Research on the Construction of a Knowledge Graph and Knowledge Reasoning Model in the Field of Urban Traffic

The integration of multi-source transportation data is complex and insufficient in most of the big cities, which made it difficult for researchers to conduct in-depth data mining to improve the policy or the management. In order to solve this problem, a top-down approach is used to construct a knowledge graph of urban traffic system in this paper. First, the model layer of the knowledge graph was used to realize the reuse and sharing of knowledge. Furthermore, the model layer then was stored in the graph database Neo4j. Second, the representation learning based knowledge reasoning model was adopted to implement knowledge completion and improve the knowledge graph. Finally, the proposed method was validated with an urban traffic data set and the results showed that the model could be used to mine the implicit relationship between traffic entities and discover traffic knowledge effectively.

Iceberg modelling with NEMO

<p>Icebergs represent around half of the yearly mass discharge from the Greenland Ice Sheet. They are not only important freshwater sources, but also pose a threat to navigation and other offshore activities. Since monitoring individual icebergs in large numbers is unfeasible, numerical models are great tools to evaluate their role in freshwater distribution and their general trajectory patterns. While large-scale iceberg modelling is in its infancy, we show recent model improvements done in the Nucleus for European Modelling of the Ocean (NEMO) iceberg module. Among those, we highlight a newly implemented iceberg-sea ice dynamic, where icebergs are locked in concentrated and strong sea ice packs, so they will move with sea ice instead of across it. Additionally, recent code modifications allow the user to choose if the iceberg melt plume is inserted in the ocean’s first model layer or distributed along the iceberg draft. Results will show if these code upgrades change the way freshwater is distributed in the ocean and if they better represent iceberg trajectories and their surge seasonality off the Labrador shelf.</p>

Versioning of 3D City Models for Municipality Applications: Needs, Obstacles and Recommendations

The use of 3D city models is changing from visualization to complex use cases where they act as 3D base maps. This requires links to registers and continuous updating of the city models. Still, most models never change or are recreated instead of updated. This study identifies obstacles to version management of 3D city models and proposes recommendations to overcome them, with a main focus on the municipality perspective, foremost in the planning and building processes. As part of this study, we investigate whether national building registers can control the version management of 3D city models. A case study based on investigations of standards, interviews and a review of tools is presented. The study uses an architectural model divided into four layers: data collection, building theme, city model and application. All layers require changes when implementing a new versioning method: the data collection layer requires restructuring of technical solutions and work processes, storage of the national building register requires restructuring, versioning capabilities must be propagated to the city model layer, and tools at the application layer must handle temporal information better. Strong incentives for including versioning in 3D city models are essential, as substantial investment is required to implement versioning in all the layers. Only capabilities required by applications should be implemented, as the complexity grows with the number of versioning functionalities. One outcome of the study is a recommendation to link 3D city models more closely to building registers. This enables more complex use in, e.g., building permits and 3D cadastres, and authorities can fetch required (versioning) information directly from the city model layer.

Hysteresis modeling of piezoelectric micro-positioning stage based on convolutional neural network

The inherent hysteresis nonlinearity of piezoelectric actuator degrades the positioning accuracy of the micro-positioning stage. Prandtl–Ishlinskii model is widely used for piezoelectric hysteresis modeling, yet it is a rate-independent model with weak generalization ability. To overcome this problem, we proposed a convolutional neural network model based on the Prandtl–Ishlinskii model, which consists of a rate-dependent Prandtl–Ishlinskii model layer and convolutional network layer. The rate-dependent Prandtl–Ishlinskii model layer extends the traditional Prandtl–Ishlinskii model to describe the rate-dependent hysteresis behavior. The convolutional network layer with deep learning ability extracts the deep features of the input signal to improve the generalization ability of the hysteresis model. The experiment results indicate that the standard error of the proposed hysteresis model to predict displacement at unmodeled frequencies has been reduced by 18.74%–36.75% in comparison with the Prandtl–Ishlinskii model, which verifies that the proposed hysteresis model has not only higher accuracy but also stronger generalization ability.

A Referenced Cyber Physical System for Compressor Manufacturing

Compressor is a typical high-end discrete product，with the shortening of product life cycle and the enhancement of the degree of product customization, the traditional compressor manufacturing system architecture cannot meet the requirements of comprehensive digital management of compressor from body scheme design to parts production line, logistics management, operation and maintenance monitoring and evaluation. This paper presents a compressor manufacturing system architecture based on digital twinning, and establishes an Internet platform for compressor industry oriented to remote coordination from three aspects of compressor design, production, operation and maintenance. The platform includes industrial Internet infrastructure layer, physical space entity model layer, virtual space multidimensional model layer, physical space and virtual space multidimensional model correlation and mapping layer, big data intelligent analysis decision-making layer, and digital twin application layer. Through the establishment of the compressor product design and simulation model of digital twin, compressor production process digital twin model, compressor fault diagnosis and remote operations digital twin model, implementation is based on the number of compressor collaboration in manufacturing industrial Internet platform twin system, leading the transformation and upgrading of intelligent manufacturing industry, compressor industry sustainable development ability and international competitiveness.

