In Industry 4.0 applications, dynamic environmental interference induces highly nonlinear and strongly coupled interactions between the environmental state and robotic behavior. Effectively representing dynamic environmental states through multimodal sensor data fusion remains a critical challenge in current robotic datasets. To address this, an industrial-grade multimodal interference dataset is presented, designed for robotic perception and control under complex conditions. The dataset integrates multi-dimensional interference features including size, color, and lighting variations, and employs high-precision sensors to synchronously collect visual, torque, and joint-state measurements. Scenarios with geometric similarity exceeding 85\% and standardized lighting gradients are included to ensure real-world representativeness. Microsecond-level time-synchronization and vibration-resistant data acquisition protocols, implemented via the Robot Operating System (ROS), guarantee temporal and operational fidelity. Experimental results demonstrate that the dataset enhances model validation robustness and improves robotic operational stability in dynamic, interference-rich environments. The dataset is publicly available at:https://modelscope.cn/datasets/Liaoh_LAB/Liaohe-CobotMagic-PnP.

Due to high-mix-low-volume production, sheet-metal workshops today are challenged by small series and varying orders. As standard automation solutions tend to fall short, SMEs resort to repetitive manual labour impacting production costs and leading to tech-skilled workforces not being used to their full potential. The COOCK+ ROBUST project aims to transform cobots into mobile and reconfigurable production assistants by integrating existing technologies, including 3D object recognition and localisation. This article explores both the opportunities and challenges of enhancing cobotic systems with these technologies in an industrial setting, outlining the key steps involved in the process. Additionally, insights from a past project, carried out by the ACRO research unit in collaboration with an industrial partner, serves as a concrete implementation example throughout.

Figure 1: Remote Human-Robot-Collaboration: a) A remote operator needs to align the body of an engine so that a robot arm can access and weld it (a linear arrangement of white points represents the required welding locations). Through this, the user controls the position and rotation of the engine, enabling her to align the engine efficiently. The concentration of unreachable locations along the task area's right side indicates to the user the need to rotate the engine further toward the robot. Human-Robot-Collaboration can enhance workflows by leveraging the mutual strengths of human operators and robots. Planning and understanding robot movements remain major challenges in this domain. This problem is prevalent in dynamic environments that might need constant robot motion path adaptation. Through a user study (n=20), we indicate the strength of the visualization relative to a point-based reachability check-up. Collaboration between human operators and robots can leverage the strengths of both. Humans can better understand ad hoc situations and control them so that they are easily accessible by the robot.

Scientific computing for large deformation of elastic-plastic solids is critical for numerous real-world applications. Classical numerical solvers rely primarily on local discrete linear approximation and are constrained by an inherent trade-off between accuracy and efficiency. Recently, deep learning models have achieved impressive progress in solving the continuum mechanism. While previous models have explored various architectures and constructed coefficient-solution mappings, they are designed for general instances without considering specific problem properties and hard to accurately handle with complex elastic-plastic solids involving contact, loading and unloading. In this work, we take stretch bending, a popular metal fabrication technique, as our case study and introduce LaDEEP, a deep learning-based surrogate model for \textbf{La}rge \textbf{De}formation of \textbf{E}lastic-\textbf{P}lastic Solids. We encode the partitioned regions of the involved slender solids into a token sequence to maintain their essential order property. To characterize the physical process of the solid deformation, a two-stage Transformer-based module is designed to predict the deformation with the sequence of tokens as input. Empirically, LaDEEP achieves five magnitudes faster speed than finite element methods with a comparable accuracy, and gains 20.47\% relative improvement on average compared to other deep learning baselines. We have also deployed our model into a real-world industrial production system, and it has shown remarkable performance in both accuracy and efficiency.

Artificial intelligence (AI) is of fundamental importance in Industry 4.0. The analysis of sensor data with AI can be utilised for the reliable recognition of complex patterns in real time, which is often a challenging task for humans [1]. For example, in predictive maintenance, AI may in this way help to identify and replace machine parts before they break. More generally, main goals in predictive maintenance are to reduce production downtime and lowering the risk of damages in a factory [2, 3, 4], which may require an exact monitoring of the status of the factory and its processing of workpieces. Other possible applications of AI in Industry 4.0 include robot automatisation, supply chain optimisation and quality control [5, 6]. The latter is significant to maintain a high-level standard and to ensure that there are no harmful components or substances introduced into a production process. Companies are facing the challenge of adopting the concepts of Industry 4.0 in their operations. To foster this development, the use of learning factories may be considered. A learning factory is a model in which learners can develop an understanding of practical problems from the real world, without tinkering with a real factory process [7].

