Past Trajectories Research Articles

Partial smoothing of delay transition semigroups acting on special functions

It is well known that the transition semigroup of an Ornstein Uhlenbeck process with delay is not strong Feller for small times, so it has no regularizing effects when acting on bounded and continuous functions. In this paper we study regularizing properties of this transition semigroup when acting on special functions of the past trajectory. With this regularizing property, we are able to prove existence and uniqueness of a mild solution for a special class of semilinear Kolmogorov equations; we apply these results to a stochastic optimal control problem.

Stratifying and predicting patterns of neighbourhood change and gentrification: An urban analytics approach

AbstractWhile recent debates have widely acknowledged gentrification's varied manifestations, success in enumerating and disentangling the process and its defining features from other forms of neighbourhood change at‐scale and across entire cities, has remained largely elusive. This paper addresses this gap and employs a novel, open and reproducible urban analytics approach to systematically examine the past and future trajectories of neighbourhood change using London, England, as a case‐study example. Using suites of datasets relating to population, house prices, and built environment development, the nature of gentrification's mutations and its spatial patterns are extracted through a multi‐stage data dimensionality reduction and classification methodology. Machine learning is subsequently adopted to model gentrification's observed trends and predict its future frontiers with interactive visualisation methods offering new insights into gentrification's projected dynamics and geographies.

Gap dynamics and disease-causing invasive species drive the development of an old-growth forest over 250 years

Investigating the disturbance regimes of unharvested forests helps us understand their past, present, and future trajectory and gives us a model for forest management. It can also clarify the relative importance of small-scale gap dynamics versus more severe disturbances. Here we used tree rings to examine the recruitment patterns, growth dynamics, and disturbance chronologies of American beech (Fagus grandifolia), sugar maple (Acer saccharum), and eastern hemlock (Tsuga canadensis) in A.B. Williams Woods, an old-growth forest in Ohio, USA, over the past 250 years. We found that beech and sugar maple recruitment peaked around 1900 and continued through the 1900s, while hemlock recruitment peaked during 1825–1875, then declined and effectively ended in the early 1900s. Hemlock grew fastest during the 1800s according to ring width and basal area increment, while sugar maple ring width surpassed beech and hemlock in the 1900s. All three species showed a dramatic increase in growth from 1980 to 2010. Beech and sugar maple established regardless of canopy gaps, but 73% of hemlocks originated in gaps. In most decades, <10% of trees experienced gap recruitment or growth release, suggesting that ongoing, endogenous canopy mortality was the primary disturbance shaping this forest. However, a more severe forest-wide disturbance occurred during the 1980s–1990s when the scale insect causing beech bark disease was introduced, with greater than 30% of living trees showing growth releases in those decades. Another synchronous release occurred in the 1930s when blight-killed chestnuts were removed; 16% of trees showed releases. Both of these intermediate-severity disturbances involved human introduction of invasive species. Thus, we documented a natural disturbance regime of small-scale gap dynamics, punctuated by more severe anthropogenic disturbances in the twentieth century. These relatively frequent, intermediate-severity events probably mean that the forest’s current composition is non-equilibrial. Hemlock may continue to decline while beech maintains its dominance by constant regeneration in both gaps and shade, and by responding to disturbance with root suckering and growth pulses. The codominance of sugar maple may be relatively recent, and perhaps temporary, as we found little sugar maple recruitment before 1875 or after 1950, and three times fewer sugar maple saplings now than in the early 1900s. Despite being protected as a park, the development of this old-growth forest has been shaped more by disease-causing invasive species than natural disturbances over the past century. This result emphasizes the pervasiveness of human impacts even in communities we look to as examples of natural pattern and process.

