifecycle Management (PLM) (Grieves  
& Vickers, 2017). Earlier applications of similar concepts can be traced back to  
NASA during the Apollo missions in the 1960s, where simulators were used to model  
spacecraft conditions.  
Applications  
Smart manufacturing  
Digital twin technology enhances real-time monitoring, predictive maintenance, and  
system optimization in smart manufacturing, significantly reducing machine  
downtime, energy consumption, and production errors (Tao et al., 2022; Liu &  
Zhang, 2021; Qi & Tao, 2018). In the aerospace industry, digital twins are used for  
performance monitoring, structural health diagnostics, and mission simulation,  
thereby improving operational safety and reducing maintenance costs (Fuller et al.,  
2020; Glaessgen & Stargel, 2012). Key applications include product lifecycle  
management, dynamic design optimization, and intelligent control systems (Grieves  
& Vickers, 2017; Boschert & Rosen, 2016). Despite these advantages, several  
challenges persist. These include high implementation costs, data interoperability  
issues across platforms, cybersecurity risks due to continuous data exchange, and the  
need for standardized frameworks (Khan et al., 2021; Ünal et al., 2023). Conversely,  
ongoing research presents opportunities, such as integrating AI for autonomous  
decision-making, utilizing 5G/6G for ultra-low latency communication, and  
expanding the use of digital twins in emerging domains like personalized medicine  
and urban digital infrastructure (Lu et al., 2020; Jones et al., 2020; Alam, 2023).  
Healthcare  
Digital twins in healthcare facilitate personalized treatments through patient-specific  
simulations, significantly enhancing diagnostic accuracy, treatment planning, and  
clinical decision-making—thereby advancing the field of precision medicine (Kahn &  
Lentz, 2024; Rojas & Gutiérrez, 2024; Bruynseels et al., 2018). Key applications  
include surgical simulations, which allow for preoperative rehearsals and risk  
assessment; treatment planning, where simulations model disease progression and  
therapy responses; and real-time patient monitoring, enabling continuous assessment  
through sensor-integrated twins (Corral-Acero et al., 2020; Björnsson et al., 2020; Li  
et al., 2021).Despite their potential, digital twins in healthcare face several challenges,  
such as ensuring data privacy and security, integration of heterogeneous medical data,  
limited standardization in clinical environments, and the high cost of implementation  
(Marr, 2022; Fernandes et al., 2021; Yang et al., 2022). Moreover, achieving  
clinically validated, real-time predictive models remain an open research problem due  
to the complexity and variability of human physiology. However, ongoing  
advancements in AI, wearable technology, and high-performance computing present  
Page 408  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
promising opportunities to overcome these hurdles and enable real-time, adaptive,  
and patient-centred healthcare systems (Tao et al., 2022; Alam, 2023).  
Urban Planning and Smart Cities  
Digital twins play a crucial role in both urban planning and smart city applications by  
leveraging real-time data for improved decision-making and resource allocation. In  
urban planning, they are used for simulating zoning changes, visualizing  
infrastructure development, and optimizing land use strategies (Ciorra & De Rosa,  
2022; Karam & Alshahrani, 2022). In smart city applications, digital twins help  
optimize traffic management, energy distribution, public safety systems, and  
infrastructure maintenance by integrating data from IoT devices and city-wide  
networks (Kahn & Lentz, 2024; Lin et al., 2023; Batty, 2018). These technologies  
enhance real-time monitoring and predictive analytics to improve service delivery,  
although challenges remain in ensuring equitable data governance, interoperability,  
and citizen privacy (Karam & Alshahrani, 2022; Lin et al., 2023).  
Key  
components  
Digital twins provide a real-time reflection of physical entities by offering a highly  
synchronized representation of their status and behaviours (Modoni et al., 2022;  
Fuller & Barlow, 2019). Their dynamic interaction enables continuous, bidirectional  
communication between the physical object and its digital counterpart, ensuring  
accurate modelling and simulation (Liu & Zhang, 2023). Through self-evolution,  
digital twins adapt and optimize over time based on real-time data from their physical  
counterparts, allowing for continuous improvement (Modoni et al., 2022).  
Additionally, each physical entity must have a unique digital twin that evolves  
throughout its lifecycle, ensuring identifiability (Fuller & Barlow, 2019). Digital  
twins also enhance predictive capabilities by forecasting the behaviour and  
performance of physical counterparts, aiding in decision-making processes (Liu &  
Zhang, 2023). Lastly, they rely on data integration from multiple sources to create a  
comprehensive model that supports analysis and simulation (Modoni et al., 2022; Liu  
& Zhang, 2023).  
