Pub Date : 2023-10-17DOI: 10.1109/JMASS.2023.3325054
Jianguo Guo;Mengxuan Li;Zongyi Guo;Zhiyong She
This article proposes a novel 3-D sliding mode interception guidance law for maneuvering targets, which explores the potential of reinforcement learning (RL) techniques to enhance guidance accuracy and reduce chattering. The guidance problem of intercepting maneuvering targets is abstracted into a Markov decision process whose reward function is established to estimate the off-target amount and line-of-sight angular rate chattering. Importantly, a design framework of reward function suitable for general guidance problems based on RL can be proposed. Then, the proximal policy optimization algorithm with a satisfactory training performance is introduced to learn an action policy which represents the observed engagements states to sliding mode interception guidance. Finally, numerical simulations and comparisons are conducted to demonstrate the effectiveness of the proposed guidance law.
{"title":"Reinforcement Learning-Based 3-D Sliding Mode Interception Guidance via Proximal Policy Optimization","authors":"Jianguo Guo;Mengxuan Li;Zongyi Guo;Zhiyong She","doi":"10.1109/JMASS.2023.3325054","DOIUrl":"10.1109/JMASS.2023.3325054","url":null,"abstract":"This article proposes a novel 3-D sliding mode interception guidance law for maneuvering targets, which explores the potential of reinforcement learning (RL) techniques to enhance guidance accuracy and reduce chattering. The guidance problem of intercepting maneuvering targets is abstracted into a Markov decision process whose reward function is established to estimate the off-target amount and line-of-sight angular rate chattering. Importantly, a design framework of reward function suitable for general guidance problems based on RL can be proposed. Then, the proximal policy optimization algorithm with a satisfactory training performance is introduced to learn an action policy which represents the observed engagements states to sliding mode interception guidance. Finally, numerical simulations and comparisons are conducted to demonstrate the effectiveness of the proposed guidance law.","PeriodicalId":100624,"journal":{"name":"IEEE Journal on Miniaturization for Air and Space Systems","volume":"4 4","pages":"423-430"},"PeriodicalIF":0.0,"publicationDate":"2023-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135002415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-26DOI: 10.1109/JMASS.2023.3319579
Chuanli Wang;Tianyu Li;Dongjun Xin;Qian Wang;Ran Chen;Chaoyi Cao
Evaporation is an important part of the moisture exchange between the earth and the air. Understanding the trend of pan evaporation can help to reveal the status of actual evaporation, which is very useful for the allocation of regional water resources. However, long short-term memory (LSTM) has become a mainstream algorithm for predicting pan evaporation, there are two issues worth considering. One of the issues is how to automatically find the optimal hyperparameters, the other is how to eliminate the correlation between prediction factors to improve prediction performance. To address the two issues, this article proposes LSTM models based on principal component analysis (PCA) factor reduction and firefly optimization algorithm. In the proposed model, fire-fly algorithm can find the optimal hyperparameters, and PCA can eliminate the correlation between prediction factors. Xiangjiang River Basin, an important Basin for China’s water resource management, is selected as a study area, the experimental results are evaluated by root mean square error (RMSE) and the coefficient of determination ( $R^{2}$