CNN explains tuning properties of anterior, but not middle, face-processing areas in macaque IT

Recent computational studies have emphasized layer-wise quantitative similarity between convolutional neural networks (CNNs) and the primate visual ventral stream. However, whether such similarity holds for the face-selective areas, a subsystem of the higher visual cortex, is not clear. Here, we extensively investigate whether CNNs exhibit tuning properties as previously observed in different macaque face areas. While simulating four past experiments on a variety of CNN models, we sought for the model layer that quantitatively matches the multiple tuning properties of each face area. Our results show that higher model layers explain reasonably well the properties of anterior areas, while no layer simultaneously explains the properties of middle areas, consistently across the model variation. Thus, some similarity may exist between CNNs and the primate face-processing system in the near-goal representation, but much less clearly in the intermediate stages, thus giving motivation for a more comprehensive model for understanding the entire system.

The speed of human social interaction perception

The ability to detect and understand other people’s social interactions is a fundamental part of the human visual experience that develops early in infancy and is shared with other primates. However, the neural computations underlying this ability remain largely unknown. Is the detection of social interactions a rapid perceptual process, or a slower post-perceptual inference? Here we used magnetoencephalography (MEG) decoding and computational modeling to ask whether social interactions can be detected via fast, feedforward processing. Subjects in the MEG viewed snapshots of visually matched real-world scenes containing a pair of people who were either engaged in a social interaction or acting independently. The presence versus absence of a social interaction could be read out from subjects’ MEG data spontaneously, even while subjects performed an orthogonal task. This readout generalized across different scenes, revealing abstract representations of social interactions in the human brain. These representations, however, did not come online until quite late, at 300 ms after image onset, well after the time period of feedforward visual processes. In a second experiment, we found that social interaction readout occurred at this same latency even when subjects performed an explicit task detecting social interactions. Consistent with these latency results, a standard feedforward deep neural network did not contain an abstract representation of social interactions at any model layer. We further showed that MEG responses distinguished between different types of social interactions (mutual gaze vs joint attention) even later, around 500 ms after image onset. Taken together, these results suggest that the human brain spontaneously extracts the presence and type of others’ social interactions, but does so slowly, likely relying on iterative top-down computations.

Experimental Study to Reinforce The Weak Subgrade Soil For Low-Volume Roads by Coir Geotextile Mats

The construction and maintenance of pavement over the weak subgrade soil become the challenging task to the pavement engineering. One of the major reasons of subgrade failure of pavement is weak subgrade. The weak subgrade soil noticed a Black Cotton (BC) soil. The BC soil subgrade poses several serious problems to the pavement such as rutting, fatigue, reflecting crack and undulation of the pavement. To minimize this problem of pavement, there are many conventional stabilization techniques were adopted and reported. But these techniques are not applied effectively into the pavement to stabilize the weak subgrade. To address this problem, to give the additional strength to the pavement geosynthetics are taken as alternate material for stabilization of pavement. In the present study, an attempt is made in the laboratory with four types of coir mats by using the fabricated mould. The study is conducted in the form of two-layer pavement system. The pavement model layer is prepared as subgrade and sub-base with BC soil and sandy gravel soil respectively. The prepared fabricated mould is tested by using the Wheel Tracking Test (WTT) under moving traffic loading condition. This study concluded that the suitable placement position and the types of coir mats can be affect the performance of the LVRs. It’s also noticed that the suitable coir mats can effectively reduce the deformation, so that it can be used over the weak subgrade to improve the performance of the LVRs.

DISASTER RECOVERY THROUGH PREDICTION OF SAFE ROUTE USING DEM LEVELS

When a disaster occurs, the normal commutation routes are disrupted. People get stuck at these disaster points and would be in trouble, hence people in those areas find it difficult to communicate and evacuation route to safe area is unknown. The aim of the paper is to predict safe routes to reach the refuge point from the disaster point. The prototype was developed using Arc geographic information system runtime SDK for Java Application and APIs in Eclipse. The system was developed with digital elevation model layer, and route layer for India basemap focused to Tamil Nadu. The safe route is found based on the elevation values of the area from the disaster point to a safe point. The developed system could be used by the relief providers to reach the disaster point and rescue victims.