Mobile Robot Localization Based on the Novel Leader-Based Bat Algorithm
| Autor: | Wolmar A. Neto, André L. M. Marcato, Daniel de Almeida Fernandes, Milena F. Pinto, Ivo C. Silva |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
0209 industrial biotechnology
Computer science Process (computing) Energy Engineering and Power Technology Particle swarm optimization Monte Carlo localization Mobile robot 02 engineering and technology Computer Science Applications 020901 industrial engineering & automation Control and Systems Engineering Position (vector) 0202 electrical engineering electronic engineering information engineering Robot 020201 artificial intelligence & image processing Electrical and Electronic Engineering Particle filter Algorithm Bat algorithm |
| Zdroj: | Journal of Control, Automation and Electrical Systems. 30:337-346 |
| ISSN: | 2195-3899 2195-3880 |
| DOI: | 10.1007/s40313-019-00453-2 |
| Popis: | This work proposes a new approach to the well-known method bat algorithm for solving the mobile robots global localization problem. The proposed method is leader-based bat algorithm (LBBA). The LBBA uses a small number of better micro-bats as leaders to influence the colony in the search for the best position, dealing satisfactorily with ambiguities during the localization process. The tests covered different scenarios aiming at comparing the proposed algorithm with other methods, such as the standard BA, the particle swarm optimization and particle filter. The results outperformed the compared methods, presenting a fast response and errors below the intended tolerance. The algorithm was tested in the robot kidnapping scenario and shows fast recovery in both simulation and in a real environment. In addition, the proposed technique showed 21% lower average error when compared with an algorithm that presents a variable quantity of particles, i.e. the adaptive Monte Carlo localization algorithm. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink