Additive manufacturing (AM), defined as the “process of joining materials to make objects from three-dimensional model data, usually layer upon layer, as opposed to subtractive manufacturing methodologies” by the American Society for Testing Materials, has been recognized as a remarkable scientific and industrial technique due to its capacity to produce complex and functional parts directly. AM is a process to build up these parts in a layer-by-layer process involving sequential deposition, melting, fusion, and binding of the successive material layers from three-dimensional design data, and has allowed the reduction of waste material, minimal production lead time, eliminated design constraints, and reduced the cost involved. Many industries are currently adopting AM to exploit these advantages, and AM products are expected to find application in a wide range of fields beyond the medical, aerospace, and mold sectors. It is important to note that not all conventionally manufactured parts can be replaced by AM.

The objective of this special issue is to collect recent research works focused on recent trends in AM. It includes four papers covering the following topics:

• Metal-based powder bed fusion using a laser beam (PBF-LB/M)

• Wire and arc-based AM (DED-Arc/W)

• Material extrusion (MEX)

This issue is expected to help readers understand the recent trends in AM, leading in turn to further research on AM. We deeply appreciate the contributions of all authors and thank the reviewers for their incisive efforts.

Triply periodic minimal surfaces (TPMSs) exhibit excellent mechanical properties. However, their complex geometries impose significant limitations on their fabrication processes. In this study, we fabricate TPMSs using wire and arc-directed energy deposition (DED-Arc) for the first time to leverage its large-scale manufacturing and low production costs. DED-Arc is generally associated with challenges such as distortion caused by high heat input and challenges in fabricating complex geometries. Moreover, TPMS structures feature curved surfaces that include a characteristic overhang in the surface extension direction, i.e., where the width of thin walls increases with height. When fabricating such overhangs, the deposited layers are generally insufficiently tall at the ends of each deposition path. Hence, we propose a novel fabrication strategy incorporating spot deposition, in which localized deposition and cooling are applied repeatedly at the end of each deposition path. Additionally, we devise a method to determine the optimal number of spot depositions based on the overhang angle in the surface extension direction to compensate for the insufficient layer height. Subsequently, we fabricate single- and 27-cell TPMS structures—including Schwarz primitive geometries—using the proposed strategy to evaluate its effectiveness. We evaluate the accuracy of the fabrication process based on X-ray computed tomography measurements, the results of which confirm high geometric fidelity. These findings demonstrate the feasibility of TPMSs fabrication using DED-Arc.

The behavior of bubble and pore formation during selective laser melting of ceramics remains unclear. In this study, in situ observations of a single laser scan over an alumina powder bed were performed using a simple near-infrared visualization system. A CO₂ laser was scanned over a pressurized powder bed under various irradiation conditions, and the interior of the molten pool was recorded with a high-speed camera. The solidified beads were then analyzed by X-ray computed tomography. The observations revealed bubbles forming at the solid–liquid interface and migrating within the molten pool, which were ultimately trapped as pores in the solidified structure. In addition, a qualitative correlation was established between the pore distribution in the solidified structure and the bubble distribution observed in situ. The influence of laser irradiation conditions on pore formation behavior was systematically clarified. These results demonstrate the effectiveness of near-infrared observation techniques in selective laser melting of oxide ceramics.

A high-efficiency, high-quality coating technology for pure copper was developed using a multi-beam metal powder deposition method with a rectangular beam profile at the processing point. Pure copper is a metal with antibacterial properties and high thermal and electrical conductivity. Its use in multi-material applications via metal powder deposition is expected to conserve resources and further expand its applications. In this study, a pure copper layer was formed on a stainless-steel substrate using a blue diode laser with high light absorption in pure copper and a rectangular top-hat beam capable of uniform heating of both the substrate and pure copper powder. Compared with a circular beam, the rectangular beam improved layer formation efficiency up to twofold and was also found to reduce dilution, the mixed layer between the layer and substrate. The pure copper layer formed using the rectangular beam exhibited a smaller difference in dilution between the center and edge of the cross-section compared to the circular beam. This demonstrates that the rectangular beam achieves uniform heating, which is beneficial for layer formation using the multi-beam metal powder deposition method.

