Język Polski
English
Русский
Slovenčina
Hrvatski
SWARM ALGORITHMS USING MICROPYTHON AND ULAB FOR MICROCONTROLLERS
Abstract
This study focuses on analyzing the possibilities of implementing the Particle Swarm Optimization (PSO) algorithm on microcontroller platforms, particularly on the PyBoard. Through rigorous testing and analysis, the behavior of the algorithm in resource-constrained environments was examined, aiming to understand its adaptability, efficiency, and stability. PSO code tests were conducted on three identical PyBoards, running the algorithm multiple times for various test functions. This comprehensive testing process allowed for capturing and evaluating the consistency of results across different microcontroller units. By repeating tests and gathering data on fitness values, execution times, and optimal solutions, valuable insights into the algorithm's behavior were obtained. The analysis results revealed several significant findings. For functions with simpler optimization landscapes, all devices achieved high effectiveness in finding the global optimum. In the case of more complex functions, the results obtained on microcontrollers were close to optimal values, suggesting good adaptability of the PSO algorithm. The study confirmed that microcontrollers like the PyBoard can be effective tools for analyzing and comparing the capabilities of optimization algorithms, despite their limited resources. In conclusion, the research results provided valuable information on the performance of the PSO algorithm on microcontroller platforms, opening new perspectives for embedded system design.
Keywords:
particle swarm optimization, microcontroller platforms, pyboard, ulab, resource-constrained environments, adaptability, efficiency, stability, test functions, fitness values, execution times, optimal solutions, optimization landscapes, embedded systems, performance analysisDetails
References
Statistics
Authors
Download files
Citation rules
Licence
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.