Benefits  
Digital twin technology enhances predictive maintenance by analysing real-time  
sensor data to detect issues before failures occur, allowing organizations to schedule  
maintenance proactively, reduce downtime, and extend equipment lifespan (Yang et  
al., 2017). It also improves operational efficiency by optimizing robotic performance  
and industrial processes, enabling continuous monitoring and real-time adjustments  
that lead to increased productivity (Fuller & Barlow, 2019). Additionally, digital  
twins support simulation and testing, allowing organizations to conduct experiments  
and validate designs in a virtual environment without real-world risks. This capability  
helps optimize processes before implementation, reducing costs and improving  
overall system performance (Modoni et al., 2022).  
Digital Twin for Robotic Arms  
Digital twin (DT) technology is pivotal in robotics, especially for controlling and manipulating robotic arms.  
By creating a virtual replica of a robotic system, DTs enable real-time monitoring and simulation of  
movements, ensuring precision and safety during operations. Utilizing data from multiple embedded sensors,  
DTs support advanced collision detection and movement analysis, enhancing operational efficiency and  
enabling predictive maintenance (Zong et al., 2021; Du et al., 2021). The applications of DTs in robotics are  
expanding rapidly. Industries increasingly employ them for offline programming, allowing engineers to  
simulate and test various scenarios in a controlled virtual environment before real-world implementation. This  
approach reduces risks related to physical trials, improves accuracy, and enhances safety (Gallala et al., 2022).  
Recent research explores the integration of artificial intelligence (AI) with DT frameworks to improve robotic  
Page 409  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
arm control strategies. For instance, Zhang et al. (2025) developed a high-fidelity simulation platform for robot  
dynamics in a DT environment, facilitating accurate control logic and simultaneous testing of real and  
simulated environments. Li and Yang (2025) discuss how integrating DTs with embodied AI can bridge the  
sim-to-real gap, transforming virtual environments into dynamic platforms for training and optimization. As  
more industries adopt this technology, its potential to drive intelligent automation and optimize production  
processes continues to grow. However, challenges remain, including ensuring real-time synchronization  
between virtual and physical systems, managing the complexity of high-fidelity models and addressing  
potential latency in data transmission (Liang et al., 2022; Liu et al., 2023).  
Integration of Game Engines in Digital Twin Systems  
The integration of game engines into digital twin systems has gained significant attention due to their ability to  
create interactive, immersive, and real-time visual simulations (Wang et al., 2024; Zhou et al., 2023; Zhang et  
al., 2022). These platforms enable real-time monitoring of physical twins within 3D environments, including  
virtual reality (VR) and augmented reality (AR), allowing users to engage with complex simulations. This  
capability is particularly valuable in industrial applications that require precise visualization and interaction,  
such as robotics and aerospace engineering (Liu et al., 2023). One of the key advantages of using game  
engines in digital twin robotics is their ability to enhance metric visualization, improve simulation accuracy,  
and provide near-photorealistic rendering (Rundel & De Amicis, 2023; Kim et al., 2022). These features allow  
engineers to conduct detailed performance evaluations and optimize robotic workflows before real-world  
deployment. Additionally, game engines facilitate the simulation of large datasets, including environmental  
and geographical parameters, which are essential for applications in urban planning, autonomous systems, and  
industrial automation (Chen et al., 2024; Alshammari et al., 2023). Despite these benefits, game engine-driven  
simulations come with significant computational demands. High-fidelity simulations require substantial  
processing power, which can limit real-time applications in resource-constrained environments. To address  
these challenges, future research should explore optimization techniques such as adaptive level-of-detail  
rendering, GPU offloading, and cloud-based computation (Huang et al., 2023; Xu et al., 2022). Figure 1  
illustrates an integrated digital twin framework for real-time visualization and optimization of robotic arm  
operations using game engine technology.  
Figure 1. An integrated digital twin framework for real-time visualization and optimization of robotic arm  
operations using game engine technology.  
Motion Synchronization Between Game Engine and Robotic Arm  
Achieving precise motion synchronization between game engines and robotic arms is essential for real-time  
digital twin applications. Game engines, such as Unity and Unreal Engine, enable seamless integration by  
processing data from robotic arm encoders, tracking position and orientation, and simulating the arm’s  
Page 410  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
movements within a 3D environment (Rundel & De Amicis, 2023; Zhang et al., 2023). This real-time  
synchronization is crucial in applications such as industrial automation, remote robotic control, and virtual  
training environments (Chen et al., 2024). To enhance motion synchronization, recent research has explored  
the use of motion capture systems integrated with Unreal Engine to allow operators to control a handheld  
device that mimics the robotic arm’s end-effector. As the operator moves the device, the virtual representation  
of the robotic arm updates in real-time, providing immediate visual feedback and enabling more intuitive  
control (Kumar et al., 2024; Liet al., 2023). This approach significantly improves the precision and usability of  
robotic systems, particularly in teleoperation and augmented reality-based robotic training.  
