Shepherding: An Immune-Inspired Robotics Approach

• Main Authors: Razali, Sazalinsyah, Meng, Qinggang, Yang, Shuang-Hua
• Format: Conference or Workshop Item
• Language: English
• Published: 2010
• Subjects: QA75 Electronic computers. Computer science; QR180 Immunology
• Online Access: http://eprints.utem.edu.my/id/eprint/4876/1/razali2010-UMIES-shepherding.pdf; http://eprints.utem.edu.my/id/eprint/4876/

In this research, basic biological immune systems and their responses to external elements to maintain an organism’s health state are used as an inspiration for solving a specific multi-robot system problem. The proposed algorithm is based on immune network theories that have many similarities with the multi-robot systems domain. The approach uses a memory-based immune network that enhances a robot’s action-selection process and can obtain an overall a quick group response. The algorithm is applied on the Player/Stage simulation platform and evaluated using the sheepdog scenario. The project proceeds to investigate the low-level shepherding behaviour, specifically the shepherds' Line Formation.