Deep learning-based human motion recognition for predictive context-aware human-robot collaboration

Timely context awareness is key to improving operation efficiency and safety in human-robot collaboration (HRC) for intelligent manufacturing. Visual observation of human workers’ motion provides informative clues about the specific tasks to be performed, thus can be explored for establishing accurate and reliable context awareness. Towards this goal, this paper investigates deep learning as a data driven technique for continuous human motion analysis and future HRC needs prediction, leading to improved robot planning and control in accomplishing a shared task. A case study in engine assembly is carried out to validate the feasibility of the proposed method.     

Micropatterning of Electrospun Polyurethane Fibers Through Control of Surface Topography

A novel technique was developed to control the deposition of electrospun polyurethane fibers using a silicone collector substrate patterned with soft lithography. This method can be used to control selective fiber deposition with broad pattern dimensions (50–500 µm) over a large area. The combination of ease of use, low cost, tunability, and generation of relatively large fiber mats available with this technique is expected to advance our ability to mimic the orientation and anisotropic properties of native tissues to generate improved tissue engineering scaffolds.

     

Enhanced wear performance of nylon 6/organoclay nanocomposite by blending with a thermotropic liquid crystalline polymer

A new approach for improving the wear performances of nylon 6 (PA6)/clay nanocomposites was examined in this study. Two hybrid nanocomposites were prepared by melt blending a thermotropic liquid crystalline polymer (TLCP) and a well‐dispersed PA6/clay nanocomposite, but with and without the incorporation of maleic‐anhydride grafted polypropylene (MAPP) as compatibilizer. The addition of MAPP improved the compatibility between TLCP and matrix and thus enhanced the fibrillation of dispersed TLCP phase. Wear‐testing results revealed that the wear resistance of the compatibilized hybrid nanocomposite could be improved effectively, as indicated by the low values of specific wear rate and frictional coefficient, especially under high‐normal load (i.e., 80 N). Based on the characterization on the worn damage and the debris, it was suggested that abrasive wear was the main‐damage mechanism for all the materials under investigation, except for the compatibilized hybrid nanocomposite. For this system, the wear damage was caused by a combination of abrasive and adhesive wearing because of the formation of transfer film on the counter pin surface from the wear debris. POLYM. ENG. SCI., 2010. © 2009 Society of Plastics Engineers     

A comparison of seven filling to packing switchover methods for injection molding

The effectiveness of seven methods for controlling switchover from the filling to packing stage were investigated, including: (1) screw position, (2) injection time, (3) machine pressure, (4) nozzle pressure, (5) runner pressure near the sprue, (6) cavity pressure near the gate, and (7) cavity temperature at the end of flow. The activation threshold for each of the seven switchover methods was iteratively determined so as to produce similar part weights relative to a standard process. A design of experiments was implemented for each of the seven switchover methods that perturbs the process settings by an amount equal to six standard deviations of the standard process so as to replicate the expected long‐term process variation. The results suggest that conventional switchover methods (e.g., screw position) had lower short‐term variation, but other methods were more robust with respect to rejecting long‐term process variation. The merits of different dimensional measurements for quality control are also discussed relative to the society of the plastics industry (SPI) standard tolerances. POLYM. ENG. SCI., 50:2031–2043, 2010. © 2010 Society of Plastics Engineers     

Using an additive to control the electrospinning of fibres of poly(ε‐caprolactone)

Electrospinning is a route to polymer fibres with diameters considerably smaller than available from most fibre‐producing techniques. We explore the use of a low molecular weight compound as an effective control additive during the electrospinning of poly(ε‐caprolactone). This approach extends the control variables for the electrospinning of nanoscale fibres from the more usual ones such as the polymer molecular weight, solvent and concentration. We show that through the use of dual solvent systems, we can alter the impact of the additive on the electrospinning process so that finer as well as thicker fibres can be prepared under otherwise identical conditions. As well as the size of the fibres and the number of beads, the use of the additive allows us to alter the level of crystallinity as well as the level of preferred orientation of the poly(ε‐caprolactone) crystals. This approach, involving the use of a dual solvent and a low molar mass compound, offers considerable potential for application to other polymer systems. Copyright © 2010 Society of Chemical Industry     

Fluorous Iminoalditols: A New Family of Glycosidase Inhibitors and Pharmacological Chaperones

