Real-world Scenarios Research Articles

Improved Road Extraction Models through Semi-Supervised Learning with ACCT

Improving the performance and reducing the training cost of road extraction models in the absence of samples is important for updating road maps. Despite the success of recent road extraction models on standard datasets, they often fail to perform when applied to new datasets or real-world scenarios where labeled samples are not available. In this paper, our focus diverges from the typical quest to pinpoint the optimal road extraction model or evaluate generalization prowess across models. Instead, we propose a method called Asymmetric Consistent Co-Training (ACCT) to train existing road extraction models faster and make them perform better in new scenarios lacking samples. ACCT uses two models with different structures and a supervision module to enhance accuracy through mutual learning. Labeled and unlabeled images are processed by both models to generate road maps from different perspectives. The supervision module ensures consistency between predictions by computing losses based on labeling status. ACCT iteratively adjusts parameters using unlabeled data, improving generalization. Empirical evaluations show that ACCT improves IoU by 2.79% to 10.26% using only 1/8 of the labeled data compared to fully supervised methods. It also reduces parameters by over 49% compared to state-of-the-art semi-supervised methods while maintaining similar accuracy. These results highlight the potential of leveraging large amounts of unlabeled data to enhance road extraction models as data acquisition technology advances.

Comparative Evaluation of Water Level Forecasting Using IoT Sensor Data: Hydrodynamic Model SWMM vs. Machine Learning Models Based on NARX Framework

This study evaluates the accuracy of water level forecasting using two approaches: the hydrodynamic model SWMM and machine learning (ML) models based on the Nonlinear Autoregressive with Exogenous Inputs (NARX) framework. SWMM offers a physically based modeling approach, while NARX is a data-driven method. Both models use real-time precipitation data, with their predictions compared against measurements from a network of IoT sensors in a stormwater management system. The results demonstrate that while both models provide effective forecasts, NARX models exhibit higher accuracy, with improved Nash–Sutcliffe Efficiency (NSE) coefficients and 33–37% lower mean absolute error (MAE) compared to SWMM. Despite these advantages, NARX models may struggle with limited data on extreme flooding events, where they could face accuracy challenges. Enhancements in SWMM modeling and calibration could reduce the performance gap, but the development of SWMM models requires substantial expertise and resources. In contrast, NARX models are generally more resource-efficient. Future research should focus on integrating both approaches by leveraging SWMM simulations to generate synthetic data, particularly for extreme weather events, to enhance the robustness of NARX and other ML models in real-world flood prediction scenarios.

Adaptive haze pixel intensity perception transformer structure for image dehazing networks.

In the realm of deep learning-based networks for dehazing using paired clean-hazy image datasets to address complex real-world haze scenarios in daytime environments and cross-dataset challenges remains a significant concern due to algorithmic inefficiencies and color distortion. To tackle these issues, we propose SwinTieredHazymers (STH), a dehazing network designed to adaptively discern pixel intensities in hazy images and compute haze residue for clarity restoration. Through a unique three-branch design, we hierarchically modulate haze residuals by leveraging the global features brought by Transformer and the local features brought by Convolutional Neural Network (CNN) which has led to the algorithm's widespread applicability. Experimental results demonstrate that our approach surpasses advanced single-image dehazing methods in both quantitative metrics and visual fidelity for real-world hazy image dehazing, while also exhibiting strong performance in cross-dataset dehazing scenarios.

Tolerance Intervals Under a Class of Unbalanced Linear Mixed Models

The study of tolerance intervals for linear mixed models under unbalanced data has been mainly focused on one-way random-effects models. In this article, we present a method to compute both one- and two-sided ( β , γ ) -tolerance intervals that is applicable to a more general class of unbalanced linear mixed models. The proposed method is based on the concept of generalized pivotal quantities and thus relies on the derivation of pivotal quantities that are shown to be independent within this particular class of models. The method employs Monte Carlo sampling methods to obtain realizations of the quantities needed to compute the tolerance intervals of interest. Extensive simulation studies confirm that coverage probabilities of the computed tolerance intervals are consistently close to their nominal levels. Furthermore, to showcase the application of our method in real-world scenarios, we provide illustrative examples of case studies that examine models falling within the specified class. Lastly, motivated by the prospective application of our method in drug development, we implement the proposed procedures to estimate shelf life of a drug product through tolerance intervals and unbalanced stability data.