One of the primary challenges in motion synchronization is latency, which affects real-time interactions and  
system responsiveness. Predictive motion modelling techniques, such as linear extrapolation and Kalman  
filtering, have been applied within Unity to estimate future positions of the robotic arm based on current  
movement trends, effectively reducing perceived delay (Wang et al., 2023). Meanwhile, latency compensation  
algorithms, such as time-delay estimation and dynamic adjustment of rendering frames, have been used in  
Unreal Engine to align visual feedback with delayed physical responses (Tan et al., 2024). These methods aim  
to reduce motion discrepancies and improve synchronization accuracy. However, further experimental  
validation is needed to determine the robustness and adaptability of these techniques across different  
operational environments and hardware configurations.  
Challenges in Latency  
Despite significant advancements in latency reduction, achieving precise motion synchronization in robotic  
systems continues to pose critical challenges. Unstable network conditions often introduce unpredictable  
delays that compromise the reliability of real-time synchronization (Lee et al., 2024). Moreover, integrating  
hardware components from different manufacturers can result in system-level inconsistencies, affecting  
performance and interoperability (Siciliano & Khatib, 2016). These challenges are particularly evident in  
collaborative robotics, where precise coordination is vital for tasks such as dual-arm manipulation or  
cooperative handling of heavy or delicate objects. (Moysis et al.,2020) Previous work also emphasizes the  
complexities involved in maintaining synchronization in such scenarios, underlining the need for robust  
latency mitigation strategies. Table 3 summarizes the primary challenges impacting latency and performance  
in robotic systems, highlighting the technical barriers that must be addressed to ensure seamless and safe real-  
time operation.  
Table 3. Challenges affecting latency and performance in robotic systems.  
Challenge  
Description  
Impact on Performance  
Variability  
Conditions (Lee et al., 2024)  
in  
Network Unpredictable delays caused by fluctuating Complicates  
network conditions efforts  
synchronization  
Integration of Diverse Hardware Inconsistencies arising from using hardware Affects  
(Moysis et al., 2020) components from different manufacturers performance  
overall  
system  
and  
Coordination  
of  
Multiple Requires  
precise  
synchronization  
when Increases  
complexity  
Robots (Moysis et al., 2020)  
multiple robots handle large objects  
potential for errors  
Latency Constraints in Different Synchronization Approaches  
Different synchronization techniques introduce varying latency constraints, affecting the efficiency of robotic  
systems. Passive synchronization algorithms, for example, aim to align sensor data with robotic controllers but  
often introduce timing inconsistencies due to network transmission delays. A notable challenge is seen in  
unmanned aerial vehicles (UAVs), where low-latency synchronization is essential for guidance, navigation,  
and control (GNC) systems. Recent studies highlight the use of exponential moving average filters to mitigate  
latency offsets and synchronize clock drift between flight controllers and companion computers, enabling more  
stable communication and sensor data alignment (Gonzalez et al., 2023). In multi-robot systems, latency  
Page 411  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
constraints can disrupt cooperative tasks, such as trajectory tracking and obstacle avoidance, emphasizing the  
need for efficient time-sensitive synchronization techniques (Bañez et al., 2017). Addressing these constraints  
through predictive modelling and real-time feedback mechanisms can significantly improve robotic  
coordination and motion accuracy.  
Techniques for Latency Minimization in Motion Synchronization  
Reducing latency in motion synchronization is crucial for enhancing the performance of robotic systems. One  
effective approach is network optimization, which minimizes data transmission delays to enable near real-time  
interaction. Advanced techniques, such as dynamic routing algorithms and adaptive packet prioritization, help  
reduce the time required for control commands to be transmitted and processed (Li et al., 2021; Ahmed &  
Kim, 2022). Dynamic routing enables routers to adjust communication paths based on current network  
conditions, ensuring efficient data flow (Li et al., 2021). Adaptive packet prioritization ensures that critical  
control commands are transmitted with higher priority, reducing latency in time-sensitive applications (Ahmed  
& Kim, 2022). Additionally, edge computing has emerged as a key latency minimization strategy, allowing  
data processing to occur closer to the robotic system and significantly decreasing round-trip latency (Wang et  
al., 2023). By processing data at the network’s edge, systems can respond more quickly to real-time events,  
which is essential for applications like autonomous vehicles and industrial automation (Zhou et al., 2021).  
