The digital transformation of production requires methods for integrating, storing, and operationalizing data across organizational boundaries, yet most existing approaches remain siloed and unidirectional, lacking a systematic loop from raw data to actionable knowledge and back. We introduce Data-to-Knowledge (D2K) and Knowledge-to-Data (K2D) pipelines as a universal production concept built on networks of Digital Shadows. The Data-to-Knowledge (D2K) pipeline is realized as a cross-organizational proof of concept that captures and semantically annotates robotic trajectory data from three independent research institutes and uses those data to train an inverse-dynamics foundation model for robot control. Centralized aggregation via an existing FAIR-compliant research data repository was chosen deliberately over federated alternatives to maximize semantic interoperability and reuse of shared infrastructure; federated and privacy-preserving extensions are identified as a promising future direction. Fine-tuning the cross-organizationally trained foundation model reduces training time by approximately 85% relative to end-to-end training from scratch, while achieving comparable accuracy on a standardized inverse-dynamics benchmark. These gains are attributable to the combination of cross-site data aggregation and transfer learning; isolating the contribution of semantic annotation alone remains a topic for future ablation work. The implementation demonstrates that semantically enriched, cross-organizational D2K pipelines can accelerate model development and reduce redundant data collection within a constrained but practically relevant class of robotics tasks. We further discuss limitations, governance challenges, and how these pipelines can contribute to a broader World Wide Lab for collaborative production research.

Articulated robotic arms in laser material processing require precise motion planning. Traditional motion planning methods face challenges in trajectory accuracy. This study demonstrates model-based reinforcement learning as an effective approach for motion planning of these robotic arms. The process involves training a neural network trajectory model based on Pilz Industrial Motion Planner, followed by training an agent to optimize motion by adjusting joint velocities. The study compares Proximal Policy Optimization and Soft Actor-Critic algorithms to the baseline Pilz motion plan. Results show that model-based reinforcement learning improves accuracy in x-direction, reducing mean absolute error to 1.75 x 10(-3) m from 6.37 x 10(-3) m. However, it slightly increases z-direction mean absolute error, from 6 x 10(-6) m to 2.5 x 10(-4) m. This leads to an increase in on-surface beam radius, from 2.9 x 10(-5) m to 3.3 x 10(-5) m, and decrease in peak intensity of 22.77 % compared to baseline. These results highlight reinforcement learning's potential to enhance trajectory accuracy in motion planning, advancing robot-based laser material processing. (c) 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0)

This paper investigates long-short-term-memory-based learning of inverse dynamics for an industrial robot from experimental data. A Franka Emika Robot is used as a case study to estimate the inverse dynamic model. An automated dataset acquisition method is presented. Furthermore, a long-short-term-memory-based neural network was built and optimized with Bayesian hyperparameter search subject to hyperband termination. Both Matlab and Python implementations of the robotics toolbox are used for comparison with conventional baseline methods as well as measured ground truth data. The generated dataset possesses near axis-limit mean centered Gaussian distributions of the command motion parameters (joint position, joint velocity and joint acceleration) and the resulting measured joint torques. Accuracy of the proposed learned inverse dynamics model (mean rmse 0.787 Nm) is comparable with state- of-the-art deep-learning-based approaches and conventional non-deep-learning-based approaches.

This paper investigates an Extended Kalman Filter (EKF) based Sensor Fusion approach for robot tool center point (TCP) position estimation using a sensing unit consisting of multiple sensors. Data from an inertial measurement unit, axis encoders and two new optical sensors for relative speed estimation in the context of laser material processing is recorded. Performance of the approach is tested experimentally. Three different test trajectories are chosen to evaluate estimation performance, including an adaption of ISO 9283 trajectory for robot accuracy. Estimation results are compared to position measurements of a Laser Tracker system with measurement accuracy of +-28 mu m and position estimation of the robot controller of the used Universal Robots UR5e.

Laser Material Processing (LMP) has gained large attraction over the last years due to its high precision, accuracy and wide range of applications. The kinematic systems currently used in LMP are adapted from conventional manufacturing machines. These kinematic systems are usually built in accordance to the size of the structural element whereas mobile robots can be used to produce arbitrarily-sized structural elements. Thus, mobile robots are more flexible in many applications and so they are cheaper to build and have lower operating costs.This work investigates whether it is possible to build a controller for a four-mecanum-wheeled mobile robot so that it can be applied in laser material processing applications. A P, PI, PID and Lyapunov controllers are designed and tested on a real four-mecanum-wheeled mobile robot. All four controllers show reasonable trajectory tracking behavior, but none of the controllers reaches superior performance in comparison to the others, considering the different application test cases. Errors occur mainly due to external disturbances like uneven terrain. Finally, possibilities for improved tracking behavior are provided.

