posted on 2022-12-16, 15:06authored byMichael Johnson
This thesis presents the design and development of a low-cost testbed for distributed
mobile robotic research, its validation in both the academic research and educational
spaces and the application of this development platform to the emerging eld of
string stability in order to propose an a ordable research tool for localisation, education,
path following and automatic convoy formation research and development.
Despite the increase in available robotic platforms and associated enabling technologies
such as localisation and communication, there exists a requirement for
an a ordable generic testbed that will facilitate entry into the distributed mobile
robotic research space. This thesis addresses this shortcoming by presenting the design
and validation of a distributed mobile robotic testbed which facilitates research
and development.
The development and realisation of the closed-loop mobile robotic testbed, complete
with
oor area, localisation and communications support, is presented. Validation
of the testbed is a two-fold process. First performance capability tests in relation
to trajectory following, path planning, localisation and real time distributed communication
overhead costs illustrate the superiority of a specialised solution such as
the MiaBot Pro robotic platform over a generic low-cost solution such as the LEGO
Mindstorm NXT robotic alternative.
Secondly, the testbed is used to address the current lack of a low-cost empirical
solution for researchers in the eld of string stability. Research in this area is still
largely based on simulation results due to the prohibitive implementation costs of
a real-world test system for string stability. This thesis proposes the adaptation of
testbeds such as the low-cost mobile robotic testbed in the Wireless Access Research
Centre at the University of Limerick for this purpose, demonstrates the feasibility of
such an undertaking and validates results that compare favourably with simulation
only based studies that have thus far dominated the literature.
Finally, the proposed testbed is used to achieve certain well de ned learning
outcomes across a wide spectrum of educational curriculae. This ranges from second
level ICT through a series of honours undergraduate and taught postgraduate
electronic engineering and robotics courses. The use of the testbed in the investigation
of teaching pedagogies such as Problem-Based Learning (PBL) and their
applicability to electronic engineering courses is also considered in the course of this
work.
Funding
A new method for transforming data to normality with application to density estimation