ABCDEFGHIJKLMNOPQRSTUVWX
1
Proposed TitleProject DescriptionProject ObjectiveHardware and Software Requirements
Course(s) Recommended to be taken
2
Development of a robust modular ground control station for fleet of Unmanned Surface VesselsTo develop a robust modular ground control station for fleet of Unmanned Surface VesselsGround Control Station for USV-
3
Development of fleet managment system for Unmanned Surface VesselsTo develop fleet managment system for Unmanned Surface VesselsGround Control Computer for USVs-
4
Development of a collaborative robotics system for teaching and learningCollaborative task between two robotic system. One SCARA and one 3DoF articulated hand robot are available. The student need to devlop proper operational code to operate these robots so that a task can be achieved using these two robots. Such as putting the cap of a bottle. 1. To develop the motion algorithm for the two robots to accomplish the collaborative task.
2. To evaluate the performance of the developed algorithm.
Micro controller coding, GUI developemnt. Robotics (but not a must)
5
IoT based data acquisition system development for river water quality monitoring. Various sensor will be integrated (temperature sensor, turbidity sensor etc ) to monitor the river water quality of the IIUM river. The fused (sensor) data need to be transmitted thru IoT to the cloud server. 1. Devlop data acquisition system for the sensors.
2. Develop node-red interface for data transmission through IoT.
3. Develop decision making algorithm using cloud based data analytic.
Micro controller, sensor interfaceMicrocontroller
6
Development of interactive turn by turn direction in Kulliyyah of EngineeringBasically student will use AR to create something like satnav at car to locate in KoE without using GPS.Develop turn by turn augmented reality for map direction in KoEsmartphone after 2019, unityProgramming
7
Gesture Recognition System using HRI Platform in Enhancing Imitation Skills1) To embed a dedicated educational module (readily developed by a PhD student) into social robots
2) To develop a gesture recognition system through bodily movements of the subjects
3) To validate the efficacy of the recognition system through observation (frame by frame)
Python, OpenCV, MediaPipe, Laptop (at least i5 microprocessor)

Currently available at BioMCT lab: RGB camera, QT robot
Mechatronics Workshop, Robotics
(Optional) Intelligent Control, Machine Vision
8
The Development of Stress Detection System using Biosignals on SASMEC Patients (in collaboration with Dr Zhafri from KOM, IIUM)The student will be required to travel to SASMEC, Kuantan for data collections (perhaps for 1 week)1) To enhance the stress detection system (previously developed by FYP students) by integrating several biosignals datasets
2) To classify the state of subjects' mental health through a developed GUI
3) To validate the developed system using questionnaire and professional medical practitioner
A laptop with installed Arduino, (will be provided) GSR sensor, piezoelectric sensor, temperature sensorElectric Circuit, Programming (Arduino)
9
Deep Learning Based Portable Dental Caries Detection Device for Increasing Oral Health AwarenessThis project is an application of Deep learning in dentistry. The goal is to build a model that assists in diagnosing teeth plaque. Having a tool that can identify the condition of someone’s teeth by simply pointing a camera into his/her mouth enables better awareness of oral health although such tools are no substitute for professional medical advice. A Deep
Learning model using Neural Network has previously been programmed, trained, and tested to provide the ability to distinguish between healthy teeth and dental plaques which are known as a cause of many other oral diseases. By collecting over 600 images of healthy and unhealthy teeth from google images models have been trained to perform classification and were implemented on Raspberry pi. In this project the student will further explore the usage of Convolutional Neural Networks and also conduct further data collection in collaboration with IIUM Kulliyah of Dentistry.
1) Develop Machine Learning Models for Detection of Dental Caries.
2) Implementation of Trained Models on a Single Board Computer.
3) Conduct study on Oral Health of Selected Subjects
Raspberry Pi.Interfacing Lab
10
Mapping of deep trenches and small objects ahead of a visually impaired person using ultrasonic sensors array for blind navigation. (Co Dr Affendy)Visually disabled peoples most of the time fail to detect small objects or deep trench in front of them during walking which lead to accidents. This project is meant for mapping sudden deep trenches below the level of the plain ground or small objects above the ground using an ultrasonic sensors array attached to a moving visually disabled person. It will require ultrasonic imaging and feature recognition for the solution to the problem. 1. To develop a mathematical model for mobile ultrasonic sensors array for mapping deep trenches and small objects below and above the ground level.
2. To develop a prototype of the ultrasonic sensors arry
2. To evaluate the performance of the developed system
Ultrasonics sensors array, microcontroller, Matlab or similar software, mechanical hardware.Signal processing, Image processing, Autonomous robotics systems..
11
Development of Tracking System for Surveillance using high speed target camera. Motion tracker has been the fundamental principle of surveillance camera. However, blind motion tracking triggers false alarm quite frequently and it is a waste of memory. Thus, there is an impending need to develop a fast motion tracking that detect specific object and ignore all others or vice versa. To develop this, a fast object detection model (like the YOLO series) is necessary.To design an object detection model that captures a specific object in high-speed motion.
To be able to simulate the object tracking and follow its path.
To analyse the performance of the model in terms of FPS and precision of the classified object.
