Jul 2026· International Scientific Unity· 0 citations
TL;DR
The paradigm shift from testing for correct instruction execution to testing for robust behavior under open‑world conditions is analyzed and practical recommendations for integrating metamorphic testing and formal robustness verification into existing validation pipelines for mPNT systems are concluded.
Abstract
The integration of AI‑based components (deep learning odometry, AI‑driven fault
detection, end‑to‑end localization) into multi‑source Positioning, Navigation and
Timing (mPNT) systems fundamentally changes the nature of testing. Unlike
traditional deterministic algorithms, neural networks operate as “black boxes” trained
on finite datasets and can fail unpredictably on unseen inputs. This paper analyzes the
paradigm shift from testing for correct instruction execution to testing for robust
behavior under open‑world conditions. Three core challenges are examined: the oracle
problem, out‑of‑distribution vulnerability, and the risk of “hallucinations”. A
comparative analysis of testing methods for classical and AI‑based navigation components is provided, followed by a detailed case study of the OdoTest framework
for deep odometry. The paper concludes with practical recommendations for
integrating metamorphic testing and formal robustness verification into existing
validation pipelines for mPNT systems.
The rapid evolution of software complexity demands more efficient and autonomous testing mechanisms. Artificial intelligence (AI) has emerged as a solution to the limitations of traditional manual testing in software development, which is time-consuming, prone to human error, and unable to scale with the increasing size and complexity of modern software systems. In this context, this paper presents an application-focused review of 35 selected empirical studies focusing on the use of AI during software testing, based on PRISMA guidelines. We introduce a comprehensive taxonomy categorizing current research into six core fields, including test case generation, defect prediction, and AI model verification. The analysis reveals that large language models, machine learning, and computer vision can significantly improve testing efficiency. Key findings demonstrate that AI can autonomously repair broken test scripts, generate robust synthetic data, enable codeless web testing, and accurately predict system defects before execution. Furthermore, advanced techniques such as reinforcement learning and deep learning successfully validate complex environments, including cloud robotics and quantum software. However, our qualitative and quantitative synthesis also highlights that challenges, such as generative AI “hallucinations” and the brittleness of Continuous Integration and Continuous Deployment (CI/CD) integration, persist. Ultimately, this review proposes a tailored research roadmap for robust industrial adoption, showing that AI is changing the way software is tested, shifting it from a predominantly reactive and static activity toward a proactive, intelligence-driven discipline.
Guilherme Martins, Nelson N. Tenório, Jorge Bernardino· Big Data and Cognitive Compu...· 1 citation
The integration of artificial intelligence into instrumentation and measurement systems has emerged as a transformative force across industrial, environmental, and scientific domains. This article provides a systematic overview of AI applications in instrumentation, encompassing intelligent sensor data processing, predictive maintenance, automated meter reading, and fault diagnosis. Drawing upon recent advances documented in the literature, we examine the methodological landscape ranging from conventional machine learning to deep learning and large language models. Key benefits include enhanced measurement accuracy through intelligent compensation, reduced downtime through predictive maintenance, and improved operational efficiency through automation. However, significant challenges persist, including the blackbox nature of AI models, data scarcity, uncertainty quantification, and the gap between laboratory performance and realworld deployment. We argue that the future of intelligent instrumentation lies in hybrid approaches that integrate datadriven AI with conventional modeldriven methods, thereby combining the pattern recognition capabilities of AI with the interpretability and rigor of physicsbased models.
Fujie Lu· Academic Journal of Manageme...· 0 citations
Deploying AI models for Non-Destructive Testing (NDT) in production environments remains challenging due to the dynamic nature of industrial conditions. Over time, data drift caused by process variations alters input distributions; label noise arises from inconsistent or imperfect expert annotations; and unseen scenarios introduce new defect types or signal patterns not represented during training. Together, these factors lead to progressive performance degradation, reducing model reliability and operator trust. In an industry such as NDT, where inspection reliability directly impacts structural integrity and safety, model failure is simply not acceptable.
To address this challenge, this study introduces an innovative human-centric framework that synergistically integrates incremental and active learning strategies in complex NDT inspections. Incremental learning methodologies are specifically designed to enable models to continuously adapt to evolving data distributions while preserving previously acquired knowledge. At the same time, AI-assisted inspection strategies based on active learning allows NDT inspection to maintain high levels of automation with a central role of the operator in the loop.
