Aims: To enhance the safety of intelligent vehicle drivers and help them achieve active collision avoidance. Background: Existing research on vehicles’ avoidance focuses reference path optimization stability tracking control. However, there is relatively little in emergency accident scenarios. Longitudinal braking technology has been mature, while steering still needs a long time to develop. The joint situations limited. Objective: Modeling longitudinal lateral distances based conditions...

Passenger flow prediction is the key to operation efficiency and safety of urban rail transit (URT). This paper combines deep learning (DL) theory support vector machine (SVM) into DL-SVM model for URT passenger prediction. Firstly, belief network (DBN) was adopted extract features inherent variation data. On this basis, an SVM regression constructed predict flow. Then, proposed compared with three shallow models through experiments on Qingdao Metro. The results show that outperformed other...

An intrabody nanonetwork (IBNN) is composed of nanoscale (NS) devices, implanted inside the human body for collecting diverse physiological information diagnostic and treatment purposes. The unique constraints these NS devices in terms energy, storage computational resources are primary challenges effective designing routing protocols IBNNs. Our proposed work explicitly considers limitations introduces a novel energy-efficient scheme based on fuzzy logic bio-inspired firefly algorithm....

An Intrabody Nanonetwork (IBNN) is constituted by nanoscale devices that are implanted inside the human body for monitoring of physiological parameters disease diagnosis and treatment purposes. The extraordinary accuracy precision these in cellular level drug delivery envisioned to advance traditional healthcare system. However, feature constraints devices, such as inadequate energy resources, topology-unawareness, limited computational power, challenges development energy-efficient routing...

The spatially distributed nanosensors move through the body for detecting different physiological information which is collected by nanorouter further disease diagnosis and treatment. presented study addresses primary challenge of data collection with minimum energy consumption proposing a novel approach that exploits moving nature Intrabody nanonetwork. adaptive self-organized mobility achieved employing Particle Swarm Optimization (PSO) model. PSO based model covers how efficiently from...

Potassium ferrate has strong oxidation in both acid and alkali environments, which attracted extensive attention. However, the impact of pH environment on this coupling process with goal resource recovery not received Under efficient organic acid, changes solid-liquid characteristics sludge after alkaline pretreatment during anaerobic digestion were discussed. The results showed that compared blank control groups, pretreatment, volatile suspended solids (VSSs) decreased most, reaching...

As a typical time series, the length of data sequence is critical to accuracy traffic state prediction. In order fully explore causality between data, this study established temporal backtracking and multistep delay model based on recurrent neural networks (RNNs) learn extract long- short-term dependencies data. With real set, coordinate descent algorithm was employed search determine optimal sequence, predictions were performed demonstrate relationship steps prediction accuracies. Besides,...

This paper studies the prediction problem based on historical time series data of base station. Established LSTM and ARIMA models to train predict traffic respectively. choose put model species into each series, forecast results table according best. At same time, we found that cell showed three patterns rising, steady, falling over instead rising all time. Our uses ‘elu’ function as activation function, number training rounds is 50, neurons 1000, observation output sequence 120, which...

In Dempster-Shafer evidence theory (DST), the determination of basic belief assignment (BBA) is an important yet challenging issue. The rational mass compound focal elements crucial for fully taking advantage DST, i.e., ability to represent ambiguity. this paper, elements, we select and construct \enquote{compound-class samples} with ambiguous class membership. Then, use these samples end-to-end model called Evidential Radial Basis Function Network (E-RBFN), input as sample output...

In order to study the role of EDR data in vehicle accident reconstruction, 15 basic types are analyzed. This paper reconstructs complete process 16 car traffic accidents and analyzes sources collision EDR. And trigger threshold range each type is explained, cases reconstructed The typical analyzed by data. speed direction force at time combined with sensitivity parameters. second reconstruction comparison error two results show that relative drops within 10%, is, adjusting parameter can...

With economic development, the intelligent logistics system is well known. This paper proposes Intelligent distribution along with a brief introduction of system. After detailed applications transportation systems (ITS) and electronic commerce (EC) in distribution, model algorithms, put forward combination ITS EC which could become development trend for smart distribution.

Whether the railway track can run smoothly for a long time directly affects safety of railway. In view nonlinear, random, and sudden characteristics series data track, we propose WT-GA-GRU model to forecast irregularity high-speed Firstly, original is decomposed by wavelet transform (WT), multiple are forecasted gated recurrent unit (GRU) networks optimized genetic algorithm (GA). Then results obtained reconstruction. Experimental show that, compared with support vector machine (SVM)...

The determination of the basic belief assignment (BBA) is an important yet difficult problem in evidence theory. In this paper, some BBA methods using density estimation that can directly generate compound focal elements for ambiguous classes are proposed, including Gaussian Mixture Model (GMM) based and Generative Adversarial Network (GAN) methods. Experimental results combination pattern classification on various UCI data sets show our new proposed rational effectively improve accuracy...

This paper presents studies on the theories aimed at using technologies of Internet Things in domain emergency management ponded water road surface. It first introduces an model based to solve problem discusses a few critical related this including monitoring condition surface and traffic simultaneously, framework data interfacing multiple sources heterogeneous data, analysis jam spreading pattern locations water, comprehensive information analysis, demonstrations GIS. In end it gives our...

Aiming at the problem of attribute reduction in rough set theory, polarization matrix and relative degree are defined, then, an approach to minimum is presented. The typical example decision table shows that it feasible effective. In oil-gas reservoir test, applied optimize seismic attributes with notable application effect, which reaches aim prediction accuracy, saves cost, improves processing speed so on. Moreover, combines optimization organically, establishes foundation optimization.

In urban transportation network, generally, traffic sensors are unable to fully cover the freeway and only local measurement data is available. To accurately timely estimate density in an this paper proposes a distributed state observer approach based on macroscopic flow model. Firstly, network modeled way of systems by using cell transmission model (CTM). Secondly, model, designed, which gains computed LMI technique according Lyapunov function S-procedure theory. Thus, densities whole can...

Feature selection is one of the key processes in pattern classification. For task feature selection, an important problem evaluation classification capability or clarity different spaces, which related to their uncertainty. In framework Dempster-Shafer evidence theory (DST), Basic Belief Assignment (BBA) can well describe this paper, Generative Adversarial Network (GAN) used generate BBAs each space, and a novel approach based on ambiguity distribution proposed, uses skewness variance BBAs'...