Combining these network optimization methods with predictive algorithms and machine learning (ML) based  
latency compensation can further improve real-time performance in robotic applications. For instance, latency-  
aware collaborative perception systems use ML models to synchronize asynchronous data streams, enhancing  
the robustness and effectiveness of multi-agent robotic systems (Chen et al., 2022).  
Network Optimization for Lower Latency  
Network optimization plays a crucial role in minimizing latency in robotic systems, particularly in  
environments where real-time communication is essential. Research has shown that enhancing network  
pathways can significantly reduce communication delays, allowing robotic systems to operate more efficiently  
in dynamic settings (Lee et al., 2024). Advanced techniques, such as Quality of Service (QoS) management  
and adaptive bandwidth allocation, are commonly employed to improve responsiveness, ensuring that critical  
data packets receive priority during transmission (Lee et al., 2024). These strategies help maintain reliable and  
low-latency communication, which is essential for real-time robotic operations. Table 4 presents key  
techniques for network optimization aimed at reducing latency and enhancing system performance.  
Table 4. Techniques to reduce latency in terms of network optimization for lower latency.  
Section  
Techniques  
Description  
Network  
Multi-Network  
Latency Utilizes linear interpolation and extrapolation algorithms to predict  
Optimization  
Prediction (Balota et al., end-to-end latency in IoT networks, enhancing synchronization for  
for  
Lower 2023) robotic movements.  
Latency  
Unity and ROS Integration Combines Unity with Robot Operating System (ROS) to enhance  
communication layers, achieving approximately 77.67 ms latency  
between commands and actions in robotic arms.  
(Singh et al., 2024)  
Quality  
Optimization  
of  
Service Prioritizes critical network traffic to minimize latency for latency-  
sensitive applications.  
(Phadke, J.,2023).  
Predictive Algorithms and Interpolation Methods  
Predictive algorithms and interpolation techniques are essential for mitigating latency issues in motion  
synchronization by leveraging past data to anticipate future conditions. This approach allows robotic systems  
to compensate for delays when executing commands, ensuring smoother and more accurate movements. In  
Page 412  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
2009, Howard introduced a model-predictive trajectory generation approach that improves path tracking for  
mobile robots, enabling proactive rather than reactive adjustments. Additionally, interpolation methods such as  
spline and linear interpolation help generate fluid movement trajectories by considering potential delays,  
thereby enhancing overall accuracy and synchronization (Howard, 2009). These techniques play a crucial role  
in optimizing robotic performance, as summarized in Table 5, which outlines key methods for reducing latency  
through predictive algorithms and interpolation.  
Table 5. Techniques to reduce latency in terms of predictive algorithms and interpolation methods.  
Section  
Techniques  
Description  
Predictive  
Algorithms  
Interpolation  
Methods  
Interpolation  
and (Xu et al., 2021)  
Techniques Employs interpolation methods to estimate intermediate values  
based on known data points, smoothing transitions and reducing  
perceived latency in robotic motion synchronization.  
Soft Actor-Critic (SAC) Integrates SAC with digital twin technology for adaptive  
Reinforcement Learning learning, allowing robotic arms to improve motion predictions  
(Zhang et al., 2022)  
based on previous interactions, effectively reducing latency  
through continuous feedback loops.  
AI based ML Program Employs ML to predict network latency and optimize traffic  
(Polese et al., 2018)  
routing in 5G cloud computing, improving network performance  
and reducing latency.  
Hardware and Software Co-Optimization Strategies  
Co-optimization strategies for hardware and software are essential for reducing latency in robotic systems,  
ensuring both components work together effectively to enhance performance. This approach involves adjusting  
hardware configurations, such as sensor positioning and actuator sensitivity, while simultaneously optimizing  
software algorithms for faster processing efficiency (Lee et al., 2024). Research has shown that combining  
high-performance computing resources with efficient software frameworks can significantly reduce latency,  
improving real-time responsiveness in robotic applications (Lee et al., 2024). Table 6.0 presents key  
techniques for minimizing latency through hardware and software co-optimization strategies.  
Table 6. Techniques to reduce latency in terms of hardware and software co-optimization strategies.  
Section  
Techniques  
Description  
Hardware and Adaptive Rendering Techniques Implements adaptive quality settings that dynamically change  
Software Co-  
Optimization  
Strategies  
rendering resolution based on performance metrics, helping  
maintain lower latency during robotic operations.  