Evolution of cooperation with asymmetric rewards

Rewards, as a form of positive reinforcement, effectively encourage cooperation. In this paper, we study a multi-population prisoner's dilemma game with asymmetric rewards, where agents in the same population play prisoner's dilemma game, and agents from the giver population can reward agents from the recipient population only if they make the same choice. In well-mixed populations, asymmetric rewards can facilitate cooperation. Similarly, asymmetric rewards on the regular square lattice can effectively prevent complete defection. In both well-mixed and structured populations, seemingly disadvantageous cooperative givers play an important role in the maintenance and spread of cooperation. Especially on lattice, cooperative givers with asymmetric rewards can be active in the system through diverse cases of cyclic dominance. Our findings provide deeper insights into the impact of asymmetry on cooperative behavior and the development of altruistic behavior in real-world scenarios.

An improved whale optimization algorithm for UAV swarm trajectory planning

For the problem of trajectory planning in the small-scale unmanned aerial vehicle (UAV) swarms, classical intelligent algorithms and newly emerged bio-inspired algorithms often suffer from being trapped in local optima, leading to suboptimal solutions. Moreover, these algorithms fail to ensure optimal solutions and convergence speed in high-speed dynamic UAV networking scenarios. In response to these challenges, this paper proposes an Improved Whale Optimization Algorithm (IWOA) that considers the global perspective. To address the issue of initial solution diversity, an opposition-based learning method is introduced by constructing a reverse population, enhancing the diversity of the initial population. To overcome the problem of the algorithm getting trapped in local optima, random convergence factors, and end-point random neighborhood perturbations are incorporated to accelerate the global convergence speed of the IWOA and prevent it from getting stuck in local optima. Additionally, a grid digital elevation model is employed when constructing the experimental environment model, considering various physical constraints of the UAV swarm. This ensures that the simulation validation of the IWOA algorithm is closer to real-world scenarios. Simulation results indicate that, compared to commonly used algorithms, the IWOA can generate more optimal trajectories for drone swarm planning under constraints resembling real-world scenarios. It exhibits better performance in trajectory evaluation, convergence speed, and stability, thereby meeting the requirements of trajectory planning for small-scale UAV swarms. The proposed IWOA enhances UAV swarm coordination and efficiency, significantly impacting real-world applications in various fields.

In-depth Analysis of Students' Mathematical Problem-Solving Skills: Influence Factors Motivation and Effective Teaching Strategies

Problem-solving (PS) skills are efforts to find solutions when faced with challenges. This study aims to examine students' mathematical PS skills by focusing on the influence of learning motivation and identifying factors that were previously overlooked. A qualitative approach with a case study design was used, where the researcher acted as the primary instrument. The research also utilized PISA-type test instruments, observation sheets, and interviews. PISA-type questions were designed around content, context, and process components, adhering to the PISA framework. The study was conducted in a school in Surakarta, chosen for its alignment with the minimum competency assessment system similar to PISA. Using snowball sampling, nine students were selected for response analysis. These students were categorized based on their motivation levels: high, moderate, and low. The results show that students with high motivation are proactive in problem-solving, directly addressing the problem, reviewing each step as they proceed, and ensuring accuracy. Moderately motivated students take more time, often using scratch paper, and review their work only at the end. Low-motivation students face greater difficulties, frequently making mistakes, relying heavily on scratch paper, and often skipping the review process due to time pressure. To improve PS skills, future research could explore the integration of interactive digital tools with instant feedback, adaptive learning platforms tailored to student motivation, and collaborative learning exercises. Additionally, incorporating real-world scenarios into problem-based learning could enhance engagement and develop deeper problem-solving abilities across different motivation levels.

Visual search and real-image similarity: An empirical assessment through the lens of deep learning.

The ability to predict how efficiently a person finds an object in the environment is a crucial goal of attention research. Central to this issue are the similarity principles initially proposed by Duncan and Humphreys, which outline how the similarity between target and distractor objects (TD) and between distractor objects themselves (DD) affect search efficiency. However, the search principles lack direct quantitative support from an ecological perspective, being a summary approximation of a wide range of lab-based results poorly generalisable to real-world scenarios. This study exploits deep convolutional neural networks to predict human search efficiency from computational estimates of similarity between objects populating, potentially, any visual scene. Our results provide ecological evidence supporting the similarity principles: search performance continuously varies across tasks and conditions and improves with decreasing TD similarity and increasing DD similarity. Furthermore, our results reveal a crucial dissociation: TD and DD similarities mainly operate at two distinct layers of the network: DD similarity at the intermediate layers of coarse object features and TD similarity at the final layers of complex features used for classification. This suggests that these different similarities exert their major effects at two distinct perceptual levels and demonstrates our methodology's potential to offer insights into the depth of visual processing on which the search relies. By combining computational techniques with visual search principles, this approach aligns with modern trends in other research areas and fulfils longstanding demands for more ecologically valid research in the field of visual search.

