No 3 (2024)

This paper focuses on the problem of reduction of the computation load for road scene text recognition by making a stopping decision which cuts off further recognition. The contribution of the paper is the construction of stopping rules for real-time text recognition systems with results combination, with an experimental evaluation on an open dataset RoadText-1k. We found that for fast-working systems the ROVER (Recognizer Output Voting Error Reduction) combination method and majority voting are best for Levenshtein and direct match metrics respectively, however, with an increase of per-frame processing time, ROVER becomes consistently better. Furthermore, while the selection of a single most focused frame is the worst strategy for fast-working systems, its comparative rank increases with the increase of processing time. Moreover, choosing one most focused frame and combining three most focused frames are preferable for fast-working systems when decreasing load on the device is needed.

Map building is one of the key tasks of autonomous mobile robots’ navigation. Traditional mapping methods build dense metric map (e.g. as an occupancy grid). Maintaining such map in case of long-term navigation is difficult because of high computational costs and odometry error accumulation. Representing the environment as a sparse topological structure (e.g. a graph of locations) lets us eliminate these drawbacks and provide fast path planning. In this work, we propose a topological mapping method which builds and updates a graph of locations without use of global metric coordinates. For localization, the proposed method uses neural network-based place recognition in pair with 2D projection-based scan matching. We carry out experiments with our method in several photorealistic simulated scenes and on data from a real robot. In simulation, we compare our method with some state-of-the-art topological mapping methods. According to the results, the proposed method significantly outperforms competitors in terms of navigational efficiency, keeping graph connectivity, high scene coverage, and low part of inconsistent edges.

The reasonability of developing a conservative type DBMS with episodic data updating is determined by the features of OLAP-technologies. The issues of creating such DBMSs require serious discussion. In this review we systematize the main results of research of the research group of Clusterix KNITU-KAI on conservative DBMSs based on computational clusters. The purpose of the performed researches is actual: development of approaches to synthesize comparatively effective by the criterion “performance/cost” domestic Big Data class DBMSs. The comparison was made with the best foreign open systems. The developed DBMSs are available for use by organizations with limited financial resources. Due attention is paid to the elements of the theory of cluster DBMS of the conservative type. The basic configurations of Clusterix systems, the dynamics of such DBMSs, and the effects of their self-organization are considered. The research is based on the constructive system modeling methodology.

The paper considers the classification of safety functions of dynamic technical systems (DTS) according to various criteria and shows the connection with the main tasks of a safety monitor. It presents accessibility as a critical property of DTS safety and proposes a risk-based methodology for calculating accessibility as a function of DTS safety in an intelligent transport environment.

In order to increase the reliability of feasibility studies for the development of intelligent transport infrastructure in the context of the sustainable development of megacities, it is necessary to more carefully take into account the actual operating conditions of single-level intersections of traffic flows. The solution of this problem requires the use of advanced economic and mathematical tools. The mathematical description is a modification of the fundamental results of E. Borel and F. Haight related to the simplest flow of events. Applications of this method will allow both assessing the economic feasibility of expanding the existing transport infrastructure and designing new transport facilities. This approach can be applied at the stage of the initial quantitative assessment of the feasibility of replacing a one-level intersection of traffic flows with a multi-level interchange to avoid economic losses.

The formation of raw material supply chains is closely related to production problems at a timber processing enterprise. This article examines a forestry enterprise without its own sources of raw materials allotments, which sets itself the goal every day of solving the problem of forming supply chains for raw materials from the exchange and the optimal loading of production workshops. A commodity exchange is considered as a source of raw materials, where lots appear every day in different regions in a random order. The paper considers a mathematical model, which represents a mechanism for making decisions on each individual day over the entire planning horizon and is distinguished by the fact that it allows taking into account cutting technology, lot travel time under conditions of uncertainty. An additional difference is that the search for a solution is based on the calculated optimal trajectories of raw material inventories in the warehouse for similar previous planning periods. The model was tested on data from the Russian Commodity and Raw Materials Exchange and one of the Primorsky Territory enterprises. Analysis of the solutions obtained showed that even with a rather complex scenario for the development of events, the model shows qualitatively high results.