Object ID Tracking in Videos: A Review

Assigning and monitoring distinct identifiers to objects or entities in a video stream is known as object ID tracking in videos. For the purpose of tracking and analyzing object movement over time, computer vision, video analytics, and surveillance systems employ this technology extensively. Object detection is the first step in the process, wherein computer vision algorithms locate and identify things within individual video frames. This may entail methods like region-based Convolutional Neural Network (Faster R-CNN) or YOLO (You Only Look Once), which are object detection models based on deep learning. Based on the object's past trajectory, the tracking algorithm assists in predicting the object's position in the frame.

Casseqgcn: Combining Network Structure and Temporal Sequence to Predict Information Cascades

One important task in the study of information cascade is to predict the future recipients of a message given its past spreading trajectory. While the network structure serves as the backbone of the spreading, an accurate prediction can hardly be made without the knowledge of the dynamics on the network. The temporal information in the spreading sequence captures many hidden features, but predictions based on sequence alone have their limitations. Recent efforts start to explore the possibility of combining both the network structure and the temporal feature for a more accurate prediction. Nevertheless, it is still a challenge to efficiently and optimally associate these two interdependent factors. Here, we propose a new end-to-end prediction method CasSeqGCN in which the structure and temporal feature are simultaneously taken into account. A cascade is divided into multiple snapshots which record the network topology and the state of nodes. The graph convolutional network (GCN) is used to learn the representation of a snapshot. The dynamic routing and the long short-term memory (LSTM) model are used to aggregate node representation and extract temporal information. CasSeqGCN predicts the future cascade size more accurately compared with other state-of-art baseline methods. The ablation study demonstrates that the improvement mainly comes from the design of the input and the GCN layer. Taken together, our method confirms the benefit of combining the structural and temporal features in cascade prediction, which not only brings new insights but can also serve as a useful baseline method for future studies.

Motion Prediction of Beating Heart Using Spatio-Temporal LSTM

In robot-assisted cardiac surgery, predicting heart motion can help improve the operation accuracy and safety of surgical robots. Different from the conventional prediction schemes which model the point of interest (POI) with only temporal correlation of past observations, this paper proposes an LSTM-based method by exploiting the spatio-temporal correlation of the 3D movements of POI and auxiliary points (APs) on the same surface of the heart. Three different LSTM models are investigated. The first two models define the POI prediction as a pure time-series forecasting problem based on past POI trajectory, and the third model combines the past observations of POI and new observations of APs to take into consideration the extra spatial correlations for prediction. Experimental comparison studies based on 3D coordinates obtained from real stereo-endoscopic videos demonstrate the superior performance of the proposed spatio-temporal LSTM model.

Methods and Means for Small Dynamic Objects Recognition and Tracking

A literature analysis has shown that object search, recognition, and tracking systems are becoming increasingly popular. However, such systems do not achieve high practical results in analyzing small moving living objects ranging from 8 to 14 mm. This article examines methods and tools for recognizing and tracking the class of small moving objects, such as ants. To fulfill those aims, a customized You Only Look Once Ants Recognition (YOLO_AR) Convolutional Neural Network (CNN) has been trained to recognize Messor Structor ants in the laboratory using the LabelImg object marker tool. The proposed model is an extension of the You Only Look Once v4 (Yolov4) 512 × 512 model with an additional Self Regularized Non–Monotonic (Mish) activation function. Additionally, the scalable solution for continuous object recognizing and tracking was implemented. This solution is based on the OpenDatacam system, with extended Object Tracking modules that allow for tracking and counting objects that have crossed the custom boundary line. During the study, the methods of the alignment algorithm for finding the trajectory of moving objects were modified. I discovered that the Hungarian algorithm showed better results in tracking small objects than the K–D dimensional tree (k-d tree) matching algorithm used in OpenDataCam. Remarkably, such an algorithm showed better results with the implemented YOLO_AR model due to the lack of False Positives (FP). Therefore, I provided a new tracker module with a Hungarian matching algorithm verified on the Multiple Object Tracking (MOT) benchmark. Furthermore, additional customization parameters for object recognition and tracking results parsing and filtering were added, like boundary angle threshold (BAT) and past frames trajectory prediction (PFTP). Experimental tests confirmed the results of the study on a mobile device. During the experiment, parameters such as the quality of recognition and tracking of moving objects, the PFTP and BAT, and the configuration parameters of the neural network and boundary line model were analyzed. The results showed an increased tracking accuracy with the proposed methods by 50%. The study results confirmed the relevance of the topic and the effectiveness of the implemented methods and tools.