A shared robot control system combining augmented reality and motor imagery brain-computer interfaces with eye tracking.

Objective: Brain-computer interface (BCI) control systems monitor neural activity to detect the user's intentions, enabling device control through mental imagery. Despite their potential, decoding neural activity in real-world conditions poses significant challenges, making BCIs currently impractical compared to traditional interaction methods. This study introduces a novel motor imagery (MI) BCI control strategy for operating a physically assistive robotic arm, addressing the difficulties of MI decoding from electroencephalogram (EEG) signals, which are inherently non-stationary and vary across individuals.
Approach: A proof-of-concept BCI control system was developed using commercially available hardware, integrating MI with eye tracking in an augmented reality (AR) user interface to facilitate a shared control approach. This system proposes actions based on the user's gaze, enabling selection through imagined movements. A user study was conducted to evaluate the system's usability, focusing on its effectiveness and efficiency.
Main results:Participants performed tasks that simulated everyday activities with the robotic arm, demonstrating the shared control system's feasibility and practicality in real-world scenarios. Despite low online decoding performance (mean accuracy: 0.52 9, F1: 0.29, Cohen's Kappa: 0.12), participants achieved a mean success rate of 0.83 in the final phase of the user study when given 15 minutes to complete the evaluation tasks. The success rate dropped below 0.5 when a 5-minute cutoff time was selected.
Significance: These results indicate that integrating AR and eye tracking can significantly enhance the usability of BCI systems, despite the complexities of MI-EEG decoding. While efficiency is still low, the effectiveness of our approach was verified. This suggests that BCI systems have the potential to become a viable interaction modality for everyday applications
in the future.

Adaptive MAC Scheme for Interference Management in Ad Hoc IoT Networks

The field of wireless communication has undergone revolutionary changes driven by technological advancements in recent years. Central to this evolution is wireless ad hoc networks, which are characterized by their decentralized nature and have introduced numerous possibilities and challenges for researchers. Moreover, most of the existing Internet of Things (IoT) networks are based on ad hoc networks. This study focuses on the exploration of interference management and Medium Access Control (MAC) schemes. Through statistical derivations and systematic simulations, we evaluate the efficacy of guard zone-based MAC protocols under Rayleigh fading channel conditions. By establishing a link between network parameters, interference patterns, and MAC effectiveness, this work contributes to optimizing network performance. A key aspect of this study is the investigation of optimal guard zone parameters, which are crucial for interference mitigation. The adaptive guard zone scheme demonstrates superior performance compared to the widely recognized Carrier Sense Multiple Access (CSMA) and the system-wide fixed guard zone protocol under fading channel conditions that mimic real-world scenarios. Additionally, simulations reveal the interactions between network variables such as node density, path loss exponent, outage probability, and spreading gain, providing insights into their impact on aggregated interference and guard zone effectiveness.

Quantum-Secured Data Centre Interconnect in a field environment

In the evolving landscape of quantum technology, the increasing prominence of quantum computing poses a significant threat to the security of conventional public key infrastructure. Quantum key distribution (QKD), an established quantum technology at a high readiness level, emerges as a viable solution with commercial adoption potential. QKD facilitates the establishment of secure symmetric random bit strings between two geographically separated, trustworthy entities, safeguarding communications from potential eavesdropping. In particular, data centre interconnects can leverage the potential of QKD devices to ensure the secure transmission of critical and sensitive information in preserving the confidentiality, security, and integrity of their stored data. In this article, we present the successful implementation of a QKD field trial within a commercial data centre environment that utilises the existing fibre network infrastructure. The achieved average secret key rate of 2.392 kbps and an average quantum bit error rate of less than 2% demonstrate the commercial feasibility of QKD in real-world scenarios. As a use case study, we demonstrate the secure transfer of files between two data centres through the Quantum-Secured Virtual Private Network, utilising secret keys generated by the QKD devices. %This demonstration marks a significant milestone in the deployment of QKD across Singapore, establishing a foundation for widespread adoption and enhancing the infrastructure for practical, quantum-safe communication in commercial environments.

