Hierarchical design for distributed MOPSO using sub-swarms based on a population Pareto fronts analysis for the grasp planning problem

Fdhila, Raja; Walha, Chiraz; Hamdani, Tarek M.; Alimi, Adel M.

doi:10.1109/his.2013.6920483

Cited by 9 publications

(21 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our main case of study consists of the comparison of performance between the proposed DPb-MOPSO and several algorithms regrouped as follow: [7] is the MOPSO-based approach using a distributed architecture for static MOPs.…”

Section: A Evolutionary and Swarm Intelligence Multi-objective Algormentioning

confidence: 99%

DPb-MOPSO: A Novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization Algorithm

Aboud¹,

Fdhila²,

Hussain³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Distributed architecture-based Particle Swarm Optimization is very useful for static optimization and not yet explored to solve complex dynamic multi-objective optimization problems. This study proposes a novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization (DPb-MOPSO) algorithm with two optimization levels. In the first level, all solutions are optimized in the same search space and the second level is based on a distributed architecture using the Pareto ranking operator for dynamic multi-swarm subdivision. The proposed approach adopts a dynamic handling strategy using a set of detectors to keep track of change in the objective function that is impacted by the problem’s time-varying parameters at each level. To ensure timely adaptation during the optimization process, a dynamic response strategy is considered for the reevaluation of all non-improved solutions, while the worst particles are replaced with a newly generated one. The convergence and<br>diversity performance of the DPb-MOPSO algorithm are proven through Friedman Analysis of Variance, and the Lyapunov theorem is used to prove stability analysis over the Inverted Generational Distance (IGD) and Hypervolume Difference (HVD) metrics. Compared to other evolutionary algorithms, the novel DPb-MOPSO is shown to be most robust for solving complex problems over a range of changes in both the Pareto Optimal Set and Pareto Optimal Front. <br>

show abstract

Section: A Evolutionary and Swarm Intelligence Multi-objective Algormentioning

confidence: 99%

DPb-MOPSO: A Novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization Algorithm

Aboud¹,

Fdhila²,

Hussain³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…This section presents the empirical study referring to the contributions [13] and [19]. The proposed DPb-MOPSO approach is compared to six transfer learning-based methods [13], five MOEAs [19], and four MOPSO-based approaches [9] , [12], [14], [20]. In the following experimentations algorithms are executed during 30 independent runs.…”

Section: Experimental Study: Preliminariesmentioning

confidence: 99%

“…In addition, reinitialization and hypermutation have been explicitly used to ensure good convergence and diversity in the dynamic search space [10], [11]. Nonetheless, genetic and population-based strategies on distributed methods are considered for static optimization [12] and have not yet been processed for DMOPs. Besides, many of the transfer learning-based methods [13] proposed for DMOPs are time-consuming, which increases the diminishing of solutions diversity.…”

mentioning

confidence: 99%

“…Including, five MOEAs, namely Multiobjective Evolutionary Algorithm Based on Decomposition (MOEA/D) [15], dynamic Competitive-cooperative coevolutionary Algorithm (dCOEA) [16], Population Prediction Strategy (PPS) [17], Dynamic Non-dominated Sorting Genetic Algorithm II (DNSGA-II) [18] and Steady-State Generational Evolutionary Algorithm (SGEA) [19]. Four PSO-based methods which are: Dual multi-objective particle swarm optimization (dMOPSO) [20], distributed Multi-Objective Particles Swarms Optimization, based on dynamic subdivision of the population using Pareto fronts (pbMOPSO) [12] and Multi-Objective Particles Swarms Optimization (MOPSO) [14], and Dynamic-MOPSO [9]. The six methods based on transfer learning [13] are; MMTL-MOEA/D, RIMOEA/D [13], KF-MOEA/D [21], PPS-MOEA/D [17], SVR-MOEA/D [22] and Tr-MOEA/D [23] approaches.…”

mentioning

confidence: 99%

See 1 more Smart Citation

DPb-MOPSO: A Novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization Algorithm