Measuring the Milky Way mass distribution in the presence of the LMC

ABSTRACTThe ongoing interaction between the Milky Way (MW) and its largest satellite – the Large Magellanic Cloud (LMC) – creates a significant perturbation in the distribution and kinematics of distant halo stars, globular clusters and satellite galaxies, and leads to biases in MW mass estimates from these tracer populations. We present a method for compensating these perturbations for any choice of MW potential by computing the past trajectory of LMC and MW and then integrating the orbits of tracer objects back in time until the influence of the LMC is negligible, at which point the equilibrium approximation can be used with any standard dynamical modelling approach. We add this orbit-rewinding step to the mass estimation approach based on simultaneous fitting of the potential and the distribution function of tracers, and apply it to two data sets with the latest Gaia EDR3 measurements of 6D phase-space coordinates: globular clusters and satellite galaxies. We find that models with LMC mass in the range $(1\!-\!2) \times 10^{11}\, \mathrm{M}_\odot$ better fit the observed distribution of tracers, and measure MW mass within 100 kpc to be $(0.75\pm 0.1)\times 10^{12}\, \mathrm{M}_\odot$, while neglecting the LMC perturbation increases it by ∼15 per cent.

When history does not matter? The rise of Quebec’s wine industry

This article contributes to the debate on new regional path development, proposing an analytical framework that accounts for new industries arising almost ex nihilo in places with weakly developed preconditions. The paper explores how seemingly adverse initial conditions can be translated into a new development path over time and casts light on the interplay between structure and agency in such settings. We find that new path development processes are not necessarily conditioned by past trajectories or by prior regional and technological capabilities, but can be initiated by forward-looking, entrepreneurial pioneers and consolidated by actors who develop the wider institutional and organizational structures to facilitate further growth of the new industry. We study the case of the wine industry in Southern Quebec, which emerged despite weakly developed preconditions and developed into a fully established, legitimized and supported path over the past forty years.

Editorial

Editorial

Reframing Chinese Business History

Business history is expanding to include a greater plurality of contexts, with the study of Chinese business representing a key area of growth. However, despite efforts to bring China into the fold, much of Chinese business history remains stubbornly distal to the discipline. One reason is that business historians have not yet reconciled with the field's unique origins and intellectual tradition. This article develops a revisionist historiography of Chinese business history that retraces the field's development from its Cold War roots to the present day, showing how it has been shaped by the particular questions and concerns of “area studies.” It then goes on to explore five recent areas of novel inquiry, namely: the study of indigenous business institutions, business and semi-colonial context, business at the periphery of empire, business during socialist transition, and business under Chinese socialism. Through this mapping of past and present trajectories, the article aims to provide greater coherence to the burgeoning field and shows how, by taking Chinese business history seriously, we are afforded a unique opportunity to reimagine the future of business history as a whole.

A context-aware pedestrian trajectory prediction framework for automated vehicles

