Adhesion always forms on the tool surface during cutting of carbon steel. The periodic evolution (formation, growth and fall-off) of adhesion has the significant effects on the cutting process, such as tool wear, finished surface roughness and so on. Because adhesion formed on the tool surface can keep the tool away from the workpiece and protect the tool from wear to some extent, so we can make use of it to realize the low-wear cutting. The aim of this study is to predict the periodic evolution of adhesion with less cutting experiment data. The model validation is performed by comparing simulation with experimental results. From the analyses, it is found that the adhesion damage model can predict the adhesive state at the tool workpiece interface with considerable accuracy. Furthermore, the relationships between the frequency of adhesion and tool wear, finished surface roughness were also discussed.

TiO₂ has unique characteristics of high chemical stability, photocatalyst etc. Crystallographically oriented or aligned nano-rod structure is preferable considering the uniformity of function or aiming at the higher performance in the application such as piezoelectric devices though it is difficult to produce. This study demonstrates a simple fabrication process for vertically aligned TiO₂ structure using hydrothermal synthesis. A fluorine-doped tin oxide (FTO) substrate was used as a substrate because its lattice constant is close to that of rutile-TiO₂. The morphology or thickness of the structure was controlled by changing synthesis conditions such as source material concentration. Vertically aligned TiO₂ nano-rods were successfully fabricated of which diameter and height was about 300 nm and 5 μm respectively after four hours synthesis. The variation of the height was examined because the smoothness is sometimes important in the application. Water contact angle on the structure was measured as about 10° and the hydrophilicity persisted for long time comparing with a glass substrate.

In polymer injection molding, to achieve surface appearance resembling metallic coating without painting the molded product, bright material such as aluminum flake is sometimes mixed into the resin. In such cases, the generation of weld lines which appear as lines with different gloss and coloring are formed at the flow front meeting area of molded products and the vicinity areas. In this study, we used an electromagnetically induced heating and cooling injection mold equipped with system to control the melt flow to mold polypropylene containing aluminum flakes. As a result, it was clarified that weld lines with inconsistent coloring can be reduced by controlling the melt flow while heating the mold, and that the preventive effect of the above weld lines improves by a rise of the heating temperature of the cavity wall, advance at the timing of melt flow control starting and increase of a bright flake diameter.

In a rough lapping process of substrates made of SiC or sapphire, a cast-iron lap is generally used. However, the cast-iron lap has some problems, which it has low corrosion resistance and low removal rate. We developed a fiber lap as an alternative lapping tool for solving these problems. The fiber lap is a tool containing stainless-steel fibers in the epoxy resin. A stainless-steel mesh (called as a “mesh lap”) was used as a lapping tool to optimize the dimensions of the fibers. As a result, the removal rate of the mesh lap was very high compared with the cast-iron lap. Then, the lapping mechanism on which a high removal rate could be obtained was investigated. The lapping performances of the mesh lap can be changed depending on the wire diameter and the pitch interval. The difference in the lapping performances was examined by observing the moving behavior of the abrasive grains on the tool. It was found that there were two types of abrasive-grain behaviors, meandering motion and linear motion. Finally, by reflecting the results obtained using the mesh lap on the fiber lap, we improved the lapping performances of a fiber lap.

A machining error correction, which is based on the predicted machining error caused by elastic deflection of tool system, is proposed. This method is based on an instantaneous force with static deflection feedback model which is classified to consider the milling process representation. The uncut chip thickness, which is required to predict cutting force, is calculated under the consideration of the elastic deflection of the tool system, which corresponds to tool and tool holder deflections caused by cutting force. In our previous study, the uncut chip thickness is calculated from voxels rem oved by each minute tool rotational angle, and the instantaneous cutting force in a minute time interval can be predicted. Furthermore, in our previous study, the elastic deflection of the tool system caused by the predicted instantaneous cutting force can be calculated. This study proposes the method to correct the tool position and posture according to the predicted horizontal displacement and the rotational displacement derived from the elastic deflection of the tool system. In order to validate the effectiveness of the proposed algorithm, the experimental milling operations were conducted. The experimental results showed that the successful error corrections can be performed. The correction of the tool position and posture according to the predicted elastic deflection of the tool system can achieve the improvement of machining accuracy and efficiency.

Pixel-based dense template matching still plays an important role in object localization tasks especially in the case of feature-less sequences. Instead of exhaustive search methods, optimization-based search strategy can converge to the global optimum faster by avoiding unnecessary tests. However, due to the presence of local optimums, conventional methods usually fail to converge in practice. In this paper, in order to prevent the optimization procedure from falling into a local optimal solution, we introduce a novel evolutionary operation called probabilistic bit-wise operation (PBO) into the framework of genetic algorithm (GA). Specifically, utilizing a natural phenomenon that the change of a higher-bit in an individual affect more than the lower-bit, the diversity of a population-based evolution algorithm can be well controlled by flipping each bit with different probabilities being assigned. Also, unlike crossover in GA, PBO only requires a single individual as the input, thus the parallelization can be easily implemented without considering the dependence between individuals. In the experiment, we compare against several classic optimization methods such as particle swarm optimization, GA, particle filter to show the superiority of PBO with a specific simulated benchmark. We also apply our method to a real sequence to show the practicality.

In this paper, we propose a method to estimate 6D pose (3D position and pose) of the everyday objects, especially Kitchen or DIY tools, even if there are not the same 3D model of the target object. We assume that the tools that belong to same category have common spatial relation in part-affordances, the role of each part. The proposed method uses this assumption for pose estimation. By minimizing two types of consistencies, the spatial relationship of part-affordances and the shapes, we have confirmed that the proposed method achieved higher performance compared with traditional pose estimation method.

With the progress of the numerical control unit, machine tool's positioning accuracy covers nano meter order. In addition, the monitoring functions of the current power and the deviation between the present position and the desired one enable to check the condition during machining and use the improvement of not only the machining accuracy but also the cutting conditions. This NC controller enables to acquire the NC table's condition from the various data during motion. However, there still remains another demand that the open NC system for NC table which enables to install an original positioning programing. In this study, a new open numerical control unit based on the FPGA (field-programmable gate array) technique and the real time OS has been developed to identify the dynamic characteristics' parameters during motion. A motion table driven by the ball screw drive system is chosen as the target object and controlled by an original numerical control system which realizes the inverse transfer function control method with the originally developed software. This control unit is able to control at 1ms period with recording the present positions, velocity and torque current etc. The semi-closed positioning control is applied for a motion table by the servomotor. In this report, the experimental control results of the proposed control method are evaluated by the real-time measurement of the position set-point and the present position during the table motion.

In the study, a novel design and fabrication method which is simple, optimized press molding for CFRP on the basis of CAD data is proposed. CFRP has been widespread use for the purpose of weight reduction of transportation equipment in recent years. However, in order to expand range of purpose, reduction of production is required. To satisfy these requirements, the authors focused on press molding technique. Provided that, partial excessive or breakage of the fiber occurred due to drawled fiber by deforming force. As a solution for these problems, the authors proposed the design and fabrication method for CFRP preform which has unfolded diagram shape of objective three-dimensional model using a tow prepreg. As the result, the calculation method for generation of unfolded diagram was proposed. And the validity of unfolded diagram was confirmed by reproducing the diagram to three-dimensional shape.