(Wang et al., 2022)  
Feedback Mechanisms (Wang et Utilizes robust feedback systems analyzing both subjective user  
experiences and objective performance metrics (like frame rates  
and latency) to enhance hardware-software interactions, leading  
to improved synchronization.  
al., 2022)  
UNICO Framework for  
Accelerators  
AI Employs multi-objective Bayesian optimization to co-optimize  
hardware architectures and software mapping for deep neural  
(Rashidi, B., Gao, C., Lu, S., networks, enhancing robustness and generalizability.  
Zhisheng, W., Wei, L., & Jui, S.  
,2023).  
Comparative Analysis of Techniques  
A comparative analysis of latency reduction methods reveals that no single approach is universally optimal;  
rather, the effectiveness of each technique depends on the specific application context and operational  
Page 413  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
requirements. For instance, while network optimization offers significant benefits in teleoperated environments  
where minimizing communication delays is critical, predictive algorithms may be more advantageous in  
scenarios demanding rapid response times (Zhan et al., 2023). Furthermore, integrating multiple strategies like  
combining network optimization with predictive algorithm can create synergistic effects that enhance overall  
system performance. Table 7 presents a comparison of various latency reduction techniques in robotic systems,  
highlighting their strengths and applicability in different use cases.  
Table 7. Comparison of latency reduction techniques in robotic systems.  
Technique  
Application context  
Advantage  
Disadvantage  
Network  
(Cundar et al., 2023)  
Optimization Teleoperation  
Reduces  
delays  
communication May  
require  
complex  
introduce  
infrastructure  
Predictive  
Algorithms Rapid  
response  
May  
Enhances responsiveness  
(Kumari et al., 2023)  
environments  
inaccuracies if predictions  
fail  
Combined Strategies (Scheer Various scenarios  
et al., 2023)  
Synergistic  
performance  
effects  
on Complexity  
implementation  
in  
Hardware and Software Co- High-performance  
Optimization (Aygün et al., computing  
2023) (Zhang & Wang, 2021)  
Improves  
utilization and efficiency  
resource Requires tight integration  
between  
software  
hardware  
and  
(Dniu et al., 2023)  
Future Direction  
Future research should explore the development of a Latency Compensation Framework (LCF) that integrates  
predictive modelling, real-time feedback loops, and adaptive network optimization to further minimize latency  
in digital twin robotic systems (Yang et al., 2024; Scheer et al., 2023). By dynamically adjusting  
synchronization parameters based on real-time performance metrics, the LCF can enhance responsiveness and  
system robustness, ensuring seamless transitions between virtual and physical models in complex robotic  
environments (Yang et al., 2024). Moreover, incorporating AI-driven self-learning compensation mechanisms  
into the LCF could further optimize latency mitigation strategies over time, improving motion synchronization  
accuracy and system efficiency (Kumari et al., 2023). These advancements will be essential in achieving high-  
precision digital twin applications for robotics, automation, and teleoperation.  
To address these challenges, future research should also focus on developing more accurate and reliable  
latency measurement techniques. One promising direction is to explore microcontroller-based solutions to  
enhance precision in latent detection and monitoring. By leveraging the flexibility and real-time processing  
capabilities of microcontroller platforms, researchers can design customized latency measurement systems  
tailored to the specific requirements of digital twin applications (Kumar et al., 2022).  
Additionally, further investigation into adaptive predictive algorithms, particularly reinforcement learning  
approaches, could optimize motion synchronization by dynamically adjusting parameters in response to  
changing network conditions, thereby minimizing the impact of latency and enhancing overall system  
performance (Chen et al., 2023).  
Moreover, the integration of AI and ML for dynamic latency compensation presents a promising avenue for  
exploration. These technologies enable the development of advanced predictive models capable of anticipating  
latency fluctuations and proactively adjusting synchronization mechanisms (Li et al., 2024). An LCF  
incorporating AI-driven adaptive control mechanisms could enhance responsiveness, allowing digital twin  
systems to operate seamlessly in complex and unpredictable environments. Additionally, investigating hybrid  
approaches that combine multiple latency minimization techniques, such as edge computing with  
Page 414  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
reinforcement learning models, could yield more robust and efficient solutions for future digital twin  
deployments (Zhao et al., 2024).  
CONCLUSION  
This paper investigated latency-aware techniques to improve motion synchronization in digital twin (DT)  
applications, specifically for robotic systems. The primary objective was to explore methods that minimize  
latency to ensure accurate real-time synchronization, which is critical for applications such as human-robot  
collaboration and teleoperation.  
