One of the problems faced by developers of artificial intelligence algorithms when creating car control systems is that the actions of other road users are difficult to predict and have a large variability. Even if we assume that all actions comply with traffic rules and participants do not make mistakes, that is, to bring the actual environment closer to the ideal, the task of automating vehicle control still contains many difficulties. This paper describes what difficulties exist in the field of predicting the trajectory of objects, shows concepts that will help in solving this problem, and also describes a particular method of forecasting, which allows you to make a forecast for cars moving along traffic lanes. The main forecasting stages and the results of testing the method collected by using a ready-made data set are given. The results presented in the form of a set of metrics, are compared with another algorithm for predicting trajectories. As a result, the advantages and disadvantages of the created solution were identified.
The problem of creating a fully autonomous vehicle is one of the most urgent in the field of artificial intelligence. Many companies claim to sell such cars in certain working conditions. The task of interacting with other road users is to detect them, determine their physical properties, and predict their future states. The result of this prediction is the trajectory of road users’ movement for a given period of time in the near future. Based on such trajectories, the planning system determines the behavior of an autonomous-driving vehicle. This paper demonstrates a multi-agent method for determining the trajectories of road users, by means of a road map of the surrounding area, working with the use of convolutional neural networks. In addition, the input of the neural network gets an agent state vector containing additional information about the object. A number of experiments are conducted for the selected neural architecture in order to attract its modifications to the prediction result. The results are estimated using metrics showing the spatial deviation of the predicted trajectory. The method is trained using the nuscenes test dataset obtained from lgsvl-simulator.
The particulars of fuel reloading in fast reactors are described. General data on a BN-1200 powergenerating unit, the principles of the construction of its reloading system, the purpose and construction of individual types of reloading equipment, and required experimental work are presented. The comparative characteristics of the BN-800 and BN-1200 reloading equipment are presented.Reloaded fuel assemblies of fast reactors, in contrast to thermal reactors, have a high specific energy release. Inside the reactor they are moved beneath the sodium level at high temperature (to 250°C). In-reactor reloading equipment is an integral part of the first loop and ensures that the loop is hermetic. For this reason, when reloading has been completed it is not removed from the reactor and when the reactor operates at power it is subject to irradiation. Such stringent operating conditions and the impossibility of monitoring the reloading visually require that the equipment be highly reliable and the reloading process fully automated.To obtain a high capacity utilization factor, the number of reactor shutdowns for reloading and the shutdown time must be decreased. The number of reactor shutdowns for reloading is determined by the nominal fuel burnup and the energy density of the core. BN-600 and -800 require two fuel reloadings per year [1]. For BN-1200, where fuel burnup is deeper and a lower energy density is used, one reloading per year is planned. The reloading time is shortened by decreasing the time required to prepare the reloading equipment and to bring the reactor up to power. The operations include decoupling and coupling of the actuating mechanisms of the safety and control system with their rods, preparation and checking of the serviceability of the reloading equipment and its control system.The BN-600 and -800 reloading systems perform reloading of the following types of assemblies -fuel assemblies, rods and bushings of the safety and control system, and assemblies of the steel and boron protection. They are characterized by the presence of a large number of pieces of equipment and associated systems. Specifically, these systems include the storage drums for fresh and spent assemblies. Sodium is used to cool a drum of spent assemblies. For this reason, the material content of the system is high and heightened safety measures must be taken because of the presence of a radioactive sodium loop.The development of a reloading system for the advanced BN-1200 fast reactor is based on the following principles: 1) maximum possible use of scientifically tested and engineered technical solutions, implemented in and 2) application of new technical solutions, which increase safety and ensure high cost-effectiveness of the facility and efficient fuel utilization.One of the principal avenues for improving the characteristics of the power-generating unit is to increase the fuel burnup as compared with BN-600 and -800 as well as to increase the diameter of the fuel elements in order to shorten their utilization. Increasing the d...
This article presents a method for recognizing key objects of the road infrastructure using a fully convolutional neural network. The result of the neural network is a segmented image, where the desired objects are highlighted in certain colors. At the post-processing stage, a section of the roadway along which the car moves is selected, as well as the calculation of the parameters of the bounding rectangles for each of the objects. This method allows you to localize the road, pedestrian crossing, cars, traffic signs, traffic lights, pedestrians. Testing of the developed algorithm was carried out on a model of the urban infrastructure at a scale of 1:18, where a wheeled robot acted as a car.
This paper presents an approach to the unmanned control of a wheeled robot, which includes recognition of road infrastructure objects, recognition of continuous and intermittent road markings, generation of control signals. Recognition of road infrastructure objects is carried out using a neural network that generates a segmented image. After that, the segmented image is identified with the found objects, including the roadway, which is used by the road marking recognition subsystem searching for continuous and intermittent lines using the computer vision library. On the basis of the information received from the considered subsystems control commands are generated indicating the direction of movement and speed. The algorithm was developed on a 1:18 scale model of the city infrastructure, where a wheeled robot simulated as a car.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.