Experience from Designing Augmented Reality Browsing Interfaces for Real-world Walking Scenarios

Mobile phones have enabled users to browse information in varying mobility contexts. For high-mobility settings such as walking, however, phones pose several usability challenges, particularly safety and limited screen sizes. While Augmented Reality (AR) has been proposed to address these issues, prior work has yet to investigate AR interface design in real-world walking conditions beyond text readability and notification design. This paper presents the first exploration of AR browsing interface design and extended usage while walking in the wild. We first conducted design sessions with 12 UI designers while walking in varied environments to design the window size, distance, opacity, anchor type, and placement for three categories of apps: text, video, and mixed content. Results show that traffic level significantly affects the designed window size, whereas content type significantly affects window size, distance, opacity, and vertical placement. To gain further insights from real-world usage, we conducted a multi-day observational study with 5 participants and observed that participants on average switched among window layouts every 3.3 minutes, for reasons such as safety and the level of extended visual attention.

Deep Learning Methods for Chest X-Ray Imaging-Based COVID-19 Pneumonia Detection

The COVID-19 pandemic currently underway has highlighted a need for fast and accurate screening tools to help identify the disease, particularly in resource-poor settings. In this paper, a deep convolutional neural network (CNN) model is proposed for the automatic detection of COVID-19 pneumonia using chest X-ray images that helps in reducing the cost and processing time compared to traditional testing procedures. It uses several deep-structured medical image data types and improvement of the deep learning model architecture to achieve good diagnostic accuracy. Our hybrid CNN architecture integrates VGG-16 for feature extraction and ResNet-50 for pattern complexity assessment in detecting fine changes characteristic of COVID-19 pneumonia. Our dataset for this study is a collection of few thousands labelled chest X-ray images categorized in three classes, which are COVID-19, viral pneumonia and healthy cases. More advanced data preprocessing methods such as normalized, augmentation, and filtered noise has also been carried out which enhances in the model performance. We have high accuracy, recall, and good F1-score in the experimental results proving model robustness are applicable in real-world clinical scenarios. These research results speak to the importance of AI for improving diagnosis workflows, especially in fast-to-deploy large scale scenarios needed during a pandemic to cope with healthcare burdens.

Word learning in the wild: App-based evidence for valence and concreteness effects

Abstract Second language (L2) learners need to acquire large vocabularies to approach native-like proficiency. Many controlled experiments have investigated the factors facilitating and hindering word learning; however, few studies have validated these findings in real-world learning scenarios. We use data from the language learning app Lingvist to explore how L2 word learning is affected by valence (positivity/negativity) and concreteness of target words and their linguistic contexts. We found that valence, but not concreteness, affects learning. Users learned positive and negative words better than neutral ones. Moreover, positive words are learned best in positive contexts and negative words in more negative contexts. Word and context valence effects are strongest on the learner’s second encounter with the target word and diminish across subsequent encounters. These findings provide support for theories of embodied cognition and the lexical quality hypothesis and point to the linguistic factors that make learning words, and by extension languages, faster.

From lab to field with machine learning – Bridging the gap for movement analysis in real-world environments: A commentary