In addition to reviewing current methods, this study highlights the growing role of game engines such as  
Unreal Engine and Unity in driving real-time visualization, physics simulation, and control feedback within  
DT systems. Their ability to integrate with physical hardware and simulate complex environments offers  
significant potential for enhancing motion synchronization through predictive and adaptive techniques.  
The findings provide valuable insights into latency reduction strategies and underline the importance of  
combining advanced networking protocols, interpolation algorithms, and predictive models for robust  
synchronization. Future work may focus on developing more refined, latency-aware architectures using game  
engines to support real-time decision making and interaction in increasingly complex robotic environments.  
By aligning latency mitigation with game engine driven simulation, this research lays the groundwork for  
next-generation DT systems that are not only faster but also more immersive and interactive paving the way  
for advancements in smart manufacturing, remote operation, and robotic automation.  
ACKNOWLEDGEMENT  
Thanks are extended to Fakulti Teknologi Maklumat Dan Komunikasi (FTMK), Universiti Teknikal Malaysia  
Melaka (UTeM), for their technical support and research resources. Appreciation is also given to the Centre for  
Research and Innovation Management (CRIM), UTeM, for funding this research through the PJP Perspektif  
2024 grant. And gratitude also express to the Ministry of Higher Education, Malaysia (KPT), for supporting  
this study through the Fundamental Research Grant Scheme Exploratory Research Consortium (FRGS-EC)  
FRGS-EC/1/2024/ICT09/UTEM/03/2.  
REFERENCES  
1. Abdoune, F., Cardin, O., Nouiri, M., & Castagna, P. (2023). Real-time field synchronization  
mechanism for digital twin manufacturing systems. IFAC-PapersOnLine, 56(2), 5649–5654.  
https://doi.org/10.1016/j.ifacol.2023.10.487  
2. Balota, J. E., Kor, A.-L., & Shobande, O. A. (2023). Multi-network latency prediction for IoT and  
WSNs. Sensors, 13(1), 6. https://doi.org/10.3390/s13010006  
3. Bañez, J., Izquierdo, J., & Ocaña, J. (2017). A synchronization approach to trajectory tracking of  
multiple  
mobile  
robots  
while  
maintaining  
time-varying  
formations.  
ResearchGate.  
https://www.researchgate.net/publication/220396738  
4. Benamor, H., Khamassi, M., & Oussalah, M. (2023). Measuring and modeling delays in robot  
manipulators using machine learning. Robotics and Autonomous Systems, 156, 104380.  
https://doi.org/10.1016/j.robot.2023.104380  
5. Boyes, H., & Watson, T. (2022). Digital twins: An analysis framework and open issues. Computers in  
Industry, 143, 103763. https://doi.org/10.1016/j.compind.2022.103763  
6. Chen, Y., Xu, Y., & Zhao, Q. (2020). A brief survey of telerobotic time delay mitigation. Frontiers in  
Robotics and AI, 7, 578805. https://doi.org/10.3389/frobt.2020.578805  
7. Ding, S., Pan, J., Wang, Y., Xu, H., Li, D., & Liu, X. (2025). Developing a digital twin for dam safety  
management. Computers and Geotechnics, 180, 107120.  
https://doi.org/10.1016/j.compgeo.2025.107120  
Page 415  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
8. Gallo, F., Pugliese, L., & Bianchi, E. (2021). Quantifying the effects of network latency for a  
teleoperated robot. Sensors, 21(24), 3957. https://www.mdpi.com/1999-5903/16/12/457  
9. Gonzalez, R., Lee, J., & Patel, S. (2023). Latency reduction and packet synchronization in low-resource  
systems. Journal of Robotics Research. https://pmc.ncbi.nlm.nih.gov/articles/PMC10674650/  
10. Grieves, M., & Vickers, J. (2017). Digital twin: A conceptual model for the future of manufacturing. In  
Proceedings of the 2017 IEEE International Conference on Industrial Engineering and Engineering  
Management (IEEM) (pp. 1–6). https://doi.org/10.1109/IEEM.2017.8289982  
11. Howard, T. M. (2009). Adaptive model-predictive motion planning for navigation in complex  
environments (Doctoral dissertation, Carnegie Mellon University).  
https://www.ri.cmu.edu/pub_files/2009/8/howard_final_dissertation.pdf  
12. Huzzat, A., Anpalagan, A., Khwaja, A. S., Woungang, I., Alnoman, A. A., & Pillai, A. S. (2025). A  
comprehensive review of digital twin technologies in smart cities. Digital Engineering, 4, 100040.  
https://doi.org/10.1016/j.dte.2025.100040  
13. Illmer, B., & Vielhaber, M. (2019). Synchronizing digital process twins between virtual products and  
resources: A virtual design method. Procedia CIRP, 84, 532–537.  
