检索结果-内蒙古大学图书馆

evolution-algorithm-based unmanned aerial vehicles path planning in complex environment

COMPUTERS & ELECTRICAL ENGINEERING 2019年 80卷 106493-000页

作者： Liu, Xiaolei Du, Xiaojiang Zhang, Xiaosong Zhu, Qingxin Guizani, Mohsen Univ Elect Sci & Technol China Sch Informat & Software Engn Chengdu Sichuan Peoples R China Temple Univ Dept Comp & Informat Sci Philadelphia PA 19122 USA Univ Elect Sci & Technol China Ctr Cyber Secur Chengdu Sichuan Peoples R China Qatar Univ Dept Comp Sci & Engn Doha Qatar

With the wide application of Unmanned Aerial Vehicles (UAVs) in production and life, more and more attention has been paid to the autonomous track planning of UAVs. When UAV path planning algorithm is dealing with flying in an unknown complex environment, there are some problems, such as inability to dynamically plan the track and slow speed to calculate the path. This paper proposes a dynamic path planning based on an improved evolutionary optimization algorithm. The experimental results show that the evolutionary optimization algorithm based on improved t-distribution can effectively deal with the problems of high computational complexity and low search efficiency encountered in UAV dynamic track planning. It has strong robustness and can dynamically plan the appropriate track. (C) 2019 Elsevier Ltd. All rights reserved.

关键词： UAV Dynamic planning Path planning evolution algorithm

来源：评论

学校读者我要写书评

暂无评论

MFNAS: Multi-fidelity Exploration in Neural Architecture Search with Stable Zero-Shot Proxy 21st

MFNAS: Multi-fidelity Exploration in Neural Architecture Sea...

引用

21st Pacific Rim International Conference on Artificial Intelligence

作者： Fu, Wei Lou, Wenqi Qin, Yunji Gong, Lei Wang, Chao Zhou, Xuehai Univ Sci & Technol China Hefei Peoples R China

ISBN: (纸本)9789819601158;9789819601165

Neural architecture search (NAS) automates the design of neural networks for specific tasks. Recently, zero-shot NAS has attracted much attention. Unlike traditional NAS, which relies on training to rank architectures, zero-shot NAS uses gradients or activation information to evaluate architecture performance. However, existing zero-shot NAS methods are limited by their inconsistent architecture ranking and the evaluation bias of their search algorithm, making it challenging to discover networks with high accuracy efficiently. To address this dilemma, this paper proposes an efficient and stable search framework for zero-shot NAS. Firstly, we design a stable zero-shot proxy, which achieves good consistency with network accuracy by utilizing filtered gradient information. On this basis, we employ a multi-fidelity evolutionary algorithm for efficient exploration. This algorithm utilizes multi-fidelity proxies to correct the bias towards certain types of networks and enhances the ability to distinguish high-performing architectures, achieving rapid convergence through performance-directed multi-point crossover and mutation. Experimental results conducted on NATS-Bench demonstrate that our framework can discover high-performance architectures within minutes of GPU time, outperforming existing training-free and training-based NAS methods. The code is available at https://***/mine7777/MFNAS.

关键词： Neural Architecture Search evolution algorithm Zero-Shot

来源：评论

学校读者我要写书评

暂无评论

Gravitational Co-evolution and Opposition-based Optimization algorithm

引用

INTERNATIONAL JOURNAL OF COMPUTATIONAL INTELLIGENCE SYSTEMS 2013年第5期6卷 849-861页

作者： Lou, Yang Li, Junli Shi, Yuhui Jin, Linpeng Xian Jiaotong Liverpool Univ Dept Elect & Elect Engn Suzhou 215123 Peoples R China Sichuan Normal Univ Coll Comp Sci Chengdu 610066 Peoples R China Chinese Acad Sci Suzhou Inst Nanotech & Nanob Suzhou 215123 Peoples R China

In this paper, a Gravitational Co-evolution and Opposition-based Optimization (GCOO) algorithm is proposed for solving unconstrained optimization problems. Firstly, under the framework of gravitation based co-evolution, individuals of the population are divided into two subpopulations according to their fitness values (objective function values), i.e., the elitist subpopulation and the common subpopulation, and then three types of gravitation-based update methods are implemented. With the cooperation of opposition-based operation, the proposed algorithm conducts the optimizing process collaboratively. Three benchmark algorithms and fifteen typical benchmark functions are utilized to evaluate the performance of GCOO, where the substantial experimental data shows that the proposed algorithm has better performance with regards to effectiveness and robustness in solving unconstrained optimization problems.

