Search

Local planning is a critical and difficult task for intelligent vehicles in dynamic transportation environments. In this paper, a new method Suppress Q Deep Q Network (SQDQN) combining traditional deep reinforcement learning Deep Q Network (DQN) with information entropy is proposed for local planning in automatic driving. In the proposed approach, local planning strategy in complex traffic environment established by the actor–critic network based on DQN, the method adopts the way of execution action-evaluation action-update network to explore the optimal local planning strategy. Proposed strategy does not rely on accurate modeling of the scene, so it is suitable for complex and changeable traffic scenes. At the same time, evaluate the update process and determine the update range by using information entropy to solve a common problem in the network that over expectation of actions damage the performance of strategies. Use this approach to improve strategic performance. The trained local planning strategy is evaluated in three simulation scenarios: overtaking, following, driving in hazardous situations. The results illustrate the advantages of the proposed SQDQN method in solving local planning problem. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Shanghai Jiaotong University (Science) is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Shanghai Jiaotong University (Science) is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

With the large-scale integration of wind power into the power grid, improving the wind speed prediction accuracy is of great significance for promoting the consumption of renewable energy. In this paper, a hybrid prediction method for multi-step wind speed prediction based on the empirical wavelet transform (EWT), multi-objective modified seagull optimization algorithm (MOMSOA), and multi-kernel extreme learning machine (MKELM) is proposed. First, EWT is used to decompose the nonstationary wind speed data into a set of stationary subsequences. Then, each subsequence of wind speed is predicted by the MKELM, and the MKELM network is optimized by the MOMSOA newly proposed in this study. Finally, the inverse empirical wavelet transform (IEWT) is adopted to reconstruct the prediction results into the final wind speed prediction results. To assess the performance of the proposed combined model, four groups of experiments are carried out on four wind speed sequences, and a comparative analysis is made with 16 comparison models. The models involved in the investigation are discussed comprehensively in terms of significance and stability. The results demonstrate that the developed combined model outperforms the comparison models. [ABSTRACT FROM AUTHOR]

This study uses deep reinforcement learning (DRL) combined with computer vision technology to investigate vehicle fuel economy. In a driving cycle with car-following and traffic light scenarios, the vehicle fuel-saving control strategy based on DRL can realize the cooperative control of the engine and continuously variable transmission. The visual processing method of the convolutional neural network is used to extract available visual information from an on-board camera, and other types of information are obtained through the vehicle's inherent sensor. The various detected types of information are further used as the state of DRL, and the fuel-saving control strategy is built. A Carla–Simulink co-simulation model is established to evaluate the proposed strategy. An urban road driving cycle and highway road driving cycle model with visual information is built in Carla, and the vehicle power system is constructed in Simulink. Results show that the fuel-saving control strategy based on DRL and computer vision achieves improved fuel economy. In addition, in the Carla–Simulink co-simulation model, the fuel-saving control strategy based on DRL and computer vision consumes an average time of 17.55 ms to output control actions, indicating its potential for use in real-time applications. [ABSTRACT FROM AUTHOR]

Grasping the future change trend of soil fertility has great significance in improving the soil quality and achieving high-quality crop production and sustainable agricultural development. However, studies predicting the future change trend of farmland soil fertility are scarce. In this paper, with Yanzhou District of Shandong Province as the research area, a study was conducted based on the sampled data from 2012 to 2017. The data extracted from 2012 to 2016 was used for prediction and that from 2017 was applied for verification. The pH, organic matter, available phosphorus, alkali-hydrolyzed nitrogen and available potassium were selected as indexes of soil fertility. From a socioeconomic perspective, the factors affecting the changes in soil fertility selected in this study include fertilization measures, crop yield, area of arable land, farmers' income, degree of mechanized operation, irrigated area, pesticide dosage, mulch dosage and rural electricity consumption. Based on this, a stochastic Petri net was used to build a model for predicting the soil fertility change trend. According to the relevant statistical data, the parameters of the model were determined, and by using the solid mathematical basis of the model, the probability of about 0.7852 was calculated out for the soil fertility to decline in the study area in the coming year. By comparing the soil fertility in 2016 and 2017, and further analyzing the changes in soil fertility from 2012 to 2016, the method of predicting the variation trend of soil fertility proposed in this study was verified to be effective. [ABSTRACT FROM AUTHOR]