With the unprecedented shift towards automated urban environments in recent years, a new paradigm is required to study pedestrian behaviour. Studying pedestrian behaviour in futuristic scenarios requires modern data sources that consider both the Automated Vehicle (AV) and pedestrian perspectives. Current open datasets on AVs predominantly fail to account for the latter, as they do not include an adequate number of events and associated details that involve pedestrian and vehicle interactions. To address this issue, we propose using Virtual Reality (VR) data as a complementary resource to current datasets, which can be designed to measure pedestrian behaviour under specific conditions. In this research, we focus on the context-aware pedestrian trajectory prediction framework for automated vehicles at mid-block unsignalized crossings. For this purpose, we develop a novel multi-input network of Long Short-Term Memory (LSTM) and fully connected dense layers. In addition to past trajectories, the proposed framework incorporates pedestrian head orientations and distance to the upcoming vehicles as sequential input data. By merging the sequential data with contextual information of the environment, we train a model to predict the future pedestrian trajectory. Our results show that the prediction error is reduced by considering contextual information extracted from the crossing environment, as well as the addition of time-series behavioural information to the model. To analyse the application of the methods to real AV data, the proposed framework is trained and applied to pedestrian trajectories extracted from an open-access video dataset. Finally, by implementing a game theory-based model interpretability method, we provide detailed insights and propose recommendations to improve the current automated vehicle sensing systems from a pedestrian-oriented point of view.

Simultaneous Past and Current Social Interaction-aware Trajectory Prediction for Multiple Intelligent Agents in Dynamic Scenes

Trajectory prediction of multiple agents in a crowded scene is an essential component in many applications, including intelligent monitoring, autonomous robotics, and self-driving cars. Accurate agent trajectory prediction remains a significant challenge because of the complex dynamic interactions among the agents and between them and the surrounding scene. To address the challenge, we propose a decoupled attention-based spatial-temporal modeling strategy in the proposed trajectory prediction method. The past and current interactions among agents are dynamically and adaptively summarized by two separate attention-based networks and have proven powerful in improving the prediction accuracy. Moreover, it is optional in the proposed method to make use of the road map and the plan of the ego-agent for scene-compliant and accurate predictions. The road map feature is efficiently extracted by a convolutional neural network, and the features of the ego-agent’s plan is extracted by a gated recurrent network with an attention module based on the temporal characteristic. Experiments on benchmark trajectory prediction datasets demonstrate that the proposed method is effective when the ego-agent plan and the the surrounding scene information are provided and achieves state-of-the-art performance with only the observed trajectories.

Wind erosion response to past and future agro-pastoral trajectories in the Sahel (Niger)

ContextWind erosion plays a major role in land degradation in semi-arid areas, especially in the Sahel. There, wind erosion is as sensitive to land use and land management as to climate factors. Future land use intensification may increase wind erosion and induce regional land degradation.ObjectiveWe aimed to estimate wind erosion responses to changing land management in a Sahelian region.MethodsWe defined land use intensification scenarios for a study site in southwestern Niger for two historical situations (1950s and 1990s), and two alternative prospective scenarios (2030s: extensive or intensive). We simulated vegetation growth and horizontal sediment flux of wind erosion for the corresponding landscapes.ResultsAnnual amounts of horizontal sediment flux increased with land management changes from 1950s (nil flux) to 1990s (176 kg m−1 yr−1) and 2030s (452 to 520 kg m−1 yr−1), mostly because of differences in land use, declining soil fertility, and practices decreasing the dry vegetation. For 2030s, intensive scenario exhibited larger vegetation yields than extensive conditions, but similar large values of horizontal sediment flux, thus questioning the sustainability of both scenarios. Realistic sets of practices had as large an influence as the largest theoretical range of practices on the variability of annual horizontal sediment flux. This variability was as large as that due to meteorological conditions.ConclusionsThis study demonstrates that the environmental impact of land use and management practices, of which wind erosion is an aspect, must be assessed at the landscape scale to account for the variability in land cover and associated land management.