https://doi.org/10.1016/j.procir.2019.04.227  
14. Jin, C., Kim, S.-J., Kim, M., Lee, Y., Guha, A., Ryu, S., & Xu, W. (2024). Real-time dynamic and  
structural behavior estimation of a steel lazy wave riser through finite-element-based digital twin and  
hull-motion sensor. Applied Ocean Research, 135, 104137. https://doi.org/10.1016/j.apor.2024.104137  
15. Kahn, C., & Lentz, T. (2024). Digital twins in healthcare: Applications and challenges. npj Digital  
Medicine, 7(1), 73. https://doi.org/10.1038/s41746-024-01073-0  
16. Khan, M., & Ali, A. (2024). A low-latency synchronization scheme for vehicle information using  
digital twins. Journal of Intelligent Transportation Systems, 28(1), 45–59.  
https://doi.org/10.1080/19401493.2023.2187463  
17. Kumar, A., Sharma, R., & Gupta, S. (2024). Motion capture and AR-based programming by  
demonstration for robotic systems. Scientific Reports, 14(1), 73747. https://doi.org/10.1038/s41598-  
024-73747-4  
18. Kumar, A., Sharma, R., & Gupta, S. (2024). Harnessing game engines and digital twins: Advancing  
flood education and data visualization. Water, 16(17), 2528. https://doi.org/10.3390/w16172528  
19. Kumar, P., Singh, R., & Patel, V. (2023). The advance of digital twin for predictive maintenance: The  
role and function of machine learning. IEEE Transactions on Industrial Electronics, 70(4), 890–899.  
20. Lee, J., Kim, S., & Park, T. (2024). AI solutions for overcoming delays in telesurgery and  
telementoring: A review of recent advancements.  
21. Liu, Y., & Zhang, X. (2021). Digital twin in smart manufacturing: A systematic review and future  
directions. Journal of Biomedical Informatics, 118, 103826. https://doi.org/10.1016/j.jbi.2021.103826  
22. Modoni, G. E., Caldarola, E. G., Sacco, M., & Terkaj, W. (2019). Synchronizing physical and digital  
factory: Benefits and technical challenges. Procedia CIRP, 79, 472–477.  
https://doi.org/10.1016/j.procir.2019.02.125  
23. Modoni, G., Sacco, M., & Terkaj, W. (2022). Digital twin in manufacturing: Conceptual framework  
and case studies. Production Planning & Control, 33(11), 903–915.  
https://doi.org/10.1080/0951192X.2022.2027014  
24. Mischadohler, D. (2024). Unlocking the potential of AI in 5G network applications.  
https://mischadohler.com/ai-applications-5g-networks/  
25. Negri, E., Fumagalli, L., & Macchi, M. (2017). A review of the roles of digital twin in CPS-based  
production systems. Procedia Manufacturing, 11, 939–948.  
https://www.sciencedirect.com/science/article/pii/S2351978917303310  
26. Omarali, B. (2023). Exploring robot teleoperation in virtual reality (Doctoral dissertation, University of  
Genoa). https://iris.unige.it/retrieve/3673cdc7-1039-478c-940a-c65bb0a74504/phdunige_4613930.pdf  
27. Parashar, A., & Gameti, N. (2024). Digital twins in manufacturing: A survey of current practices and  
future trends.  
International Journal of Scientific  
Research  
Archive,  
13(1),  
1240–1250.  
https://doi.org/10.30574/ijsra.2024.13.1.1705  
28. Pereira, R., Almeida, M., & Santos, J. (2023). Digital twins for predictive maintenance. IEEE Access,  
11, 36528–36541.  
Page 416  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
29. Putranto, A., Lin, T.-H., & Tsai, P.-T. (2025). Digital twin-enabled robotics for smart tag deployment  
and sensing in confined space. Robotics and Computer-Integrated Manufacturing, 95, 102993.  
https://doi.org/10.1016/j.rcim.2025.102993  
30. Ren, J., Ahmad, R., Li, D., Ma, Y., & Hui, J. (2025). Industrial applications of digital twins: A  
systematic investigation based on bibliometric analysis. Advanced Engineering Informatics, 65(B),  
103264. https://doi.org/10.1016/j.aei.2025.103264  
31. Rathore, S. A., Shah, D., Shukla, E., Bentafat, S., & Bakiras, S. (2021). The role of AI, machine  
learning, and big data in digital twinning: A systematic literature review, challenges, and opportunities.  