Biomechanical data collection was largely confined to controlled laboratory setups, relying on marker-based systems or force platforms. However, the emergence of wearable sensors and markerless motion capture has revolutionized this field, enabling data collection in real-world scenarios. This shift has also sparked interest in integrating machine learning (ML) into biomechanical workflows, promising to revolutionize data acquisition in field settings and refine data analysis (Halilaj et al., 2018). Figure 1 provides an overview of corresponding ML applications for key tasks in the biomechanical workflow. This commentary explores the transformative potential of ML in biomechanics, focusing on enhancing data collection and analysis in real-world environments. Pose estimation (a) is the process of automatically tracking and determining the body’s anatomical landmarks, body segments, or joint centre locations in video images using ML, enabling the quantification of human movement without marker and sensor attachments to the human body. Feature estimation (b) employs ML to predict complex biomechanical data, e.g. joint moments from more accessible data sources, including IMUs, pressure insoles, and RGB video cameras. Event detection (c) in time series data is the annotation of certain events that are used to extract useful and vital information or to remove unwanted and unnecessary data for further analysis. Clustering (d) involves grouping instances or individuals with similar biomechanical characteristics using unsupervised ML, thereby revealing underlying structures and subgroups within complex biomechanical data. Finally, automated classification (e) refers to the process of developing a predictive model that assigns input features of data samples to predefined categories or classes using supervised ML. Despite the advancements of ML in biomechanics, central challenges are faced. Estimation errors remain critically high depending on the task and application field, necessitating a careful reflection on data acquisition potentials. Furthermore, especially complex Deep Learning models, while showing promising performances, exhibit a lack of transparency in understanding their decision-making processes and the underlying patterns and rules learned from the data. This phenomenon, often termed as the black-box nature of these models, poses a considerable obstacle. In response, Explainable Artificial Intelligence (XAI), a field that encompasses different explainability approaches to shed light on the inner workings of complex, non-linear ML models, has gained increasing attention in recent years (Slijepcevic et al., 2023). A further central challenge is the availability of data and annotations, which describes the limited availability or insufficiency of relevant and comprehensive datasets, as well as task-specific annotations, for conducting thorough analyses and research in biomechanics. The lack of large-scale benchmark datasets available restricts the widespread adoption of ML-based approaches in biomechanics. Privacy concerns present significant ethical and legal issues, e.g. with identifiable video data. Additionally, model validation remains a critical problem, as many studies fail to validate their models on diverse datasets, often relying on limited data from a single laboratory. Furthermore, there is a risk that the fundamental mechanical understanding of biomechanical processes might be overshadowed by an over-reliance on ML techniques. In conclusion, the integration of ML into biomechanics presents a transformative opportunity for understanding human movement by enabling and improving data collection and analysis in real-world settings. Challenges in data accessibility and methodological transparency necessitate collaborative efforts for future advancements.

Revolutionizing Roads: The Impact of Autonomous Vehicles on Future Travel

This project focuses on developing an autonomous self-driving car by integrating advanced algorithms with cutting-edge hardware. The vehicle is designed to perceive its environment, make informed decisions, and navigate complex traffic scenarios autonomously by leveraging machine learning and computer vision technologies. Essential methods include using deep learning algorithms for object detection and classification, sensor fusion techniques to combine data from cameras, LiDAR, and radar, and reinforcement learning for decision-making in dynamic environments. The project emphasizes the critical role of AI in enabling the car to interpret and respond to dynamic road conditions while ensuring reliable performance through rigorous testing and validation in simulated and real-world scenarios. Initial results from testing indicate that the car can successfully navigate complex traffic situations, such as merging onto highways and responding to pedestrians, with a high degree of accuracy. By pushing the boundaries of autonomous vehicle technology, this research aspires to contribute to the ongoing evolution of self-driving cars and their potential to revolutionize mobility. With a focus on safety, efficiency, and innovation, this endeavor represents a significant step forward in developing autonomous vehicles, offering new possibilities for the future of transportation. The project promises to improve travel convenience and significantly reduce traffic accidents, ultimately contributing to a safer and more efficient road system.

Hybrid Deep Learning Model for Traffic Flow and Speed Forecasting at Urban Junction

The hybrid deep learning model H-CNN-LSTM (Hybrid Convolution Neural Network and Long Short-Term Memory) is introduced in this paper. The model combines CNN to extract traffic flow and speed features, LSTM networks to model temporal dependencies, and employs a combined matrix and a rolling forward input method during data input, enhancing the coarse-grained extraction of traffic data and further refining the utilization of adjacent data. Our research focuses on the complex dynamic traffic conditions of Fuli East Road in Xigu District, Lanzhou City of China. Using detailed traffic video data collected during one month of evening rush hours, we predict future traffic using the H-CNN-LSTM model. By comparing it with other prediction methods, we validate the effectiveness of our approach in predicting traffic flow and speed in real-world traffic scenarios