关键词： Gravitation evolution algorithm Co-evolution Opposition-based Optimization

来源：评论

学校读者我要写书评

暂无评论

Optimization of the Mean Radius Flow Path of a Multi-Stage Steam Turbine with evolution algorithms

引用

Journal of Thermal Science 2011年第4期20卷 318-323页

作者： Jukka Huttunen Jaakko Larjola Teemu Turunen-Saaresti Jari Backman Savonia Power ljytie 678710 VarkausFinlandjukka Lappeenranta University of Technology P.O.Box 2053851 LappeenrantaFinland

A prototype model of the mean radius flow path of a four-stage, high speed 1 MWe axial steam turbine was optimized by using evolution algorithms, DE （differential evolution） algorithm in this case. Also the cost-benefits of the optimization were inspected. The optimization was successfully performed but the accuracy of the optimization was slightly less than hoped when compared to the control modeling executed with the CFD （computational fluid dynamics）. The mentioned inaccuracy could have been hardly avoided because of problems with an initial presumption involving semi-empiric calculations and of the uncertainty concerning the absolute areas of qualification of the functions. This kind of algebraic modeling was essential for the success of the optimization because e.g. CFD-calculation could not have been done on each step of the optimization. During the optimization some problems occurred with the adequacy of the computer capacity and with finding a suitable solution that would keep the algorithms within mathematically allowable boundaries but would not restrict the progress of the opti- mization too much. The rest of the problems were due to the novelty of the application and problems with pre- ciseness when handling the areas of qualification of the functions. Although the accuracy of the optimization re- suits was not exactly in accordance with the objective, they did have a favorable effect on the designing of the turbine. The optimization executed with the help of the DE-algorithm got at least about 3.5 % more power out of the turbine which means about 150 000 ε cost-benefit per turbine in the form of additional electricity capacity.

关键词： steam turbine evolution algorithm optimization mean radius flow path semi-empirical loss correlations

来源：评论

学校读者我要写书评

暂无评论

A Community Detection algorithm Using Differential evolution 3

A Community Detection Algorithm Using Differential Evolution

引用

3rd IEEE International Conference on Computer and Communications (ICCC)

作者： Sun, Hanlin Ma, Sugang Wang, Zhongmin Xian Univ Posts & Telecommun Sch Comp Sci & Technol Xian 710121 Shaanxi Peoples R China Xian Univ Posts & Telecommun Shaanxi Key Lab Network Data Intelligent Proc Xian 710121 Shaanxi Peoples R China

ISBN: (纸本)9781509063529

evolution algorithm (EA) is an attractive way for community detection (CD) on complex networks since it does not require any prior knowledge about the community structure, which is usually not available for real-world networks. The differential evolution (DE) is claimed as one of the most powerful EA currently in use for real-parameter optimization. One fundamental issue of EA is how to encode a problem solution into an evolution individual. In this paper, the indexed locus-based adjacency representation (ILAR) for evolution individual encoding in CD problem is proposed. As a result, a CD problem could be converted to a discrete integer optimization problem. Based on the ILAR, a community detection algorithm using DE as the search engine is developed. A number of experiments are conducted on synthesized and real-world networks, and the results show that the proposed DE based algorithm is very competitive for community detection as the community structure of a network is sound.

关键词： community detection community structure analysis differential evolution evolution algorithm complex network

来源：评论

学校读者我要写书评

暂无评论

Sensor Deployment Method Through Maximizing Reconfigurability to Reduce Degradation in TDOA Location System

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Wei, Juhui Bai, Zhenzu Wang, Jiongqi Zhou, Xuanying Hou, Bowen Natl Univ Def Technol Coll Sci Changsha 410073 Peoples R China

Measurement anomalies may cause a sudden system performance degradation, especially in location systems based on time difference of arrivals (TDOAs) due to its limited number of redundant measurement instruments. Conventional approaches mitigate measurement anomalies through data processing techniques such as anomaly detection and system reconfiguration algorithms. However, these approaches may prove inadequate when anomalies affect critical system configurations. This study presents an innovative reconfigurability-enhanced deployment to reduce degradation in the presence of abnormal measurements. First, we introduce the model, algorithm, and performance evaluation of the TDOA location, along with a geometric dilution of precision (GDOP) metric to quantify the impact of system deployment independently of measurement uncertainty. Then, we elaborate on measurement anomalies and the existing strategies for system diagnosis and reconfiguration. Following this, we provide detailed analyses of system degradation risks under various strategies. Subsequently, we establish reconfigurability to assess the system's tolerance to anomalies. An adapted evolutionary algorithm is proposed to design an optimal deployment that enhances reconfigurability. Finally, in accordance with real applications, we conduct simulations and semiphysical experiments to evaluate the performance of different deployments under anomalies. The results demonstrate that the proposed deployment effectively mitigates degradation risks while maintaining system performance.

关键词： Diagnosis evolution algorithm measurement anomalies performance evaluation reconfiguration time difference of arrival (TDOA) localization Diagnosis evolution algorithm measurement anomalies performance evaluation reconfiguration time difference of arrival (TDOA) localization

来源：评论

学校读者我要写书评

暂无评论

The University Teacher Title Stable Structural Model and the evolution Solutions

The University Teacher Title Stable Structural Model and the...

