Сценарное моделирование движения беспилотных транспортных средств в искусственной дорожной сети с использованием FLAME GPU
This article presents a model of the ground autonomous vehicles (AVs) motion in the Artificial Road Network (ARN) belonging to the "Manhattan Lattice" type with the implementation of the large-scale agent-based modeling framework FLAME GPU. The most important scenarios of the traffic situation development are investigated, in particular, which are associated with reducing visibility on the roads, especially in conditions of unusual behaviour of some agents of the traffic system, e.g. the unexpected appearance of obstacles such as agent-pedestrians and chaotic maneuvering of usual (i.e. manned) vehicles (MVs) having abnormal characteristics. A new approach to designing large-scale agent-based transportation simulations based on ARNs with a complex configuration and implementation using supercomputer technologies is proposed.
This work is devoted to the development of an evolutionary algorithm for fuzzy clustering of an ensemble of interacting conventional and unmanned vehicles in order to identify the relationship between stable groups of agents and initial modeling parameters.
During the last years, new technologies have been developing at a rapid pace; however, new technologies carry risks and uncertainties. Technology forecasting by analogy has been used in the case of emerging technologies; nevertheless, the use of analogies is subject to several problems such as lack of inherent necessity, historical uniqueness, historically conditioned awareness, and casual analogies. Additionally, the natural process of selecting the analogy technology is based on subjective criteria for technological similarities or inductive inference. Since many analogies are taken qualitatively and rely on subjective assessments, this paper presents a quantitative comparison process based on the Social Network Analysis (SNA) and patent analysis for selecting analogous technologies. In this context, the paper presents an analysis of complex patent network structures using centrality and density metrics in order to reduce the lack of information or the presence of uncertainties. The case of Autonomous Vehicles (AVs) is explored in this paper, comparing three candidate technologies which have been chosen based on the similarities with the target technologies. The best candidate technology is selected based on the analysis of two main centrality metrics (average degree and density).
Urban greenery such as trees can effectively reduce air pollution in a natural and eco-friendly way. However, how to spatially locate and arrange greenery in an optimal way remains as a challenging task. We developed an agent-based model of air pollution dynamics to support the optimal allocation and configuration of tree clusters in a city. The Pareto optimal solutions for greenery in the city were computed using the suggested heuristic optimisation algorithm, considering the complex absorptive-diffusive interactions between agent-trees (tree clusters) and air pollutants produced by agent-enterprises (factories) and agent-vehicles (car clusters) located in the city. We applied and tested the model with empirical data in Yerevan, Armenia, and successfully found the optimal strategy under the budget constraint: planting various types of trees around kindergartens and emission sources.
In our recent papers, we proposed a new family of residual convolutional neural networks trained for semi-dense and sparse depth reconstruction without use of RGB channel. The proposed models can be used in low-resolution depth sensors or SLAM methods estimating partial depth with certain distributions. We proposed using perceptual loss for training depth reconstruction in order to better preserve edge structure and reduce over-smoothness of models trained on MSE loss alone.
This paper contains reproducibility companion guide on training, running and evaluating suggested methods, while also presenting links on further studies in view of reviewers comments and related problems of depth reconstruction.
The effects of autonomous vehicles (AVs) on urban forms are modeled, calibrated, and analyzed. Vehicles are used for commute between peripheral home and central work, and require land for parking. An advantage of AVs is that they can optimize the location of day parking, relieving downtown land for other uses. They also reduce the per-kilometer cost of commute. Increased AV availability increases worker welfare, travel distances, and the city size. Land rents increase in the center but decrease in the periphery. Possible locations of AV daytime parking are analyzed. The effects of AV introduction on traffic and on mass transit coverage are discussed.
With advances of recent technologies, augmented reality systems and autonomous vehicles gained a lot of interest from academics and industry. Both these areas rely on scene geometry understanding, which usually requires depth map estimation. However, in case of systems with limited computational resources, such as smartphones or autonomous robots, high resolution dense depth map estimation may be challenging. In this paper, we study the problem of semi-dense depth map interpolation along with low resolution depth map upsampling. We present an end-to-end learnable residual convolutional neural network architecture that achieves fast interpolation of semi-dense depth maps with different sparse depth distributions: uniform, sparse grid and along intensity image gradient. We also propose a loss function combining classical mean squared error with perceptual loss widely used in intensity image super-resolution and style transfer tasks. We show that with some modifications, this architecture can be used for depth map super-resolution. Finally, we evaluate our results on both synthetic and real data, and consider applications for autonomous vehicles and creating AR/MR video games.
We consider certain spaces of functions on the circle, which naturally appear in harmonic analysis, and superposition operators on these spaces. We study the following question: which functions have the property that each their superposition with a homeomorphism of the circle belongs to a given space? We also study the multidimensional case.
We consider the spaces of function on the circle whose Fourier transform is p-summable. We obtain estimates for the norms of exponential functions deformed by a C1 -smooth phase.
We consider the spaces of functions on the m-dimensional torus, whose Fourier transform is p -summable. We obtain estimates for the norms of the exponential functions deformed by a C1 -smooth phase. The results generalize to the multidimensional case the one-dimensional results obtained by the author earlier in “Quantitative estimates in the Beurling—Helson theorem”, Sbornik: Mathematics, 201:12 (2010), 1811 – 1836.