Digraph k-Coloring Games: New Algorithms and Experiments

We study digraph k-coloring games where strategic agents are vertices of a digraph and arcs represent agents' mutual unidirectional conflicts/idiosyncrasies. Each agent can select, as strategy, one of k different colors, and her payoff in a given state (a k-coloring) is given by the number of outgoing neighbors with a color different from her one. Such games model lots of strategic real-world scenarios and are related to several fundamental classes of anti-coordination games. Unfortunately, the problem of understanding whether an instance of the game admits a pure Nash equilibrium (NE), i.e., a state where no agent can improve her payoff by changing strategy, is NP-complete. Thus, in this paper, we focus on algorithms to compute an approximate NE: informally, a coloring is an approximate γ-NE, for some γ ≥ 1, if no agent can improve her payoff, by changing strategy, by a multiplicative factor of γ. Our contribution is manifold and of both theoretical and experimental nature. First, we characterize the hardness of finding pure and approximate equilibria in both general and special classes of digraphs. Second, we design and analyze three approximation algorithms with different theoretical guarantees on the approximation ratio, under different conditions; (i) algorithm APPROX-1 which computes, for any k ≥ 3, a Δo-NE for any n vertex graph having a maximum outdegree of Δo, in polynomial time; (ii) algorithm LLL-SPE, a randomized algorithm that, for any constant k ≥ 2, determines a γ-NE for some constant γ but only in digraphs whose minimum outdegree is sufficiently large, in polynomial time in expectation; (iii) algorithm APPROX-3 which, for any ε, computes a (1+ε)-NE by using O(log(n)/ε) colors, for any n-vertex digraph. Note that, the latter shows that a (1+ε)-NE exists and can be computed in polynomial time for k = O(log(n)). Finally, to assess how proposed algorithms behave in the typical case, we complete our study with an extensive experimental evaluation showing that, while newly introduced algorithms achieve bounded worst case behavior, they generally perform poorly in practice. Motivated by such unsatisfactory performance, we shift our attention to the best-response paradigm, successfully applied to other classes of games, and design and experimentally evaluate it a heuristic based on such paradigm. Our experiments provide strong evidences of such approach outperforming, in terms of approximation and computational time, all other methods and hence identify it as the most suited candidate for practical usage. More remarkably, it is also able to compute exact, pure NE in the great majority of cases. This suggests that, while these games are known to not always possess a pure NE, such an equilibrium often exists and can be efficiently computed, even by a distributed uncoordinated interaction of the agents.

Escrow-free and efficient dynamic anonymous privacy-preserving batch verifiable authentication scheme for VANETs

To enhance road safety, Vehicular Ad-hoc Networks (VANETs) facilitate the exchange of safety-critical messages between smart vehicles and road traffic authorities. However, VANET’s wireless channels are prone to several attacks, such as replay or modification. Therefore, to protect the links, robust authentication and message integrity mechanisms are required. Previously, several robust authentication schemes have been devised. However, those designs often struggle with complex certificate management, the key escrow problem, and the necessity for secure channels to establish user keys. Additionally, prior methods rely on pseudonyms to ensure user privacy. To implement it, several pseudonyms are stored in the vehicle’s device, which burdens the device. To overcome these limitations, this study introduces an efficient and escrow-free dynamic anonymous authentication scheme tailored for VANETs. By utilizing the paradigm of certificate-based cryptography and fuzzy identity generation, the proposed design eliminates the limitations. Through rigorous security analysis, the proposed design’s effectiveness against various threats is demonstrated. Furthermore, a detailed performance analysis, including computational and communication cost comparisons, showcases the scheme’s feasibility for VANET deployment. An NS-3 simulation further confirms the suitability of the proposed scheme for real-world VANET communication scenarios.

Exploring run-and-tumble movement in confined settings through simulation.

Motion in bounded domains is a fundamental concept in various fields, including billiard dynamics and random walks on finite lattices, and has important applications in physics, ecology, and biology. An important universal property related to the average return time to the boundary, the Mean Path Length Theorem (MPLT), has been proposed theoretically and experimentally confirmed in various contexts. We investigated a wide range of mechanisms that lead to deviations from this universal behavior, such as boundary effects, reorientation, and memory processes. This study investigates the dynamics of run-and-tumble particles within a confined two-dimensional circular domain. Through a combination of theoretical approaches and numerical simulations, we validate the MPLT under uniform and isotropic particle inflow conditions. This research demonstrates that although the MPLT is generally applicable for different step length distributions, deviations occur for non-uniform angular distributions, non-elastic boundary conditions, or memory processes. These results underline the crucial influence of boundary interactions and angular dynamics on the behavior of particles in confined spaces. Our results provide new insights into the geometry and dynamics of motion in confined spaces and contribute to a better understanding of a broad spectrum of phenomena ranging from the motion of bacteria to neutron transport. This type of analysis is crucial in situations where inhomogeneity occurs, such as multiple real-world scenarios within a limited domain.