ALBERT

Description: Travel time prediction is critical for advanced traveler information systems (ATISs), which provides valuable information for enhancing the efficiency and effectiveness of the urban transportation systems. However, in the area of bus trips, existing studies have focused on directly using the structured data to predict travel time for a single bus trip. For state-of-the-art public transportation information systems, a bus journey generally has multiple bus trips. Additionally, due to the lack of study on data fusion, it is even inadequate for the development of underlying intelligent transportation systems. In this paper, we propose a novel framework for a hybrid data-driven travel time prediction model for bus journeys based on open data. We explore a convolutional long short-term memory (ConvLSTM) model with a self-attention mechanism that accurately predicts the running time of each segment of the trips and the waiting time at each station. The model is more robust to capture long-range dependence in time series data as well.

Description: It was previously shown that Bone Morphogenetic Protein (BMP)-9 is constitutively produced and secreted by hepatic stellate cells (HSC). Upon acute liver damage, BMP-9 expression is transiently down-regulated and blocking BMP-9 under conditions of chronic damage ameliorated liver fibrogenesis in C57BL/6 mice. Thereby, BMP-9 acted as a pro-fibrogenic cytokine in the liver but without directly activating isolated HSC in vitro. Lipopolysaccharide (LPS), an endotoxin derived from the membrane of Gram-negative bacteria in the gut, is known to be essential in the pathogenesis of diverse kinds of liver diseases. The aim of the present project was therefore to investigate how high levels of BMP-9 in the context of LPS signalling might result in enhanced liver damage. For this purpose, we stimulated human liver sinusoidal endothelial cells (LSEC) with LPS and incubated primary human liver myofibroblasts (MF) with the conditioned medium of these cells. We found that LPS led to the secretion of factors from LSEC that upregulate BMP-9 expression in MF. At least one of these BMP-9 enhancing factors was defined to be IL-6. High BMP-9 in turn, especially in combination with LPS stimulation, induced the expression of certain capillarization markers in LSEC and enhanced the LPS-mediated induction of pro-inflammatory cytokines in primary human macrophages. In LSEC, pre-treatment with BMP-9 reduced the LPS-mediated activation of the NfkB pathway, whereas in macrophages, LPS partially inhibited the BMP-9/Smad-1 signaling cascade. In vivo, in mice, BMP-9 led to the enhanced presence of F4/80-positive cells in the liver and it modulated the LPS-mediated regulation of inflammatory mediators. In summary, our data point to BMP-9 being a complex and highly dynamic modulator of hepatic responses to LPS: Initial effects of LPS on LSEC led to the upregulation of BMP-9 in MF but sustained high levels of BMP-9 in turn promote pro-inflammatory reactions of macrophages. Thereby, the spatial and timely fine-tuned presence (or absence) of BMP-9 is needed for efficient wound-healing responses in the liver.

Description: In recent years, multirotor unmanned aerial vehicles (UAVs) have become more and more important in the field of plant protection in China. Multirotor unmanned plant protection UAVs have been widely used in vast plains, hills, mountains, and other regions, and become an integral part of China’s agricultural mechanization and modernization. The easy takeoff and landing performances of UAVs are urgently required for timely and effective spraying, especially in dispersed plots and hilly mountains. However, the unclearness of wind field distribution leads to more serious droplet drift problems. The drift and distribution of droplets, which depend on airflow distribution characteristics of UAVs and the droplet size of the nozzle, are directly related to the control effect of pesticide and crop growth in different growth periods. This paper proposes an approach to research the influence of the downwash and windward airflow on the motion distribution of droplet group for the SLK-5 six-rotor plant protection UAV. At first, based on the Navier-Stokes (N-S) equation and SST k–ε turbulence model, the three-dimensional wind field numerical model is established for a six-rotor plant protection UAV under 3 kg load condition. Droplet discrete phase is added to N-S equation, the momentum and energy equations are also corrected for continuous phase to establish a two-phase flow model, and a three-dimensional two-phase flow model is finally established for the six-rotor plant protection UAV. By comparing with the experiment, this paper verifies the feasibility and accuracy of a computational fluid dynamics (CFD) method in the calculation of wind field and spraying two-phase flow field. Analyses are carried out through the combination of computational fluid dynamics and radial basis neural network, and this paper, finally, discusses the influence of windward airflow and droplet size on the movement of droplet groups.

Description: Pollination success is essential for hybrid oilseed rape (OSR, Brassica napus) seed production, but traditional pollination methods are not efficient. The unmanned agricultural aerial system (UAAS) has developed rapidly and has been widely used in China. When flying, the wind field generated by the rotors overcomes the UAAS gravity, and it blows and disturbs the crops below, which helps the pollen spread. In order to investigate the distribution law of the three-dimensional (direction x, y, z) airflow field, experiments involving three levels of flight speed (FS) at 4.0, 5.0, and 6.0 m/s, and three levels of flight height (FH) at 1.5, 2.0, and 2.5 m were conducted in the OSR field by using an electric four-rotor UAAS P20. The effects of FS and FH on airflow velocities (vx, vy, vz) were analyzed. High-speed dynamic camera (HSDC) technology was used to capture the swings of OSR plants under airflow field disturbance. OSR pollen samples were collected during the experiments. The results showed that the airflow field in the direction x was mainly concentrated on the center of the flight path (S3), and the maximum wind velocity of direction x was 8.01 m/s (T1, S3). The direction x airflow field width was distributed almost symmetrically, but the center position shifted easily, due to crosswind. The airflow field in the direction y was distributed on both sides of the center flight path, and the velocity was generally larger, with the maximum at 7.91 m/s (T1, S2). The airflow field in the direction z was distributed irregularly, and the velocity was small. The FH had highly significant impacts on  vx (p 〈 0.01), and the interaction of FS and FH had significant impacts on  vx (0.01 〈 p 〈 0.05), while the FS had no significant impact on vx (p = 0.70804 〉 0.05). The FS, FH, and interaction of FS and FH all had highly significant impacts on vy (p 〈 0.01). The swings of the OSR plant captured by the HSDC proved that the UAAS airflow field could effectively blow the OSR plant. The swing amplitude changes showed a positive correlation with airflow velocities (vx) in general. Although the observed OSR plant swung forward and backward repeatedly, there was a law of first forward, and then backward, and forward again at the beginning of each swing. The pollen collected on the sampler verified that the UAAS airflow field could help with pollen spread. The research results provide technical support for UAAS application on supplementary pollination for hybrid OSR seed production.