Fused filament fabrication is a type of material extrusion process category of additive manufacturing. Because the material is prepared in the form of filaments, a wide variety of materials including composite materials with continuous fibers can be used. With the recent increase in environmental consciousness, the use of biodegradable composites using polylactic acid (PLA) and continuous cellulose nanofiber (CNF) strings has garnered attention from the scientific community. However, the fiber content of the composites remains low, preventing full exploitation of the ultrahigh intrinsic strength of CNFs. In this study, we proposed two methods to increase fiber content. First, we modified the cross-sectional shape of the filament from circular to rectangular to increase the fiber content. Second, we optimized the process parameters during filament fabrication by adjusting the feed rates of continuous CNFs and PLA, resulting in a higher fiber content up to 35%. Following this, filaments with higher fiber content were used in fabrication experiments. Tensile tests of the fabricated specimens showed that the mechanical strength increased with the fiber content. Finally, we present an application example of the developed CNF filament for reinforcing finger orthosis.

Ensuring consistent mesh geometry is essential for mechanical reliability and quality control of knotted nets in factory production line inspections. This paper presents a fully automated visual inspection system that integrates knot detection, mesh topology reconstruction, and real-world dimensional measurements within a distortion-aware deep learning framework. The knots were detected using a YOLOv11-based object detector trained to localize small and densely arranged targets under varying illumination, deformation, and partial occlusion. To enable accurate metric measurements over large net areas, the images captured using a 150° ultrawide-angle lens were rectified through camera calibration and distortion correction. Knot centroids were extracted from the detected bounding boxes and organized using a row-wise directional linking strategy that reconstructed a diamond-shaped mesh topology while avoiding physically implausible connections. Interknot distances were computed in the rectified image plane and converted into real-world measurements using a K-nearest neighbors (KNN) regression model, which compensated for the residual nonlinearities remaining after distortion correction. The scope of inspection in this study was limited to mesh size and knot spacing evaluation. However, missing knots, broken strands, or other structural defects may affect the dimensional consistency of the net and may be indicated indirectly through abnormal mesh measurements. The proposed system is not intended for direct detection or classification of such defects. It was evaluated on a custom dataset of 1,054 industrial fishing net images, achieving a high detection performance with an mAP50 of 0.99 and mAP50–95 of 0.86. The calibration stage achieved distance estimation errors within ±3.53% relative to the nominal mesh size. The results demonstrated that the proposed approach enabled the automated and quantitative assessment of mesh uniformity, providing a practical and scalable solution for industrial net quality inspection.

In various manufacturing processes, feedback control systems have been investigated to enhance productivity. However, applications in practical mass production are still limited. This study aimed to develop a closed-loop feedback system involving a new die set for nonstop mass production using a transfer press machine and to demonstrate its effectiveness in the real production of millions of pieces. Displacement and load sensors were embedded in the die set to monitor the product dimension and status of the upsetting punch. An actuator was also installed to adjust the product dimension controlled by threshold-type discrete feedback. Complex wiring from the sensors and their power sources were replaced with dedicated wireless data and power transmission devices to improve the mountability and operability when changing the die set. In the demonstration tests, the product thickness was precisely adjusted to 0.016 mm using the actuator controlled with a feedback command of 0.02 mm. The proposed system has been successfully used for the production of more than 6.5 million pieces during the period from the middle of 2023 to the end of 2025, demonstrating good durability and practicality. To detect a partial punch fracture, the Mahalanobis method is effective even when the fracture area is relatively small. To apply this method when a step-like change in the Mahalanobis distance is inevitable owing to interruptions according to the production schedule, the training data of the first several hundreds of shots need to be used for the production schedules.

