Institutional-Repository, University of Moratuwa
Welcome to the University of Moratuwa Digital Repository, which houses postgraduate theses and dissertations, research articles presented at conferences by faculties and departments, university-published journal articles and research publications authored by academic staff. This online repository stores, preserves and distributes the University's scholarly work. This service allows University members to share their research with a larger audience.
![]() Research Publications | ![]() Thesis & Dissertation | ![]() E- Books |



![]() UoM Journal Publications | ![]() UoM Conference Proceedings | ![]() Articles published in Scimago's Q1 journals | ![]() UoM Research Reports | ![]() Other Articles authored by UoM staff |
Recent Submissions
item: Conference-Full-text
Proposing a Novel Grading Method for Women’s Wear Upper Torso Contour Garments
(Department of Textile and Apparel Engineering University of Moratuwa, 2024) Balasooriya, L.; Withanaarachchi, S; De Silva; Nandasiri, G.K.
Women’s wear body contour garments are critical in fit
and construction due to the considerable number of body
curves involved. Fit of the women’s upper torso contour
garments is affected by the shoulder angle, armscye, sleeve
crown, bust and necklines [1], [2]. It becomes more
complicated with fabric stretch as fabrics incorporate stretch
as a part of their fit and function [3].
Patternmaking theory plays a vital role in achieving
perfect fit. Flat pattern creation and fabric draping are the
most common methods of patternmaking [4]. Pattern grading
is also important in achieving clothing fit. Pattern grading is
the process of adjusting a pattern to fit within a range of sizes
[3]. An accurate pattern grading method is necessary to
ensure accurate garment fit across the sizes. Although there
are established pattern grading methods for some garments
produced using rigid fabrics, no such grading methods are
published or practiced in the industry for garments produced
using stretch fabrics.
Incorporation of fabric stretch during grading needs to be
considered during to have a proper fit. According to Moore
et al. [3], reducing ease and darts are necessary to obtain the
perfect fit with the stretch fabrics. With the different fabric
stretch percentages, it is difficult to determine a common
grading rule for each fabric. However, the industry demands
a precise guide for stretch-knit pattern grading, to minimize
existing problems relevant to the current trial-and-error
method.
Establishment of a grading method for stretch-knit pattern
grading is useful for the ready-to-wear industry, and this tacit
knowledge can be applied in developing an automated pattern
development system that can be used in customized clothing
production in future. This study aims to examine the grading
values of contoured patterns in developing the size set of the
women’s upper torso garments and develop a relationship
between fabric stretch and pattern grading values for the same
product type.
This study will examine pattern grading values of
women's contour garments, analyze pattern curve shapes and
body landmarks, and develop a relationship between fabric
stretch and grading values of women's upper torso garments.
The proposed method will be validated both digitally and
physically.
item: Conference-Full-text
Development of 3D Knitted Fabrics for Triboelectric Energy Generation
(Department of Textile and Apparel Engineering, University of Moratuwa, 2024) Dissanayaka, D.M.V.U; Madhuranga, R.S; Lanarolle, W.D.G.; Nandasiri,, G.K.
The demand for sustainable energy solutions has been driving
the exploration of alternative energy generation technologies.
Triboelectric nanogenerators (TENGs) are one such
promising solution, capable of converting mechanical energy
into electrical energy by exploiting[1] the triboelectric effect.
TENGs offer a clean, renewable source of energy by
harnessing movements, vibrations, and friction from human
activity. With the increasing popularity of wearable
technology, integrating energy-harvesting fabrics into
clothing presents a significant opportunity to create selfpowered
systems. [2]
This study focuses on developing 3D knitted fabrics
specifically designed for triboelectric energy generation.
Traditional fabrics are limited in their surface area and
mechanical properties, leading to suboptimal energy
harvesting efficiency. By utilizing advanced knitting
techniques to create three-dimensional structures, the
potential for maximizing the triboelectric effect is enhanced.