Dual-stage systems find extensive use in the manufacturing sector due to their ability to enable more precise, reliable, and rapid production processes. Laser material processing stands out as a relevant field employing such redundant combinations. This article outlines a new algorithm designed for planning the trajectories of dual stage positioning systems for laser applications utilizing a combination of model predictive and sliding mode control. In comparison to other techniques found in literature, the method presented here offers the advantage of generating feasible trajectories by explicitly considering the kinematic constraints of each actuator. Additionally, this research describes a technique for automatically tuning the parameters of the model predictive problem, aiming to enhance the feed rate of the laser processing. Finally, the approach is validated experimentally showing an improvement of 69.2% in feed rate compared to state of the art.

AbstractThe Internet of Production (IoP) promises to be the answer to major challenges facing the Industrial Internet of Things (IIoT) and Industry 4.0. The lack of inter-company communication channels and standards, the need for heightened safety in Human Robot Collaboration (HRC) scenarios, and the opacity of data-driven decision support systems are only a few of the challenges we tackle in this chapter. We outline the communication and data exchange within the World Wide Lab (WWL) and autonomous agents that query the WWL which is built on the Digital Shadows (DS). We categorize our approaches into machine level, process level, and overarching principles. This chapter surveys the interdisciplinary work done in each category, presents different applications of the different approaches, and offers actionable items and guidelines for future work.The machine level handles the robots and machines used for production and their interactions with the human workers. It covers low-level robot control and optimization through gray-box models, task-specific motion planning, and optimization through reinforcement learning. In this level, we also examine quality assurance through nonintrusive real-time quality monitoring, defect recognition, and quality prediction. Work on this level also handles confidence, verification, and validation of re-configurable processes and reactive, modular, transparent process models. The process level handles the product life cycle, interoperability, and analysis and optimization of production processes, which is overall attained by analyzing process data and event logs to detect and eliminate bottlenecks and learn new process models. Moreover, this level presents a communication channel between human workers and processes by extracting and formalizing human knowledge into ontology and providing a decision support by reasoning over this information. Overarching principles present a toolbox of omnipresent approaches for data collection, analysis, augmentation, and management, as well as the visualization and explanation of black-box models.

AbstractToday’s industrial world is characterized by ever-shortening product development cycles and increasing degrees of product individualization which demand tools and enablers for accelerated prototyping. In addition, the existing uncertainty in the product development cycle should be reduced by involving stakeholders as early as possible. However, should an engineering change request (ECR) be necessary in the product development cycle, a fast iteration step into production is inevitable. The methodological description of such an ECR in the product development cycle is described in the previous chapter. Together with researchers from the Internet of Production (IoP), information from the product development process will be transferred to the digital shadow established in the IoP. The digital shadow collects information from all areas of the product lifecycle and provides it to the appropriate departments, adapted to the corresponding task. To tackle this challenge, a new type of product development process, the method of agile product development, is applied. Within the Enablers and Tools project, the development of various advanced manufacturing technologies (AMTs) for agile product development are at the forefront of the work. The enablers and tools are further developed with the principles of agile product development. They also serve to map the requirements for rapidly available and specific prototypes which are used to answer specific questions that arise during the product development cycle. To answer these questions, the concept of the Minimum Viable Product (MVP), an approach to reduce development time and increase customer satisfaction, is introduced and applied to all development tasks.

Laser-based production systems have become more and more popular in recent years due to their potential to achieve high precision and accuracy in a wide range of different applications. However, the kinematic systems used for laser materials processing (LMP) are often inherited from other production technologies such as milling. The use of mobile robots (MRs) equipped with laser processing optics could disprove the current paradigm of adapted kinematic systems: scaling the size of the material processing system with the size of the components being processed and, thus, the resources used. The trend of autonomous MRs replacing classical kinematic systems in the field of material handling in industrial applications has been evident for years due to their higher flexibility, efficiency, and lower operating costs. In this paper, the prototype of a corresponding MR system is presented. In addition, the general design of the MR is presented. One challenge is the accuracy of an MR; for a common LMP such as laser cutting, the MR must be able to follow a predefined trajectory as accurately as possible. For this purpose, two different measurement systems are presented and compared. To demonstrate the potential of the mobile robot, an exemplary LMP process is also performed and evaluated. Finally, possibilities for improvement or further development, such as integration of scanner optics or the use of several autonomous MRs to increase productivity, are shown.