Reinforcement learning is generating considerable interest in terms of building guidance law and solving optimization problems that were previously difficult to solve. Since reinforcement learning-based guidance laws often show better robustness than a previously optimized algorithm, several studies have been carried out on the subject. This paper presents a new approach to training missile guidance law by reinforcement learning and introducing some notable characteristics. The novel missile guidance law shows better robustness to the controller-model compared to the proportional navigation guidance. The neural network in this paper has identical inputs with proportional navigation guidance, which makes the comparison fair, distinguishing it from other research. The proposed guidance law will be compared to the proportional navigation guidance, which is widely known as quasi-optimal of missile guidance law. Our work aims to find effective missile training methods through reinforcement learning, and how better the new method is. Additionally, with the derived policy, we contemplated which is better, and in which circumstances it is better. A novel methodology for the training will be proposed first, and the performance comparison results will be continued therefrom.
This paper proposes that the deep neural network-based guidance (DNNG) law replace the proportional navigation guidance (PNG) law. This approach is performed by adopting a supervised learning (SL) method using a large amount of simulation data from the missile system with PNG. Then, the proposed DNNG is compared with the PNG, and its performance is evaluated via the hitting rate and the energy function. In addition, the DNN-based only line-of-sight (LOS) rate input guidance (DNNLG) law, in which only the LOS rate is an input variable, is introduced and compared with the PN and DNNG laws. Then, the DNNG and DNNLG laws examine behavior in an initial position other than the training data.
This paper presents two amplitude comparison monopulse algorithms and their covariance prediction equation. The proposed algorithms are based on the iterated least-squares estimation method and include the conventional monopulse algorithm as a special case. The proposed covariance equation is simple but predicts RMS errors very accurately. This equation quantitatively states estimation accuracy in terms of major parameters of amplitude comparison monopulse radar, and is also applicable to the conventional monopulse algorithm. The proposed algorithms and covariance prediction equations are validated by the numerical simulations with 100,000 Monte Carlo runs.
This paper presents the optimal control approach to solve both Lambert’s problem and Gibbs’ method, which are commonly used for preliminary orbit determination. Lambert’s problem is reinterpreted with Hamilton’s principle and is converted to an optimal control problem. Various extended Lambert’s problems are formulated by modifying the weighting and constraint settings within the optimal control framework. Furthermore, Gibbs’ method is also converted to an extended Lambert’s problem with two position vectors and one orbit energy with the help of the proposed orbital energy computation algorithm. The proposed extended Lambert’s problem and Gibbs’ method are numerically solved with the Lobatto pseudospectral method, and their accuracies are verified by numerical simulations.
This paper conducts a comparative analysis of three wind farm simulators, examining the influence of wake on the local wind speed and power output for downstream turbines using experimental data. The study features experiments in three distinct scenarios, evaluating differences among the simulators by calculating the local wind speed and power for each. Each simulator employs a unique wake model, which substantially affects the local wind speed experienced by downstream turbines. Furthermore, the experiment involves adjusting parameter values for each simulator to assess their respective impacts on wind farm performance. The findings of this research are expected to play an important role in investigations related to power optimization and wake effects in the wind farm control.
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.