Research on Indoor Robot Localization Method Based on Clustering Optimizes Genetic Algorithm

To reduce the operation time of the visible light positioning system, a visible light 3D positioning system based on clustering optimization genetic algorithm (COGA) is proposed. This method uses genetic algorithm (GA) to assess the fitness of each individual of the population, then applies density...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2023 2nd International Conference on Artificial Intelligence and Intelligent Information Processing (AIIIP) s. 153 - 158
Hlavní autoři: Shi, Yuanyuan, Qin, Ling, Zhao, Desheng, Xu, Yanhong
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 27.10.2023
Témata:
Clustering algorithms
density based noise application spatial clustering algorithm
genetic algorithm
Genetic algorithms
Location awareness
photodiode
Robot localization
Sociology
Statistics
Three-dimensional displays
trilateral positioning
visible light localization
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:To reduce the operation time of the visible light positioning system, a visible light 3D positioning system based on clustering optimization genetic algorithm (COGA) is proposed. This method uses genetic algorithm (GA) to assess the fitness of each individual of the population, then applies density based noise application spatial clustering algorithm (DBSCAN) to analyze the population structure and then adjust the population structure. By combining these two algorithms, we can search for the optimal solution on a global scale and adjust it on a local scale, integrated use of global and local information. Simulation results show that the location speed of single point is 43ms, which improves the real-time performance of the positioning system. In a simulation space of 3\times 3\times 3.6\mathrm{m}^{3} , the average positioning error of this method is about 3.8cm, which achieves the precision of centimeter positioning and satisfies the need of indoor robot positioning.
DOI:10.1109/AIIIP61647.2023.00035