A collection of new reversible glycosidase inhibitors of the iminoalditol type featuring N‐substituents containing perfluorinated regions has been prepared for evaluation of physicochemical, biochemical and diagnostic properties. The vast variety of feasible oligofluoro moieties allows for modular approaches to customised structures according to the intended applications, which are influenced by the fluorine content as well as the distance of the fluorous moiety from the ring nitrogen. The first examples, in particular in the D ‐galacto series, exhibited excellent inhibitory activities. A preliminary screen with two human cell lines showed that, at subinhibitory concentrations, they are powerful pharmacological chaperones enhancing the activities of the catalytically handicapped lysosomal D ‐galactosidase mutants associated with GM1 gangliosidosis and Morquio B disease.     

Nonwetting,nonrolling, stain resistant polyhedral oligomeric silsesquioxane coated textiles

Cotton/polyester fabric surfaces were modified using nanostructured organic‐inorganic polyhedral oligomeric silsesquioxane (POSS) molecules via solution dip coating. Surface wetting characteristics of coatings prepared from two chemically and structurally different POSS molecules, a closed cage fluorinated dodecatrifluoropropyl POSS (FL‐POSS) and an open cage nonfluorinated trisilanolphenyl POSS (Tsp‐POSS), were evaluated with time and compared with Teflon. Surface analysis, including Atomic Force Microscopy, SEM/EDAX, and NMR revealed the presence of POSS aggregates on the fabric surface leading to a spiky topography, high roughness, and hysteresis. POSS coated fabrics showed complete reversal of surface wetting characteristics with contact angles higher than the benchmark Teflon surface. Water contact angle measured as a function of time showed equivalent or better performance for POSS‐coated surfaces in comparison to Teflon. Furthermore, FL‐POSS coated fabric exhibited exceptional stain and acid resistance along with a 38% reduction in relative surface friction. Additionally, “nonsliding” and high surface adhesion behavior of water droplets on the FL‐POSS coated fabric are reported. © 2009 Wiley Periodicals, Inc. J Appl Polym Sci, 2010     

Enhanced nanofluidization by alternating electric fields

We show experimental results on a proposed technique to enhance the fluidization of nanoparticle beds. This technique consists of the application of an alternating electric field to the nanofluidized bed. Three different field configurations have been tested: co‐flow field, cross‐flow field, and variable field configurations. Nanoparticle agglomerates are naturally charged by contact and tribo charging mechanisms and therefore are agitated by the action of the externally applied field, which enhance fluidization. According to our observations, the best results are obtained for the variable field configuration. In this configuration, the electric field strength is higher at the bottom of the bed, whereas it is almost negligible at the free surface. Thus, the larger agglomerates, which tend to sink at the bottom of the bed due to stratification, and usually impede uniform fluidization, are strongly agitated. It is thought that the strong agitation of the bigger agglomerates that usually sink to the bottom of the bed contributes to further homogenize the distribution of the gas flow within the bed by destabilizing the development of gas channels close to the gas distributor. On the other hand, the smaller agglomerates at the vicinity of the free surface are just weakly excited. Consequently, fluidization is greatly enhanced, whereas at the same time excessive elutriation is avoided. It is demonstrated that this technique is even suitable to achieve highly expanded fluidization of unsieved nanopowder samples even though the fluidization state returns to be heterogeneous upon the electric field being turned off. © 2009 American Institute of Chemical Engineers AIChE J, 2010     

采用混合智能方法评估球墨铸铁摩擦焊接接头的抗拉强度(英文)

采用混合智能方法评价球墨铸铁的最佳摩擦焊接工艺参数。在自动循环中使用支持向量回归(SVR)、遗传算法(GA)和帝国竞争算法(ICA)优化焊接工艺参数。该方法被用来确定焊接工艺参数,得到了理想的球墨铸铁摩擦焊接抗拉强度。在加热力40kN,加热时间300s,顶锻压力10.12kN条件下,使用SVR加上GA方法得到了最高抗拉强度为256.93MPa。将摩擦焊接样品进行拉伸强度测试,并比较了采用混合智能方法得到的优化值与实验结果。结果表明,混合智能方法可以使ZT14型摩擦焊机拉伸强度从211MPa增加到258MPa。