Aboud¹,

Rokbani²,

Fdhila³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Particle swarm optimization system based on the distributed architecture has shown its efficiency for static optimization and has not been studied to solve dynamic multiobjective problems (DMOPs). When solving DMOP, tracking the best solutions over time and ensuring good exploitation and exploration are the main challenging tasks. This study proposes a novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization (DPb-MOPSO) algorithm including two parallel optimization levels. At the first level, all solutions are managed in a single search space. When a dynamic change is successfully detected, the Pareto ranking operator is used to enable a multiswarm subdivisions and processing which drives the second level of enhanced exploitation. A dynamic handling strategy based on random detectors is used to track the changes of the objective function due to time-varying parameters. A response strategy consisting in re-evaluate all unimproved solutions and replacing them with newly generated ones is also implemented. Inverted generational distance, mean inverted generational distance, and hypervolume difference metrics are used to assess the DPb-MOPSO performances. All quantitative results are analyzed using Friedman's analysis while the Lyapunov theorem is used for stability analysis. Compared with several multi-objective evolutionary algorithms, the DPb-MOPSO is robust in solving 21 complex problems over a range of changes in both the Pareto optimal set and Pareto optimal front. For 13 UDF and ZJZ functions, DPb-MOPSO can solve 8/13 and 7/13 on IGD and HVD with moderate changes. For the 8 FDA and dMOP benchmarks, DPb-MOPSO was able to resolve 4/8 with severe change on MIGD, and 5/8 for moderate and slight changes. However, for the 3 kind of environmental changes, DPb-MOPSO assumes 4/8 of the solving function on IGD and HVD. <br>

show abstract

“…The reinitialization of certain percentage of population, and the hypermutation are explicitly used to ensure good convergence and diversity in the dynamic search space [14], [15]. Nonetheless, genetic and population-based strategies on distributed methods are considered for static optimization [16] and have not yet been processed for DMOPs. Besides, many of the transfer learning-based methods proposed for DMOPs are time-consuming, and increases the diminishing of solutions diversity [17].…”

Section: Introductionmentioning

confidence: 99%

DPb-MOPSO: A Novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization Algorithm

Aboud¹,

Rokbani²,

Fdhila³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Particle swarm optimization system based on the distributed architecture over multiple sub-swarms has shown its efficiency for static optimization and has not been studied to solve dynamic multi-objective problems (DMOPs). When solving DMOP, tracking the best solutions over time and ensuring good exploitation and exploration are the main challenging tasks. This study proposes a novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization (DPb-MOPSO) algorithm including two parallel optimization levels. At the first level, all solutions are managed in a single search space. When a dynamic change is successfully detected in the objective values, the Pareto ranking operator is used to enable a multiple sub-swarm’ subdivisions and processing which drives the second level of enhanced exploitation. A dynamic handling strategy based on random detectors is used to track the changes of the objective function due to time-varying parameters. A response strategy consisting in re-evaluate all unimproved solutions and replacing them with newly generated ones is also implemented. Inverted generational distance, mean inverted generational distance, and hypervolume difference metrics are used to assess the DPb-MOPSO performances. All quantitative results are analyzed using Friedman's analysis of variance while the Lyapunov theorem is used for stability analysis. Compared with several multi-objective evolutionary algorithms, the DPb-MOPSO is robust in solving 21 complex problems over a range of changes in both the Pareto optimal set and Pareto optimal front. For 13 UDF and ZJZ functions, DPb-MOPSO can solve 8/13 and 7/13 on IGD and HVD with moderate changes. For the 8 FDA and dMOP benchmarks, DPb-MOPSO was able to resolve 4/8 with severe change on MIGD, and 5/8 for moderate and slight changes. However, for the 3 kind of environmental changes, DPb-MOPSO assumes 4/8 of the solving function on IGD and HVD metrics.<br>

show abstract

Hierarchical design for distributed MOPSO using sub-swarms based on a population Pareto fronts analysis for the grasp planning problem

Cited by 9 publications

References 18 publications

DPb-MOPSO: A Novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization Algorithm

DPb-MOPSO: A Novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization Algorithm

DPb-MOPSO: A Novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization Algorithm

DPb-MOPSO: A Novel Dynamic Pareto bi-level Multi-Objective Particle Swarm Optimization Algorithm

Contact Info

Product

Resources

About