Hepatic steatosis and development of type 2 diabetes: Impact of chronic hepatitis B and viral specific factors

Chronic hepatitis B (CHB) was associated with a lower prevalence of nonalcoholic fatty liver disease (NAFLD). The impact of CHB on the link between NAFLD and type 2 diabetes (T2D) and related virological implications remain unclear. We recruited 2255 middle-to older-aged individuals who were examined serially for hepatic steatosis by ultrasonography and blood biochemistry as part of a population-based hepatocellular-carcinoma cohort study. In CHB patients, hepatitis B surface antigen (HBsAg) seroclearance and variation in viral load trajectory were also evaluated. During the average follow-up of 6 years, 168 participants developed T2D. CHB, as compared with uninfected subjects, was associated with lower risks for both new development and persistence of hepatic steatosis. Furthermore, the risk of steatosis decreased with higher levels of past viral load trajectories (p for trend=0.0002). However, concomitant steatosis at baseline in CHB patients was still significantly associated with a 1.98-fold increased risk for T2D after multivariate adjustment including age, impaired fasting glucose, cirrhosis, and time-varying body mass index, although CHB reduced the propensity of hepatic steatosis to develop diabetes, especially for patients with high levels of past viral-load trajectory. In CHB, the functional cure of HBV infection, as indicated by HBsAg seroclearance, was associated with a 1.41-fold (95% CI 1.12-1.79) increased risk of steatosis. In addition, the increased risk for progressive impairment of glucose metabolism due to steatosis was especially prominent after HBsAg seroclearance. The data showed that HBV interferes with fatty liver disease and modulates its related T2D risk, offering additional insight into the interplay between NAFLD and CHB.

Demographic inferences and climatic niche modelling shed light on the evolutionary history of the emblematic cold-adapted Apollo butterfly at regional scale.

Cold-adapted species escape climate warming by latitudinal and/or altitudinal range shifts, and currently occur in Southern Europe in isolated mountain ranges within "sky islands". Here, we studied the genetic structure of the Apollo butterfly in five such sky islands (above 1,000m) in France, and infer its demographic history since the last interglacial, using single nucleotide polymorphisms (ddRADseq SNPs). The Auvergne and Alps populations show strong genetic differentiation but not alpine massifs, although separated by deep valleys. Combining three complementary demographic inference methods and species distribution models (SDMs) we show that the LIG period was highly unfavourable for Apollo that probably survived in small population in the highest summits of Auvergne. The population shifted downslope and expanded eastward between LIG and LGM throughout the large climatically suitable Rhône valley between the glaciated summits of Auvergne and Alps. The Auvergne and Alps populations started diverging before the LGM but remained largely connected till the mid-Holocene. Population decline in Auvergne was more gradual but started before (~7kya vs. 800 ya), and was much stronger with current population size ten times lower than in the Alps. In the Alps, the low genetic structure and limited evidence for isolation by distance suggest a nonequilibrium metapopulation functioning. The core Apollo population experienced cycles of contraction-expansion with climate fluctuations with largely interconnected populations overtime according to a "metapopulation-pulsar" functioning. This study demonstrates the power of combining demographic inferences and SDMs to determine past and future evolutionary trajectories of an endangered species at a regional scale.

Towards Predictive Analysis on Disease Progression: A Variational Hawkes Process Model.

Massively available longitudinal data about long-term disease trajectories of patients provides a golden mine for the understanding of disease progression and efficient health service delivery. It calls for quantitative modeling of disease progression, which is a tricky problem due to the complexity of the disease progression process as well as the irregularity of time documented in trajectories. In this study, we tackle the problem with the goal of predictively analyzing disease progression. Specifically, we propose a novel Variational Hawkes Process (VHP) model to generalize disease progression and predict future patient states based on the clinical observational data of past disease trajectories. First, Hawkes Process captures the intensity of irregular visits in a trajectory documented to medical facilities and controls the aforementioned information flowing into future visits. Thereafter, the captured intensity is incorporated into a Variational Auto-Encoder to generate the representation of the future partial disease trajectory for a target patient in a predictive manner. To further improve the prediction performance, we equip the proposed model with a disease trajectory discriminator to distinguish the generated trajectories from real ones. We evaluate the proposed model on two public datasets from the MIMIC-III database pertaining to heart failure and sepsis patients, respectively, and one real-world dataset from a Chinese hospital pertaining to heart failure patients with multiple admissions. Experimental results demonstrate that the proposed model significantly outperforms state-of-the-art baselines, and may derive a set of practical implications that can benefit a wide spectrum of management and applications on disease progression.

Patterns and determinants of carbon emission flows along the Belt and Road from 2005 to 2030

The Belt and Road Initiative (BRI) has promoted economic growth of participating countries while giving rise to profound environmental consequences. To steer the BRI towards a low-carbon and green development, it is necessary to analyze past trajectories and future trends of BRI's CO2 emissions. To this end, we assess the patterns and determinants of emission flows along the BRI during 2005–2030 using the multi-region structural decomposition analysis technique. For the period 2005–2015, we show that intermediates export of the BRI embodied more CO2 emissions than final goods export. The significant technological improvement only partly offset the emission growth stemming from the deteriorated cross-border production structure and the surging final demand of the BRI in the past. For the period 2015–2030, our prospective analysis indicates that emissions embodied in exports of participating countries increase by over 20% in the reference scenario where historical development patterns of the BRI continue. The rise might be even higher if the initiative ends. On the contrary, enhancing diffusion and adoption of low-carbon technologies and promoting green trade within the BRI show substantial emission mitigation potential. Our empirical results reveal directions and priorities for policymaking in the pursuit of a green BRI.

Predicting humans future motion trajectories in video streams using generative adversarial network

Understanding the behavior of human motion in social environments is important for various domains of a smart city, e.g, smart transportation, automatic navigation of service robots, efficient navigation of autonomous cars and surveillance systems. Examining past trajectories or environmental factors alone are not enough to address this problem. We propose a novel methodology to predict future motion trajectories of humans based on past attitude of individuals, crowd attitude and environmental context. Many researchers have proposed different techniques based on different features extraction and features fusion to predict the future motion trajectory. They used traditional machine learning algorithms like SVM,social forces, probabilistic models and LSTM to analyze the heuristic motion trajectories but they didn’t consider the other environmental factors e.g relative positions of other humans present in environment and positions of objects present in environment which can affect the motion trajectories of humans. We intend to achieve this goal by employing Long Short Term Memory(LSTM) units to analyze motion histories, convolution neural networks to environmental facts e.g. human-human, human-object interaction and relative positioning of 80 different objects including pedestrians and generative adversarial networks(GANs) to predict possible future motion paths. Our proposed method achieved 70% lower Average Displacement Error(ADE) and 41% lower Final Displacement Error(FDE) in comparison to other state of the art techniques.

Defining the “Ice Shed” of the Arctic Ocean's Last Ice Area and Its Future Evolution

AbstractSea ice will persist longer in the Last Ice Area (LIA), north of Canada and Greenland, than elsewhere in the Arctic. We combine earth system model ensembles with a sea‐ice tracking utility (SITU) to explore sources of sea ice (the “ice shed”) to the LIA under two scenarios: continued high warming (HW) rates and low warming (LW) rates (mean global warming below ca. 2°C) through the 21st century. Until mid‐century, the two scenarios yield similar results: the primary ice source shifts from the Russian continental shelves to the central Arctic, mobility increases, and mean ice age in the LIA drops from about 7 years to less than one. After about 2050, sea ice stabilizes in the LW scenario, but continues to decline in the HW scenario until LIA sea ice is nearly entirely seasonal and locally formed. Sea ice pathways through the ice shed determine LIA ice conditions and transport of material, including biota, sediments, and pollutants (spilled oil and industrial or agricultural contaminants have been identified as potential hazards). This study demonstrates that global warming has a dramatic impact on the sources, pathways and ages of ice entering the LIA. Therefore, we suggest that maintaining ice quality and preserving ice‐obligate ecologies in the LIA, including the Tuvaijuittuq Marine Protected Area north of Nunavut, Canada, will require international governance. The SITU system used in this study is publicly available as an online utility to support researchers, policy analysts, and educators interested in past and future sea ice sources and trajectories.