Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.741710
Title: Self-organising assembly using swarm robots
Author: Syafitri, Niken
ISNI:       0000 0004 7225 5037
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
When a swarm of small robots is required to operate in an unstructured and unknown environment, adaptability is required, particularly in the case of traversing a void (typically a crevasse or gap between two stable platforms). Adaptability is required when robots face a void of an unknown size so that they can safely cross. By applying adapt- ability, the robustness of the swarm can be maintained either by crossing a void or safely retreating after failing to cross. However, the relatively simple characteristics of swarm robots limit their performance in a complex task when they work individually, but it can be accomplished when they work collaboratively. Solving the problem arised, a strategy of self-organising assembly has been proposed supported by simulation and practical experiments. Algorithms with simple rules have been developed that allows the robot swarm to reach the target area from an initial zone separated by the void, where the size and location of the void are unknown. The configuration for the dynamic structure have been proposed built by the robot swarm to bridging two separated zones. To measure the performance of the swarm, metrics have been identified within the simulation environment. A number of 3D printed platforms have been designed and evaluated to investigate the emerging physical problems including the required docking system, platform shape, approaches to actuation and communication system. Several contributions have been achieved, in particular the approach that proposes a dynamic structure with simple rules that is constructed by a robotic swarm to cross the void, where the size and location of the void are unknown. This is in accordance with the concept of self-organise and self-assembly. The solution developed is scalable, with any number of robots can be added into the swarm. Any robot can also be the leader who initiates the construction or detects the target. When the size of the swarm does not satisfy with the size of the void, the swarm will retreat to the initial zone to prevent the robots lost. To support the rigid dynamic structure, a number of docking mechanisms are proposed.
Supervisor: Crowder, Richard Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.741710  DOI: Not available
Share: