Multiple odor source localization using diverse-PSO and group-based strategies in an unknown environment
Journal of Computational Science(2019)
摘要
This work presents a diverse particle swarm optimization based multi-robot cooperative approach for multiple odor source localization. Major contributions of this paper are with respect to group related tasks and plume following through the introduction of methods for group formation, maintenance of group aggregation, group closeness measurement, group dismantling, and next movement calculation of the robot. Specifically, this paper introduces methods for: (1) localizing multiple odor sources in parallel, (2) maintaining aggregation degree of a group by limiting the maximum number of robots a group can have, (3) measuring the closeness of the formed groups based on which group merging behavior is employed, (4) group dismantling to ensure better resource utilization, and (5) calculating the next move of a robot within the group by diverse- PSO. To bridge the gap between simulation and real-time experiments, sensor odometric error along with localization error in robot positioning is introduced, and the working of the proposed framework is evaluated. Contaminant release is simulated in the 3D indoor environment using Ansys Fluent. Performance of the proposed approach is compared with three state-of-art approaches considering the erroneous and error-free cases. Results validate the effectiveness of the proposed approach.
更多查看译文
关键词
Odor source localization,Diverse particle swarm optimization,Group formation,Limiting group size,Group merging,Group dismantling
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络