Fault Classification in Robotic Arm with Ensemble Based Model-Agnostic Meta-Learning

Bootstrapping Self-Supervised Learning of Binary Classification Using Error Bounds: A Case Study on a Robotic Insertion Task

A data engine which gathers data and improves its performance while executing the task, and demonstrates the ability to learn and reduce the need for expensive verifications over time, while staying within the set error-rate.

Electromagnetic Environment Classification for Robust UAV and Robotic Operation Using Ensemble and Deep Learning Models

LLM Ensemble Fault Classification for Automotive HiL Validation

Automotive HiL validation generates large multivariate test recordings whose analysis remains challenging due to manual review effort, rule-based limitations, and the need for explainable diagnostic decisions. Recent machine-learning and deep-learning approaches have improved fault diagnosis, but they often require large labelled datasets, generalise poorly across operating conditions, and provide limited insight into their predictions. This paper proposes an explainable multi-LLM ensemble framework for sensor-level fault classification in automotive validation. The framework uses compact evidence representations of fault-injection recordings and combines the outputs of heterogeneous large language models to improve diagnostic robustness, ranking quality, confidence reliability, and interpretability. The approach is evaluated on gasoline-engine and electric-vehicle HiL systems across three driving settings and ten single-fault classes. Among the individual models, Mistral Small~24B provides the strongest overall single-model trade-off, achieving 0.903 Top-1 accuracy, 0.887 MCC, and the lowest Brier score of 0.102. The final Top-3 ensemble combines Mistral Small~24B, Qwen2.5~32B, and Phi-4~14B using confidence-weighted voting, improving the scenario-averaged results to 0.917 Top-1 accuracy, 0.913 macro F1, and 0.902 MCC, while also providing the best calibration among the tested ensemble strategies. A Top-5 ensemble does not improve over the Top-3 configuration, indicating that model complementarity is more important than ensemble size. The results show that coordinated multi-LLM reasoning can support robust, calibrated, and engineer-interpretable fault classification for automotive HiL validation.

A Consumption-Based Approach for Fault Diagnosis in Multi-DOF Robots

Fault detection and diagnosis (FDI) in multi-degree-of-freedom (multi-DOF) robotic systems is essential for ensuring operational integrity in life-critical applications, such as robotic-assisted surgery and advanced bionics. Traditional methods often struggle with limited data sources and the masking effects of complex motion dynamics on fault localization. The theoretical innovation of this work lies in a novel, hierarchical FDI architecture that synergistically integrates frequency-domain signature modeling with bidirectional temporal learning to decouple motion-induced power fluctuations from subtle fault signals. We utilize the Bode Equation Vector Fitting (BEVF) method to precisely model non-stationary dynamic fault signatures, providing a high-fidelity reference baseline. A two-stage classifier is then employed: a Bidirectional Long Short-Term Memory (BiLSTM) network first localizes faults to a specific joint with 94.4% accuracy by exploiting bidirectional temporal dependencies in the power residuals. Subsequently, a Support Vector Machine (SVM) diagnoses the fault type (mechanical or electrical) with an overall accuracy of 76.3%. This approach successfully identifies high-impact electrical faults while capturing subtle mechanical deviations often masked by the robot’s internal compensatory control loop. Our framework demonstrates a robust and non-invasive solution for FDI, significantly improving diagnostic granularity and providing actionable insights for high-reliability robotic systems.

Target Label-Free Diagnosis of Rotor Bolted Joints via a Hybrid Physical-Data-Driven Partial Domain Adaptation Framework

Flange bolted joints are widely used in the rotor systems of aeroengines. Intelligent fault diagnosis of these bolted joints is crucial for ensuring operational safety. However, due to the extreme scarcity of real-world fault cases, deep learning-based fault diagnosis methods are difficult to apply in engineering practice. Although simulation-to-reality (Sim2Real) transfer learning provides a promising solution, its application in rotor systems faces three challenges: 1) bolt looseness at different locations can excite the similar combination resonance frequencies, making feature differences weak and difficult to distinguish; 2) various inevitable uncertainties introduced during rotor system assembly are difficult to fully account for in mechanical modeling; and 3) the types of faults occurring in actual operation rarely cover all categories present in computational simulations, and this asymmetric label space ( $\mathcal {Y}_{t} \subset \mathcal {Y}_{s}$ ) leads to negative transfer. To overcome these difficulties, this article proposes a novel hybrid physical-data-driven framework, called physics-guided domain randomization and adversarial domain adaptation (Phys-DRADA). This method integrates physical model-based domain randomization (Phys-DR) and frequency-based normalization (F-Norm), eliminating absolute amplitude interference and establishing a robust physical baseline. Subsequently, a physics-guided operational-spatial representation network is proposed. It employs a harmonic-aware dilated spectrum extractor (HADS) to match the equidistant comb-like topologies in the vibration spectrum that shift with operating conditions, and utilizes an operational-spatial coupled transformer to extract the implicit rotor mode shape information from the vibration spectrum and characterize its dynamic variations with rotational speed, thereby distinguishing bolt looseness at different locations. Furthermore, a partial adversarial domain adaptation (PADA) mechanism is introduced to dynamically cutoff the feature alignment of source-private classes. Extensive experimental validations demonstrate that, entirely without prelabeled real-world fault data, Phys-DRADA achieves diagnostic accuracies of 99.33 % and 93.71 % under standard and partial domain adaptation (PDA) scenarios, respectively, providing a highly reliable algorithmic paradigm with practical engineering value for complex rotating machinery.

Development and Modeling of Ensemble Machine Learning Algorithms in Colored Petri Nets for Execution Failure Diagnosis in Robotic Manipulators

The growing adoption of Artificial Intelligence (AI) in industry has automated tasks and interconnected processes, enabling autonomous operational diagnostics. However, due to their often non-transparent nature, machine learning algorithms used in AI (particularly ensemble models) have limited comprehensibility, especially when it comes to combining them for complex applications. This limitation can compromise process reliability and hinder the recognition of failure patterns. To mitigate this limitation, this work investigates the modeling of algorithm combinations in hierarchical colored Petri nets. The feature importance by the permutation technique is also modeled and employed to quantify the contribution of each attribute in the classification process, aiming to provide a clear and dynamic visualization for predicting execution failures in robotic manipulators. This methodology provides a precise identification of the steps executed by the model to make decisions, addressing visualization gaps in robotic manipulator failure applications. The validation was conducted through two experiments: the first explored non-failure-centric scenarios for the developed model, and the second utilized a database already tailored to the problem of interest. In this context, the proposed model achieved results similar to those reported in the literature, demonstrating its applicability. The results of this work aim to improve transparency and trust in automated industrial systems.