In-process measurement of changes in grinding wheel condition is expected to prevent machining defects. A method for monitoring grinding wheel condition through grinding vibration analysis was examined in this study. To achieve this, a measurement device was developed by housing two accelerometers, a compact microcontroller, a wireless transmitter, and two batteries in an acrylic case mounted on the grinding wheel, enabling vibration acquisition and wireless transmission during operation. Two measurement methods were implemented in the developed system. The first is a raw data transmission method, which measures biaxial acceleration at a sampling frequency of 24 kHz and simultaneously transmits the data to a personal computer (PC). The second is an edge computing method, which calculates anomaly scores using a neural network within the microcontroller and transmits only the results. The advantage of the raw data transmission method is that it allows various analyses using the raw acceleration data received by the PC. However, the disadvantage is the heavy load on the network due to the large volume of data transmitted. In contrast, the edge computing method significantly reduces data volume and power consumption by transmitting only anomaly scores, thereby extending battery life. Experiments were conducted using a surface grinding machine. In the raw data transmission method, it was found that the integrated value of the absolute acceleration measured by the developed device strongly correlates with the magnitude of the grinding force. This result indicates that the developed device can estimate the grinding force. In the edge computing method, the anomaly scores calculated within the microcontroller correlated with the grinding force. This confirms that the developed device can estimate the grinding force using this method as well. Furthermore, it was demonstrated that this method can detect abnormalities in wheel rotational speed, changes in wheel condition, and the occurrence of grinding burn.

Heat generated in machine tools during operation causes thermal deformation and deteriorates machining accuracy. To suppress thermal deformation and the resulting displacement, strategies such as reducing heat generation, cooling machine elements that generate heat, and compensating for thermal displacement are generally adopted. However, existing methods for estimating and compensating for thermal deformation and displacement can only be adopted in limited situations. In particular, there are few reports on real-time estimation methods that take into account moving parts and actual machine operating conditions. For high-accuracy machining, it is important to understand the temperature field and the associated displacement. In this study, real-time estimation of the temperature field and displacement is carried out on a test apparatus equipped with a ball screw drive system and a linear guideway system. In the estimation process, heat generation values are first calculated from servo data corresponding to the machine operation. Using these calculated values, the temperature field is then derived through the finite volume method. Finally, the thermal displacement is obtained from the temperature field using a simplified estimation method. The results are validated by comparison with experimental data and with results from commercial finite element method software, demonstrating the effectiveness and real-time performance of the proposed approach.

Environmental temperature changes are a major factor in machining accuracy degradation in machine tools. As maintaining a constant environmental temperature is costly, reducing thermal deformation through machine design is important. For this purpose, it is essential to clarify the mechanisms of thermal deformation caused by environmental temperature variations. In recent years, machine tools have increasingly been required to incorporate covers for chip and coolant management as well as compliance with safety regulations. However, covers influence the heat transfer between the machine and the environment and complicate the mechanism of thermal deformation. In this study, the influence of covers on the heat transfer and thermal displacement of a turning center was investigated. The relationship between thermal displacement and temperature variations in the ambient air and machine structure was examined for different cover configurations under environmental temperature variations. Experimental results demonstrated that, while the influence of environmental temperature changes was suppressed by the machine outer cover, the temperature difference of the air within the machine outer cover increased due to the mechanical configuration, resulting in thermal deformation. Additional covers suppressed the temperature rise of the air around the structure and the structure itself, but the temperature distribution of the air and the structure was not improved, and thermal deformation was not reduced. Simulation results showed that the heat transfer rate from the air to the structure was reduced by 20% with additional covers. These results indicate that thermally robust machine tools under environmental temperature changes can be developed by appropriately designing the arrangement of covers.

Separating moving and static objects from a moving camera viewpoint is essential for 3D reconstruction, autonomous navigation, and scene understanding in robotics. Existing approaches often rely primarily on optical flow, which struggles to detect moving objects in complex, structured scenes involving camera motion. To address this limitation, we propose Focus of Expansion Likelihood and Segmentation (FoELS), a method based on the core idea of integrating both optical flow and texture information. FoELS computes the focus of expansion (FoE) from optical flow and derives an initial motion likelihood from the outliers of the FoE computation. This likelihood is then fused with a segmentation-based prior to estimate the final moving probability. The method effectively handles challenges including complex structured scenes, rotational camera motion, and parallel motion. Comprehensive evaluations on the DAVIS 2016 and FBMS-59 datasets, along with real-world traffic videos including parallel, cross-direction, opposite-direction, and crowded scenes, demonstrate its effectiveness and state-of-the-art performance.