引用

3rd International Symposium on Intelligent Information Technology Application

作者： Qu Wen-jian Xu Xiao-fang Tan Guang-xing Wei Ying Jiangxi Univ Finance & Econ Sch Informat Management Nanchang Peoples R China Nanchang Univ Sch Informat Engn Nanchang Peoples R China Jiangxi Univ Finance & Econ Inst Higher Educ Res Nanchang Peoples R China

ISBN: (纸本)9780769538594

The teachers' title be divided into the professor, the associate professor, the lecturer and the teaching assistant in china's university, the different titles form a structure. How to seek the stable structure, to be very difficulty, and very few people studied on the question. In this paper, we establish a Markov model for the university teachers' title structure, and establish the optimized model for the ideal title structure, and we use the evolution algorithm to solve it directly, and we analyze teacher title structure under changed condition of university in China for the recent 5 years, the result is effective. Our work should be applied to forecast the title structure and to seek the ideal title structure for various china universities in the future.

关键词： Markov process Stable title structure Transition probability evolution algorithm

来源：评论

学校读者我要写书评

暂无评论

Pseudo Multi-Population Differential evolution for Multimodal Optimization 10

Pseudo Multi-Population Differential Evolution for Multimoda...

引用

10th International Conference on Natural Computation (ICNC)

作者： Li, Hao-Feng Gong, Yue-Jiao Zhan, Zhi-Hui Chen, Wei-Neng Zhang, Jun Sun Yat Sen Univ Engn Res Ctr Supercomp Engn Software Dept Comp Sci Key Lab Machine Intelligence & Adv CompMinist Ed Guangzhou 510006 Guangdong Peoples R China

ISBN: (纸本)9781479951512

Multimodal optimization aims at locating multiple optima in a run, which has two main advantages over traditional single objective global optimization. First, it would be useful to provide multiple solutions since some solutions may be hard to realize physically. Second, a multimodal algorithm is not so easy to get stuck in a local optimum. In recent years, multi-population evolutionary algorithms have been used for multimodal optimization. However, their ability to locate multiple peaks is limited by the number of populations used. It is difficult to find out all the peaks if the populations are fewer than the peaks. When algorithms increase the number of populations, they have to maintain huge population sizes and hence encounter lower search efficiency. This paper overcomes such deficiencies by proposing a pseudo multi-population differential evolution (p-MPDE). The p-MPDE employs a small exemplar population to conduct normal DE operation. Each other individual uses the differential of two randomly chosen members in the exemplar population to mutate themselves and evolve. Each such individual represents a pseudo population and promises to find a global or local optimum. In the experiment, p-MPDE was compared to other state-of-the-art multimodal algorithms and the result shows that p-MPDE outperforms R3PSO, LIPS and CDE on CEC2013 niching benchmark.

关键词： evolution algorithm Multimodal Optimization Differential evolution

来源：评论

学校读者我要写书评

暂无评论

Research on Regional Collaborative Innovation evolution from Complex Network Perspective 9

Research on Regional Collaborative Innovation Evolution from...

引用

9th International Conference of Information and Communiation Technologies (AICT)

作者： Wang, Kun Sun, Duoyong Natl Univ Def Technol Coll Informat Syst & Management Changsha Hunan Peoples R China

ISBN: (纸本)9781467368551

This paper proposed a novel algorithm to analyze the evolution of regional collaborative innovation systems from the complex network perspective. At first, we illustrate the structure of collaborative networks, which contains several elements. And we found out the system could be regarded as a complex network. As to the evolution algorithm we proposed, in order to match the actual situation, each node in the network can only connect to less than a specific number of its neighbors. Specifically, for a certain node, maximum number of the final edges is only determined by its node importance. We collect some raw data from the Internet. Finally, we carried out a study to compare the evolution situation in both small and large scare network. Experimental results have demonstrated the proposed algorithm is valid. Simulation results are consistent well with the realities of situation. Moreover, some beneficial conclusions can be summarized from our paper.

关键词： Regional collaborative innovation Complex network evolution algorithm Statistical yearbook

来源：评论

学校读者我要写书评

暂无评论

Winner-Take-All Memetic Differential evolution for Genetic Interaction: Parameter Identification

Winner-Take-All Memetic Differential Evolution for Genetic I...

引用

14th ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)

作者： Wu, Shinq-Jen Wu, Cheng-Tao Da Yeh Univ Dept Elect Engn Changhua Taiwan Natl Chiao Tung Univ Dept Elect & Control Engn Hsinchu 300 Taiwan

ISBN: (纸本)9780769550053

The emerging large-scale biological tools (e. g., microarray) challenge biologists to realize the connectivity of genes and/or proteins at the system level (global view). Having advantages in good generalization and showing the direct interaction of genes and/or proteins, the S-system becomes one of the popular models, which is able to capture the dynamic behavior of the biological system. Differential evolution (DE) and its variants have recently applied to solve various optimization problems in engineering fields. However, the exploitative and explorative abilities are insufficient. In this study, we propose a winner-take-all memetic differential evolution scheme to infer the parameters of the S-type gene regulatory networks. This method was tested with a genetic-branch pathway and a twenty-gene network. The learning was implemented in a wide search space ([0, 100] for rate constants and [-100, 100] for kinetic orders) with a bad initial start (All parameters were randomly initialized at the neighborhood of 80). Simulation results show high-accuracy solutions are obtained.

关键词： inverse problem memetic algorithm evolution algorithm S-system

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：