Investigating Multiple Pheromones in Swarm Robots - A Case Study of Multi-Robot Deployment

Liu, Tian, Sun, Xuelong, Hu, Cheng , Fu, Qinbing, Isakhani, Hamid and Yue, Shigang (2020) Investigating Multiple Pheromones in Swarm Robots - A Case Study of Multi-Robot Deployment. In: 2020 5th International Conference on Advanced Robotics and Mechatronics (ICARM).

Full content URL: https://doi.org/10.1109/ICARM49381.2020.9195311

Documents
Investigating Multiple Pheromones in Swarm Robots - A Case Study of Multi-Robot Deployment
Authors' Accepted Manuscript
[img]
[Download]
[img]
Preview
PDF
Tian 2020 Investigating Multiple Pheromones in Swarm Robots - A Case Study of Multi-Robot Deployment.pdf - Whole Document

3MB
Item Type:Conference or Workshop contribution (Presentation)
Item Status:Live Archive

Abstract

Social insects are known as the experts in handling complex task in a collective smart way although their small brains contain only limited computation resources and sensory information. It is believed that pheromones play a vital role in shaping social insects' collective behaviours. One of the key points underlying the stigmergy is the combination of different pheromones in a specific task. In the swarm intelligence field, pheromone inspired studies usually focus one single pheromone at a time, so it is not clear how effectively multiple pheromones could be employed for a collective strategy in the real physical world. In this study, we investigate multiple pheromone-based deployment strategy for swarm robots inspired by social insects. The proposed deployment strategy uses two kinds of artificial pheromones; the attractive and the repellent pheromone that enables micro robots to be distributed in desired positions with high efficiency. The strategy is assessed systematically by both simulation and real robot experiments using a novel artificial pheromone platform ColCOSΦ. Results from the simulation and real robot experiments both demonstrate the effectiveness of the proposed strategy and reveal the role of multiple pheromones. The feasibility of the ColCOSΦ platform, and its potential for further robotic research on multiple pheromones are also verified. Our study of using different pheromones for one collective swarm robotics task may help or inspire biologists in real insects' research.

Keywords:Robot sensing systems, Robot kinematics, Task analysis, Insects, Mechatronics, Collision avoidance
Subjects:H Engineering > H671 Robotics
Divisions:College of Science > School of Computer Science
ID Code:43680
Deposited On:21 Jan 2021 10:54

Repository Staff Only: item control page