Visual odometry in local underwater navigation problems

The work processes of the ORB-SLAM algorithm are presented. The results of experimental studies on temporal comparisons of the operation of the algorithm with different parameters and cameras are presented. The necessity of forming a visual odometry (VO) system as a local navigation of remote-controlled and autonomous underwater robots has been substantiated. The two most suitable odometry methods in the underwater environment are described, such as their advantages and disadvantages. The work processes of the ORB-SLAM algorithm are presented. The results of experimental studies on temporal comparisons of the operation of the algorithm with different parameters and cameras are presented. The procedure for preparing video data is described: processing a video stream, adjusting camera parameters for calibration. The experiments represent the testing of the ORB-SLAM3 algorithm on a sample of video filmed as part of the ecological monitoring of the Caspian shelf in 2020.

Local territorial navigation system for regional airfields of the Arctic region RF

The specific conditions peculiar to the Northern regions create a number of difficulties in the functioning of the information airfield structure based on the traditional complex of navigation aids. These difficulties are caused by the unstable operation of satellite navigation systems in the Arctic region and territorial fragmentation. There is an urgent need to develop new approaches and technical means for the deployment of local navigation systems that ensure the operation of airfields with low traffic and difficult flight conditions. The aim of the study is to develop approaches to the creation of a new generation of multifunctional radar systems for the organization of a local navigation and information system at the airfields of the Arctic region, consisting of a ground-based multifunctional station, a portable small-sized terminal installed on board the aircraft and a board-to-ground communication line. This article proposes a promising navigation and territorial system for Arctic airfields, based on a new generation of information tools, including multifunctional radar stations. The proposed local territorial navigation and system is based on the information of the ground-based multifunctional airfield control radar. The information is transmitted to a portable small-sized terminal installed on board the aircraft. In this system, when transmitting the current situation to the on-board terminal from the multifunctional radar, which is linked to the airfield control center, it provides a safer implementation of the approach and landing modes at Arctic airfields, including in the complete absence of satellite navigation. The implementation of the proposed principles and technical solutions will enable the creation of a new generation of technical tools and the introduction of new organizational approaches to the creation of local navigation systems.

Application of genetic algorithm for determining the locations of reference points of local navigation system and minimizing their number

In this paper, the problem of minimization of the number of local navigation system (LNS) reference points (RPs) and RPs’ location optimization is posed according to the criterion of ensuring the factor of geometric dilution of precision (GDOP) averaged over the space is not worse than a given one. According to authors’ best knowledge, this is the first time such a problem has been posed and solved. An approximate numerical solution to the problem based on a genetic algorithm is proposed. The effectiveness of the proposed method is assessed by simulating the placement of a different number of ranging LNS RPs in rooms of various configurations and comparing the result with the quasi-optimal result of a complete numerical search of all possible locations of the LNS anchors. The conclusion is made on the applicability of the proposed method for solving the problem and the prospects of building on its basis a tool for automating the development of positional LNS to reduce the development time and the cost of infrastructure.

Deep reinforcement learning for map-less goal-driven robot navigation

Mobile robots that operate in real-world environments need to be able to safely navigate their surroundings. Obstacle avoidance and path planning are crucial capabilities for achieving autonomy of such systems. However, for new or dynamic environments, navigation methods that rely on an explicit map of the environment can be impractical or even impossible to use. We present a new local navigation method for steering the robot to global goals without relying on an explicit map of the environment. The proposed navigation model is trained in a deep reinforcement learning framework based on Advantage Actor–Critic method and is able to directly translate robot observations to movement commands. We evaluate and compare the proposed navigation method with standard map-based approaches on several navigation scenarios in simulation and demonstrate that our method is able to navigate the robot also without the map or when the map gets corrupted, while the standard approaches fail. We also show that our method can be directly transferred to a real robot.

Simulation Model of Crowd Evacuation Navigation in a Multi-Obstacle Environment

The simulation of a crowd evacuating public buildings can be an important reference in planning the layout of buildings and formulating evacuation strategies. This paper proposes an agent-based crowd model; a crowd evacuation navigation simulation model is proposed for the multi-obstacle environment. We introduce the concept of navigation factor to describe the proximity of the navigation point to the exit. An algorithm for creating navigation points in multi-obstacle environment is proposed along with the global navigation and local navigation control algorithms of the crowd. We construct a crowd evacuation simulation prototype system with different simulation scenes using the scene editor. We conduct the crowd evacuation simulation experiment in the multi-obstacle scene, recording and analyzing the relevant experimental data. The simulation prototype system can be used to derive the evacuation time of the crowd and analyze the evacuation behavior of the crowd. It is expected to provide a visual deduction method for crowd management in an evacuation emergency.