Results show that the proposed approach provides critical AI-based support, enabling inspectors to certify component acceptance or rejection with greater reliability and confidence. This facilitates next-generation NDT 5.0 systems and empowers inspectors in highly automated environments.
J. Mendikute, Jose Luis Lanzagorta, I. Sanchez et al.· e-Journal of Nondestructive...· 0 citations
Flexible manufacturing requires rapid deployment of solutions and minimal setup time to remain competitive. An essential attribute is the ability to control error levels, as failures can range from minor performance degradation to severe equipment damage. However, conventional deployment often involves extensive setup, data collection, model training or parameter tuning, and system testing, resulting in significant delays that hinder commercial feasibility. We propose a data engine which gathers data and improves its performance while executing the task. The data engine consists of two classifiers, a fast model prediction and expensive verification. First, a model prediction is performed and based on the confidence level of the prediction, the expensive verification can be used. By adjusting the confidence level, users can control the level of tolerable error. Our method is implemented on a real-world robotic insertion task, which uses force data for the model prediction. The system applies UMAP dimensionality reduction and uses Wilson-Score to compute the confidence bounds of the prediction. Results demonstrate the ability to learn and reduce the need for expensive verifications over time, while staying within the set error-rate. The results highlight the potential of confidence bounds in self-improving models to enhance reliability in robotic classification task.
Zebin Duan, Norbert Krüger, Juan Heredia et al.· 0 citations
Autonomous driving systems rely on a sophisticated pipeline of artificial intelligence models to perceive, predict, and plan in dynamic environments. This review presents a systematic analysis of the machine learning and deep learning models underpinning vehicle autonomy, spanning classical convolutional neural networks (CNNs) for object detection and semantic segmentation to recurrent and Transformer-based architectures for trajectory prediction and motion planning. It also provides a critical examination of the autonomous vehicle sensor stack, including cameras, LiDAR, radar, ultrasonics, and GNSS/IMU as data acquisition systems, highlighting modality-specific AI challenges such as monocular depth estimation, 3D point cloud processing, and radar Doppler interpretation. The evolution of perception and decision-making pipelines is reviewed, contrasting modular architectures with end-to-end learning paradigms that directly map raw sensor data to control commands, and discussing their trade-offs in interpretability, safety assurance, and robustness to rare edge cases. We further survey specialized hardware accelerators and heterogeneous automotive SoCs designed to meet stringent real-time and power constraints. Industrial strategies are compared, including multi-modal sensor fusion and vision-centric approaches based on large-scale imitation learning. Finally, we identify open challenges related to robustness under adverse conditions, domain shift, causal ambiguity, and the need for interpretable and certifiable AI in safety-critical autonomous driving systems.
Esraa Khatab, Fares Fathy, Abdallah AlKholy et al.· Machine Learning and Knowled...· 0 citations
To maintain high efficiency and reduce operational downtime in industrial manufacturing, effective Predictive Maintenance (PdM) for robotic manipulators is essential. Although combining Model-Agnostic Meta-Learning (MAML) with digital twin technology offers a solid basis for quickly identifying faults, conventional methods often face challenges regarding parameter sensitivity and generalizing to new domains. To mitigate these issues, we introduce an ensemble-based metalearning framework that combines MAML with majority voting and operational grouping. This methodology improves generalization, stabilizes performance across diverse conditions, and strengthens few-shot learning capabilities. We validated the framework using a synthetic vibration dataset generated via a digital twin to simulate various robotic arm faults. Our findings demonstrate that this method achieves 93.8% accuracy and 93.1% precision in the ten-shot regime, outperforming the MAML baseline by 11.1%, across a broad range of mechanical faults, showing strength in cross-domain few-shot (CDFS) scenarios. Comparisons with established frameworks - including Reptile, Protonet, and ANIL, confirm the effectiveness of our model. By employing ensemble learning, we attain greater robustness and classification accuracy, establishing the method as a viable solution for industrial PdM. Furthermore, the integration of digital twins bridges the gap between simulation and real-world deployment, reducing data dependency and enabling effective fault classification even in dynamic environments with limited labeled data.
Mainak Mallick, Seung-Kyum Choi· 2026 6th International Confe...· 0 citations