This paper presents a CAD-based approach for automated surface defect detection. We leverage the a-priori knowledge embedded in a CAD model and integrate it with point cloud data acquired from commercially available stereo and depth cameras. The proposed method first transforms the CAD model into a high-density polygonal mesh, where each vertex represents a state variable in 3D space. Subsequently, a weighted least squares algorithm is employed to iteratively estimate the state of the scanned workpiece based on the captured point cloud measurements. This framework offers the potential to incorporate information from diverse sensors into the CAD domain, facilitating a more comprehensive analysis. Preliminary results demonstrate promising performance, with the algorithm achieving convergence to a sub-millimeter standard deviation in the region of interest using only approximately 50 point cloud samples. This highlights the potential of utilising commercially available stereo cameras for high-precision quality control applications.

Robotic systems are increasingly employed for industrial automation, with contact-rich tasks like polishing requiring dexterity and compliant behaviour. These tasks are difficult to model, making classical control challenging. Deep reinforcement learning (RL) offers a promising solution by enabling the learning of models and control policies directly from data. However, its application to real-world problems is limited by data inefficiency and unsafe exploration. Adaptive hybrid RL methods blend classical control and RL adaptively, combining the strengths of both: structure from control and learning from RL. This has led to improvements in data efficiency and exploration safety. However, their potential for hardware applications remains underexplored, with no evaluations on physical systems to date. Such evaluations are critical to fully assess the practicality and effectiveness of these methods in real-world settings. This work presents an experimental demonstration of the hybrid RL algorithm CHEQ for robotic polishing with variable impedance, a task requiring precise force and velocity tracking. In simulation, we show that variable impedance enhances polishing performance. We compare standalone RL with adaptive hybrid RL, demonstrating that CHEQ achieves effective learning while adhering to safety constraints. On hardware, CHEQ achieves effective polishing behaviour, requiring only eight hours of training and incurring just five failures. These results highlight the potential of adaptive hybrid RL for real-world, contact-rich tasks trained directly on hardware.

This paper delves into various robotic manipulation control methods designed for dynamic contact tooling operations on a robotic repair platform. The explored control strategies include hybrid position-force control, admittance control, bilateral telerobotic control, virtual fixture, and shared control. Each approach is elucidated and assessed in terms of its applicability and effectiveness for handling contact tooling tasks in real-world repair scenarios. The hybrid position-force controller is highlighted for its proficiency in executing precise force-required tasks, but it demands contingent on an accurate model of the environment and structured, static environment. In contrast, for unstructured environments, bilateral teleoperation control is investigated, revealing that the compliance with the remote robot controller is crucial for stable contact, albeit at the expense of reduced motion tracking performance. Moreover, advanced controllers for tooling manipulation tasks, such as virtual fixture and shared control approaches, are investigated for their potential applications.

This study presents a novel approach for predicting wall thickness changes in tubes during the nosing process. Specifically, we first provide a thorough analysis of nosing processes and the influencing parameters. We further set-up a Finite Element Method (FEM) simulation to better analyse the effects of varying process parameters. As however traditional FEM simulations, while accurate, are time-consuming and computationally intensive, which renders them inapplicable for real-time application, we present a novel modeling framework based on specifically designed graph neural networks as surrogate models. To this end, we extend the neural network architecture by directly incorporating information about the nosing process by adding different types of edges and their corresponding encoders to model object interactions. This augmentation enhances model accuracy and opens the possibility for employing precise surrogate models within closed-loop production processes. The proposed approach is evaluated using a new evaluation metric termed area between thickness curves (ABTC). The results demonstrate promising performance and highlight the potential of neural networks as surrogate models in predicting wall thickness changes during nosing forging processes.

Regression is a fundamental prediction task common in data-centric engineering applications that involves learning mappings between continuous variables. In many engineering applications (e.g.\ structural health monitoring), feature-label pairs used to learn such mappings are of limited availability which hinders the effectiveness of traditional supervised machine learning approaches. The current paper proposes a methodology for overcoming the issue of data scarcity by combining active learning with hierarchical Bayesian modelling. Active learning is an approach for preferentially acquiring feature-label pairs in a resource-efficient manner. In particular, the current work adopts a risk-informed approach that leverages contextual information associated with regression-based engineering decision-making tasks (e.g.\ inspection and maintenance). Hierarchical Bayesian modelling allow multiple related regression tasks to be learned over a population, capturing local and global effects. The information sharing facilitated by this modelling approach means that information acquired for one engineering system can improve predictive performance across the population. The proposed methodology is demonstrated using an experimental case study. Specifically, multiple regressions are performed over a population of machining tools, where the quantity of interest is the surface roughness of the workpieces. An inspection and maintenance decision process is defined using these regression tasks which is in turn used to construct the active-learning algorithm. The novel methodology proposed is benchmarked against an uninformed approach to label acquisition and independent modelling of the regression tasks. It is shown that the proposed approach has superior performance in terms of expected cost -- maintaining predictive performance while reducing the number of inspections required.