Immunity-Based Dynamic Reconfiguration of Mobile Robots in Unstructured Environments

• Published in: Journal of intelligent & robotic systems Vol. 96; no. 3-4; pp. 501 - 515
• Main Authors: Raza, Ali, Ali, Shujat, Akram, Maria
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.12.2019; Springer; Springer Nature B.V
• Subjects: Antibodies; Arbitration; Artificial Intelligence; Configurations; Control; Electrical Engineering; Engineering; Immunity; Immunology; Mechanical Engineering; Mechatronics; Reconfiguration; Robotics; Robotics industry; Robots; Unknown environments; Viral antibodies; Clonal selection; Artificial immune systems; Dynamic reconfiguration; Robot reconfiguration
• ISSN: 0921-0296; 1573-0409
• DOI: 10.1007/s10846-019-01000-6

Mobile robots have shown the capability to address difficult challenges ranging from disaster response to autonomous delivery of goods. However, the robots designed for one environment are often unable to perform effectively in other settings, and frequently require reconfiguration on software and/or hardware levels. Therefore, authors have designed a new immunity-based arbitration scheme to dynamically reconfigure mobile robots during their activity in unknown environments. The technique continuously selects the most suitable modules from its library e.g. sensor-configuration and navigation approach, on the basis of immunological mechanism of clonal selection. The technique successfully arbitrates different robot-configurations because of its ability to clone, mutate and select the suitable antibodies as well as to dynamically change the antibody concentrations. Moreover, maintenance of low inflammation levels alongside low collision count and slow battery drainage clearly illustrates the effectiveness of the presented technique.