Parallelized Multiple Swarm Artificial Bee Colony (PMS-ABC) Algorithm for Constrained Optimization Problems

Subotic, Milos; Manasijevic, Aleksandar; Kupusinac, Aleksandar

doi:10.24846/v29i1y202008

Cited by 4 publications

(3 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This algorithm was used to solve 24 benchmarks and 3 engineering problems, and high-quality results were obtained. When M. Subotic [29] improved ABC, the population was divided into several subspecies and then calculated. The modified algorithm is called the parallel multiple-swarm artificial bee colony (PMS-ABC).…”

Section: Related Workmentioning

confidence: 99%

A Modification of the Imperialist Competitive Algorithm With Hybrid Methods for Constrained Optimization Problems

2021

View full text Add to dashboard Cite

This paper studies a modification of the imperialist competitive algorithm to solve constrained optimization problems with hybrid methods. The imperialist competitive algorithm is a kind of evolutionary algorithm based on the colonial competition mechanism of imperialism, which is a type of social heuristic optimization algorithm. However, this algorithm needs to be modified because of some problems, including the decreasing number of empires, which leads to easily falling into a local optimum, and a lack of information exchange among countries leading to the search ability being insufficiently strong.To solve these problems and improve the search ability of the algorithm, a newly improved imperialist competitive algorithm with hybrid methods is proposed based on the imperialist competitive algorithm in this paper. In this algorithm, a crossover mechanism is introduced as the empire interaction to boost the information exchange among countries. Two groups of combinatorial operators are constructed in the process of empire assimilation and revolution to improve the algorithm's global and local search ability. Moreover, a step of empire splitting is added based on the imperialist competitive algorithm in which the number of empires is decreased rapidly by empire competition in imperialist competitive algorithm, which leads to a decrease in the population diversity. Then, the proposed algorithm is tested on 12 benchmark functions and 4 engineering problems, and the constraints are addressed by Deb's rules. The results obtained by the proposed algorithm are compared with those of algorithms that have obtained good results in recent years, and it is shown that the proposed method obtains promising and comparable results on the constrained optimization benchmarks in terms of the solution quality and robustness.INDEX TERMS Constrained optimization problems, Hybrid methods, Imperialist competitive algorithm.

show abstract

Section: Related Workmentioning

confidence: 99%

A Modification of the Imperialist Competitive Algorithm With Hybrid Methods for Constrained Optimization Problems

2021

View full text Add to dashboard Cite

show abstract

“…Among them, the serial hybrid performance is restricted by the sequence of the two algorithms and the switching point of the two algorithms, while embedded hybrid performance is related to the embedding mode of the two related algorithms. Parallel hybrid makes full use of the performance of PSO and GA, has a wide search range and faster convergence, which is usually used in the case of parameter optimization (Subotic, Manasijevic, & Kupusinac, 2020).…”

Section: Pso-ga Hybrid Optimization Algorithmmentioning

confidence: 99%

Design of Improved PID Controller Based on PSO-GA Hybrid Optimization Algorithm in Vehicle Lateral Control

GAO¹,

Gao²,

Pan³

et al. 2021

STUD INFORM CONTROL

View full text Add to dashboard Cite

Considering that it is difficult for a single PID controller to adapt to different vehicle speeds in the context of vehicle path lateral tracking control, this paper proposes a segmented fuzzy PID controller based on particle swarm optimization (PSO) and on the genetic algorithm (GA), namely the hybrid optimization algorithm PCAG. Firstly, the vehicle speed is divided into several intervals, and different PID controller parameters are used for each interval. Secondly, in order to reduce the overshoot and stabilization time, the proposed PCAG algorithm is employed, which is a combination of PSO, particle swarm optimization with convergence factor (PSO-CF), adaptive particle swarm optimization (APSO) and GA. Further on, the PID controller parameters of different speed ranges are adjusted through this algorithm. Finally, in order to make up for the shortcomings of a single PID controller in the context of time-varying vehicle speed control, a fuzzy controller is employed with the purpose of compensating the parameters of the PID controller, so that the controller could adapt to a wider range of vehicle speeds. The simulation results show that the convergence speed and optimization ability of the proposed PCAG are higher than those of PSO. In addition, the segmented fuzzy PID controller optimized by PCAG can adapt to different vehicle speeds and features an excellent path tracking accuracy.

show abstract

“…Farenzena, et al (2010) presented a symmetry-driven accumulation of local features (SDALF) method with symmetry and asymmetry to address viewpoint variability. The metric system is a very important element in person Re-ID, and many methods have been applied in computer vision and proved to be effective (Subotic et al, 2020, Yousuf Uddin et al, 2021, Rădulescu & Rădulescu, 2020, 2021b, Ma et al, 2020, for example, Keep It Simple and Straightforward Metric learning (KISSME) (Kostinger et al, 2012), Locally-Adaptive Decision Function (LADF) (Li et al, 2013) and Cross-View Quadratic Discriminant Analysis (XQDA) (Liao et al, 2015). These algorithms have shown excellent results in face recognition and person Re-ID.…”

Section: Introductionmentioning

confidence: 99%

Fused-Grain Feature Learning for Unsupervised Person Re-identification

Han¹,

Huang²,

Zhang³

et al. 2022

STUD INFORM CONTROL

View full text Add to dashboard Cite

Most supervised learning methods are currently used to solve the task of person re-identification (Re-ID) and yield excellent results. But these methods usually need manual annotation of training data. Especially for large data sets, they need too high cost of manual annotation and the data are difficult to obtain for fully pairwise labeling. So unsupervised learning becomes a necessarily trend for person Re-ID. This paper is trying to solve the problem by unsupervised learning method. Moreover, global features focus on spatial integrity of person features, and local ones help to highlight discriminative features of different patches. Therefore, fused-grained unsupervised (FGU) learning framework of global and local branches' feature learning is proposed to solve Re-ID task. Specifically, for the local branch, one extracts patches from a feature map which learned on a PatchNet network of images, and learns their fine-grained features to pull close the similar patches and push away the dissimilar ones. For the global branch, one maximizes the diversity between classes by repelled loss and similarity within classes through attracted loss, then similarity and diversity in the unlabeled data sets are used as information for unsupervised cluster merging and learning their coarse-grained features. The two branches are used to jointly achieve the effect of increasing inter-class differences and intra-class similarity. A large number of experiments verify the superiority of the proposed method for unsupervised person re-identification.

show abstract

Parallelized Multiple Swarm Artificial Bee Colony (PMS-ABC) Algorithm for Constrained Optimization Problems

Cited by 4 publications

References 27 publications

A Modification of the Imperialist Competitive Algorithm With Hybrid Methods for Constrained Optimization Problems

A Modification of the Imperialist Competitive Algorithm With Hybrid Methods for Constrained Optimization Problems

Design of Improved PID Controller Based on PSO-GA Hybrid Optimization Algorithm in Vehicle Lateral Control

Fused-Grain Feature Learning for Unsupervised Person Re-identification

Contact Info

Product

Resources

About