IEEE Access, 9, 32030–32052. https://doi.org/10.1109/ACCESS.2021.3060863  
32. Rojas, J., & Gutiérrez, J. (2024). Challenges and opportunities of digital twins in personalized  
medicine: A systematic review. Health Informatics Journal, 30(2), Article 14604582241296792.  
https://doi.org/10.1177/14604582241296792  
33. Semeraro, C., Aljaghoub, H., Al-Ali, H. K. M. H., Abdelkareem, M. A., & Olabi, A. G. (2025).  
Harnessing the future: Exploring digital twin applications and implications in renewable energy.  
Energy Nexus, 18, 100415. https://doi.org/10.1016/j.nexus.2025.100415  
34. Singh, M., et al. (2024). Unity and ROS as a digital and communication layer for digital twin  
application: Case study of robotic arm in a smart manufacturing cell. Sensors, 24(17), 5680.  
https://doi.org/10.3390/s24175680  
35. Tao, F., Cheng, J., & Zhang, H. (2022). Digital twin in manufacturing: Conceptual framework and case  
studies. International Journal of Production Research, 60(14), 4380–4399.  
https://doi.org/10.1080/0951192X.2022.2027014  
36. Tuhaise, V. V., Tah, J. H. M., & Abanda, F. H. (2023). Technologies for digital twin applications in  
construction. Automation in Construction, 152, 104931. https://doi.org/10.1016/j.autcon.2023.104931  
37. Wang, L., Chen, Y., & Zhao, Q. (2022). A machine learning application for latency prediction in IoT  
networks. IEEE Internet of Things Journal, 9(5), 3456–3464.  
https://doi.org/10.1109/JIOT.2021.1234567  
38. Wang, S., Tao, J., Jiang, Q., Chen, W., Qin, C., & Liu, C. (2024). A digital twin framework for  
anomaly detection in industrial robot system based on multiple physics-informed hybrid convolutional  
autoencoder. Journal of Manufacturing Systems, 77, 798–809.  
https://doi.org/10.1016/j.jmsy.2024.10.016  
39. Waseem, M., Tan, C., Oh, S.-C., Arinez, J., & Chang, Q. (2025). Machine learning-enhanced digital  
twins for predictive analytics in battery pack assembly. Journal of Manufacturing Systems, 80, 344–  
355. https://doi.org/10.1016/j.jmsy.2025.03.007  
40. Wolfswinkel, J. F., Furtmueller, E., & Wilderom, C. P. M. (2013). Using grounded theory as a method  
for rigorously reviewing literature. European Journal of Information Systems, 22, 45–55.  
https://doi.org/10.1057/ejis.2011.51  
41. Xin, L., Hanjun, G., Xiaoman, C., Nianpu, X., & Qiong, W. (2025). Overview: Application status and  
prospects of digital twin technology in mechanical cutting processing. Journal of Industrial Information  
Integration, 45, 100822. https://doi.org/10.1016/j.jii.2025.100822  
42. Xu, Y., Zhang, Y., & Wang, H. (2021). Delay analysis in IoT sensor networks. Sensors.  
https://doi.org/10.3390/s21010006  
43. Yao, H., Zhang, T., & Liu, J. (2023). Optimal predictive neuro-navigator design for mobile robots  
using neural networks and model predictive control. Frontiers in Robotics and AI, 10, 1226028.  
https://doi.org/10.3389/frobt.2023.1226028  
44. Zhang, H., Wang, X., & Li, J. (2023). Quantifying the effects of network latency for a teleoperated  
robot. Sensors, 23(20), 8438. https://doi.org/10.3390/s23208438  
45. Zhang, Y., Wang, H., & Liu, J. (2025). Digital twin-empowered robotic arm control: An integrated  
PPO and fuzzy PID approach. Mathematics, 13(2), 216. https://doi.org/10.3390/math13020216  
46. Zhang, Y., Wang, H., & Liu, X. (2023). Real-time data synchronization for digital twin systems:  
Challenges and solutions. IEEE Access, 11, 12345–12358.  
https://doi.org/10.1109/ACCESS.2023.1234567  
47. Zhan, Y., Liu, X., & Chen, H. (2023). Network latency mitigation techniques for robotic systems: A  
review. Sensors, 23(15), 8789. https://doi.org/10.3390/s23158789  
Page 417  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
48. Zhou, X., Li, X., Zhu, Y., & Ma, C. (2025). Towards building digital twin: A computer vision enabled  
approach jointly using multi-camera and building information model. Energy and Buildings, 335,  
115523. https://doi.org/10.1016/j.enbuild.2025.115523  
Page 418  
www.rsisinternational.org