The primary goal of this project is to develop durable, flexible,
and efficient 3D knitted fabrics that can power low-energy
electronics, such as sensors, through human movement.[3]
item: Thesis-Abstract
Development of an energy efficiency rating system for building lighting systems
(2023) Randimali, HDS; Rodrigo, AS
Lighting systems consumes more than 10-20% of total energy consumption of the buildings. Lighting systems significantly affect for the energy efficiency of the building. The energy efficiency rating systems were developed by various countries including Sri Lanka to enhance the energy efficiency in buildings. This research is aimed at developing a new rating system by identifying the existing rating systems and investigating the criteria stated to rate the efficiency of lighting systems and shortcomings and issues.
To implement a new performance rating system, the studies were carried out modeling the lighting systems using DIALux evo software. A perfect model was developed optimizing the luminaire and building envelope parameters using the software simulations for an office area. The ratings of the optimum model normalized to develop the energy score between 0-100. The lighting system efficiency rate proposed in this thesis can be used to rate the lighting systems of buildings with various functional requirements.
item: Thesis-Abstract
Autonomous navigation of mobile robot in a dynamic environment using deep reinforcement learning
(2023) Seetharaman, D; Pathirana, CD
Autonomous navigation of a mobile robot in a dynamic environment is a highly challenging application because the path to the goal frequently changes due to unpredictable movements of humans with different velocities. Deep Reinforcement completely trains a model in a simulator using a trial-and-error technique by exploring and collecting required data automatically and cheaply from the customized environment. This research develops a customized environment with robot and pedestrian models in OpenAI Gym, replicating humans' real-world measurements and motion patterns. A mathematical model was developed to encapsulate the navigation norms of humans' to teach the robot about socially compliant routes using a reward function in order to smooth the robot’s navigation in a dynamic environment. A recently evolved algorithm, H-PPO, has been selected to train the model by considering the agent's hybrid action space, which consists of discrete actions parametrized by continuous values.
First, the model failed to learn due over fit in a simple environment, and then it learned when the task was randomized. The various approaches have been investigated to enhance the model's generalizability as much as possible in the simulator. Finally, the agent is trained in each environment separately. Despite this research has not considered the complex scenario as randomizing the whole environment initially, the developed model was able to scrutinize the performance of the recently evolved algorithm H-PPO in obstacle avoidance applications, and the developed model can learn obstacle avoidance in a dynamic environment by respecting social norms in the long-range motion for laboratory application. However, the success rate of the model trained later in a fully randomized 3-pedestrians environment was 86.67% out of 30 episodes of testing, which is higher than the previous research [1]. Further investigation has to be carried out in future work by adding memory ability to the model in order to enhance the performance, reduce the training time and mitigate the performance collapse during the learning phase.
item: Conference-Full-text
Few-shot multispectral segmentation with representations generated by reinforcement learning
(The British Machine Vision Association, 2024) Jayakody, D; Ambegoda, T
The task of segmentation of multispectral images, which are images with numerous channels or bands, each capturing a specific range of wavelengths of electromagnetic radiation, has been previously explored in contexts with large amounts of labeled data. However, these models tend not to generalize well to datasets of smaller size. In this paper, we propose a novel approach for improving few-shot segmentation performance on multispectral images using reinforcement learning to generate representations. These representations are generated as mathematical expressions between channels and are tailored to the specific class being segmented. Our methodology involves training an agent to identify the most informative expressions using a small dataset, which can include as few as a single labeled sample, updating the dataset using these expressions, and then using the updated dataset to perform segmentation. Due to the limited length of the expressions, the model receives useful representations without any added risk of overfitting. We evaluate the effectiveness of our approach on samples of several multispectral datasets and demonstrate its effectiveness in boosting the performance of segmentation algorithms in few-shot contexts. The code is available at https://github.com/dilithjay/IndexRLSeg.