Berry phase effects have significant influences on the electronic properties of condensed matter. In particular, the anomalous Hall conductivity has been recognized as an intrinsic property of the systems with non-zero Berry curvature. Here, we present the anomalous Hall effect observed in the non-magnetic material ZrTe5, which hosts a large Zeeman splitting with Landé g-factor of 26.49. The quantum oscillation analysis reveals non-linear band dispersion near the top of valence band in bulk band structure, and no Weyl node forms with applied magnetic field. The anomalous Hall conductivity reaches 129 Ω−1 cm−1 at 2 K, and shows weak temperature dependence. All these combined with theoretical analysis suggest that the anomalous Hall effect observed in ZrTe5 originates from the non-vanishing Berry curvature induced by combining large Zeeman splitting and strong spin–orbit coupling. Remarkably, the anomalous Hall resistivity reverses its sign from negative to positive at a hydrostatic pressure P = 1.3 GPa, which confirms that the anomalous Hall effect in ZrTe5 is highly related to the band structure-dependent Berry curvature. Our results have verified the anomalous Hall mechanism in ZrTe5 and offer a new platform to study the anomalous Hall effect. [ABSTRACT FROM AUTHOR]

Application server cluster (cluster for short) is generally designed to handle peak load, resulting in an excessive waste of energy; hence, its deployment should be dynamically adjusted according to changing load. In this paper, we propose a power-aware dynamic deployment scheme for application server cluster, which aims to minimize cluster’s power consumption under given target of CPU utilization guarantee. Firstly, we present an approach to guarantee server’s CPU utilization equal to a given target value through load allocation and Dynamic Frequency Scaling (DFS), deduce the power models of Discrete DFS (D-DFS) and Equivalent Continuous DFS (EC-DFS) manner, and analyze the power and continuity advantages of EC-DFS. Then, combining DFS, server Dynamic Switching On/Off and probability-based request scheduling, we transform cluster’s power-aware dynamic deployment problem to a programming problem, whose variable number is less than other similar researches. Finally, an efficient Differential Evolution-based solving algorithm is proposed to solve the problem. Test results verify the advantages of EC-DFS, and demonstrate the effectiveness of our deployment scheme and solving algorithm. [ABSTRACT FROM AUTHOR]

By coupling the phase field model of dendrite growth with the Lattice Boltzmann method (LBM), a Phase field LBM model (PF-LBM) that can simulate 3D dendrite non-isothermal growing from the under-cooled melt of a pure substance in forced flow has been established. Taking high-purity succinonitrile (SCN) as an example, the effects of flow on 3D dendritic growth were studied. The calculation efficiency of LBM and SOLA algorithm was compared under the same condition. The results show that, with the PF-LBM model, the effects of forced flow on 3D dendrite growth morphology and the dendrite asymmetric growth can be quantitatively modeled in large scale, and the dendrite growth is consistent with the crystallization theory. The LBM algorithm overcomes the weakness of a large amount of calculation and slow computation speed that exists in the traditional algorithm, and it achieves a speedup of 11.7 when the calculation region size is 200. [ABSTRACT FROM AUTHOR]

The Jinding Pb-Zn deposit in Yunnan Province is the representative of a Cd-enriched area and mining activities lead to the release of Cd into the hypergenic ecosystem, resulting in Cd pollution. The concentrations of Cd vary greatly from one type to another type of rocks in the mining district. In the host rock, Cd concentrations range from 50×10−6 to 650×10−6 with an average of 310×10−6. In primary ores, Cd concentrations range from 14×10−6 to 2800×10−6 with an average of 767×10−6. However, in oxidized ores, Cd concentrations are highest, varying within the range of 110×10−6 to 8200×10−6, averaging 1661×10−6. It is shown that the oxidized ores are the main carrier and environmental source of Cd. Leaching test showed that Pb/Zn ores are easy to oxidize and thereafter release Cd and other harmful elements. These leached elements in the leachate may be precipitated rapidly in the order of Zn>Pb>Cd. As for the concentration distribution of Cd in the Bijiang River, it is estimated to be 15.7 μg/L Cd in water, 49.3 mg/L in suspended substances, and 203.7 mg/L in sediments. The average value of Cd in soil from the polluted area is 83.0 mg/kg. Natural weathering of Cd-rich rocks and minerals imposes a potential environmental risk on the aquatic ecosystem of the Bijiang catchment. [ABSTRACT FROM AUTHOR]

Hypertrophic cardiomyopathy (HCM) represents one of the most common heritable heart diseases. However, the signalling pathways and regulatory networks underlying the pathogenesis of HCM remain largely unknown. Here, we present a strand-specific RNA-seq dataset for both coding and lncRNA profiling in myocardial tissues from 28 HCM patients and 9 healthy donors. This dataset constitutes a valuable resource for the community to examine the dysregulated coding and lncRNA genes in HCM versus normal conditions. Design Type(s) transcription profiling design • disease state design • sequence analysis objective Measurement Type(s) transcription profiling assay Technology Type(s) RNA sequencing Factor Type(s) experimental condition Sample Characteristic(s) Homo sapiens • heart Machine-accessible metadata file describing the reported data (ISA-Tab format) [ABSTRACT FROM AUTHOR]