Cloud-Based Task Distribution System Infrastructure for Group of Mobile Robots

Migranov, A.B.

doi:10.1007/978-981-15-5580-0_33

Cited by 1 publication

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The article considers the problem of constructing an optimal algorithm (strategy) for the behavior of a group of robots on a working field containing several tasks, therefore, the objective function is constructed in the class of multi-criteria, and one of the modifications of the ant algorithm was used for synthesis. The authors have previously solved a similar problem using neural network computing [12] and genetic algorithms [13], therefore, in order to continue the research and compare the different approaches, it was decided to investigate the quality of multi-criteria optimization of the behavior strategy of a group of mobile robots using the ant algorithm. The works [14,15] show the applicability of ant algorithms related to swarm algorithms [16] for solving this class of problems when optimizing one efficiency criterion.…”

Section: Introductionmentioning

confidence: 99%

Multi-criteria Optimization of the Mobile Robot Group Strategy Using the Ant Algorithm

Darintsev

Airat

2021

Electromechanics and Robotics

View full text Add to dashboard Cite

One of the problems of group control, the distribution, of tasks in a group of mobile robots, is considered. The basic solution algorithm is proposed to use the "swarm intelligence" implemented on the basis of ant algorithms. This article is a development of the previous authors' work, which reviewed the known approaches to the distribution of problems using swarm algorithms, described the ant algorithm in detail, and analyzed the resulting solution to the problem of single-criteria optimization. The problem statement, the working space model are presented, the robot function objectives and parameters characterizing their work are formalized. To solve the multi-criteria problem, the following quality criteria were selected: The amount of energy expended, the time it took to complete the tasks, and the number of robots involved. For multi-criteria optimization, the resulting vector optimality criterion is linearly converged by introducing additional parameters characterizing group control: The total efficiency of a group of robots, the specific amount of energy for the operation of the support group and the energy for moving one robot to the given coordinates. To implement the ant algorithm, the problem was presented in the form of a set of undirected weighted graphs on which the "ants" will build solutions. An example is given that shows a detailed transition from the initial problem setting to graphs. The experiments were carried out on the basis of a software-implemented algorithm for finding the optimal group behavior strategy for a multi-criteria objective function with weight coefficients. For a group of three robots, experimental data were obtained and the results analysed, and a solution was modeled for several groups of robots with different weights.

show abstract

Section: Introductionmentioning

confidence: 99%