Camera & PC-
12
Development of shape memory alloy (SMA) based artificial muscle in lower limb exoskeleton for rehabilitationActuators used in exoskeleton are often bulky such as motor, pneumatic-type etc. Shape memory alloy actuator has the advantage of having simple structure, lightweight and requires only heat (from current supply) to actuate, but suffers from low actuating force and slow response. In this project, student will develop a SMA based artificial muscle using different wire diameters and spring configurations bundled in a fabric, and measuring its capability to lift weight and response time. Upon having the best muscle design, a PID controller for the artificial muscle is developed to control its position/displacement. Wire encoder is used to measure the muscle contraction and current is the input to the system. Finally, this fabric muscle will be attached to a simple knee brace/exoskeleton to show proof of concept .1) To investigate response time and force-lifting capability of shape memory alloy (SMA) artificial muscle using different wire diameter and spring configuration
2) To develop PID controller of the artificial muscle to control its position/displacement
3) To test the performance of the SMA artificial muscle
shape memory alloy, Labview, NI acquisition system (all provided)None
13
Development of Vehicle Simulator System for the training of driversThe purpose of this project was to research, design and implement motion cues for the 2/3-DOF motion platform. This motion platform is able to provide translational motion along the z-axis (Heave), rotational motion about the x-axis (Roll) and rotational motion about the y-axis (Pitch). The research aimed to create the best possible fidelity in the vehicle simulator system by creating realistic motion cues that work in cohesion with visual cues.The research objectives are as follows:
1) To investigate, design, implement and test the classical algorithm for use in 2/3-DOF motion platform
2) To evaluate the position control system performance to provide motion cues against the performance of the Matlab/Simulink simulation system results.
2/3D car simulator platform & PCMCT workshop
14
Normalization of Upper Trapezius sEMG Signal upon Stress InductionInvestigation on electromyograph (EMG) signal to identify differences in the activation levels and patterns between subjects with or without shoulder neck pain during stress with the aim of understanding the cause of the pain is paramount to develop an improved training program to treat the pain. Trapezius muscle is one of the most important stabilizers of the shoulder and any alteration in the activation level captured in the EMG signal is always associated with musculoskeletal pain around the area.
Therefore, a good normalization method on the signal is important to reduce any susceptible influences caused by electrode placement, skin resistance and type of muscle contraction. The study will look at different methods to do the signal normalization for the purpose to standardize and reproduce muscle contraction procedure. (co supervised by Dr Aimi dan Jazlan)
To reproduce a standardized, normalized EMG signal reading from trapezius muscleEMG signal DAQ, electrode patch and MatlabDSP
15
Development of a haptic feedback system to assist blind and visually impaired (BVI) students in tactile graphics readingBlind and visually impaired (BVI) students require instruction and guidance to gain
information from tactile graphics. Without any teachers or instructors will lead them into
frustration and confusion to work with tactile graphics. Facilitating the exploration of tactile
graphics can be in many ways such as moving their hand, tracing along paths, or verbal
instructions. There are also assistive devices that use audio feedback to convey information
about the tactile graphics when hand and finger move and touch specific areas of interest.
Nevertheless, the haptic feedback method can be a complement to the audio feedback method
and act as a replacement for the sighted individual to guide BVI students in exploring tactile
graphics. Therefore, this current project aims to explore the potential of a haptic feedback
system to support reading and exploring tactile graphics. It can be a teaching and learning
tool for teachers in special education schools for the blind, resulting in more accessibility to
tactile graphics for BVI people.
1. To design a haptic feedback device for a single hand that can be remotely controlled
2. To design a haptic guidance strategy for tactile graphics reading
3. To evaluate the reading performance of the BVI students with and without the haptic
feedback device
a. Arduino
b. Vibrating Mini Motor Disc / Micro motor or actuator
c. Suitable electronic components
Instrumentation and Measurement, DSD and Microprocessors
16
Intention Detection for Lower-Limb Exoskeleton Control in RehabilitationThis project is initiated as a result to the discussion with an Orthopedic specialist in IIUM Kuantan, that current rehabilitation process for patients who just undergo Total Knee Replacement (TKR) operation is very manual. Thus, a study to improve current system is deemed necessary.
The student is expected to detect motion intention (the signal before the motion is made or very early stage of movement) via EMG sensor signal and feed the output to controller to move the motor.
Hardware: Leg Exoskeleton prototype equipped with motor
Software: Pattern recognition and classification
Control : Position control

Co-SV : Dr Azni Nabela
To develop a control system to move lower-limb exoskeleton based on motion intention signal for rehabilitation purposes. Hardware: 3D printer, controller, motor, signal conditioning circuit.
Software: any software to analyze signal and can do classification (e.g. Matlab, phyton, scilab)
Control System, Signal processing
17
Sensor development for identification of acetone vapour for detecting diabetesCurrently, prick method to measure glucose is widely use and acceptable around the world. However, it become trauma especially to the senior citizen and children. We intend to explore this technology to measure from the actual acetone and from exhaled breath of diabetes patient. 1- Design and simulation of the sensor using finite element method.
2- Fabrication of the sensor and prepare experimental setup.
3- Performance evaluation of the modeling and fabricated sensor.
Gas sensor, acetone Interfacing
18
Real Time Object Detection of Dental Instruments for Inventory ManagementThis project involves extracting images from a series of videos of dental instruments to form a dataset. Object detection using methods such as CNN and YOLO will be used to detect these dental tools on top of existing pretrained models. This research will explore the possibility of incorporating automation in a dental clinic practice. Validation of the developed object detection will be performed in a dental clinic.1) To preprocess videos of dental instruments and build the required dataset.
2) To explore object detection algorithms for recognizing various dental tools.
3) To validate the proposed method in a real clinical setting.
Google Colab (Jupyter Notebook), Raspberry pi, any suitable camera.Machine Vision
19
Lightweight Deep Learning Model for Resource Constrained Robot NavigationObstacle avoidance and place recognition are key capabilities of an autonomous robots such as unmanned surface vessel (USV) and mobile robots. Navigation task of robots becomes increasing challenging when deployed in resource constraint environment with limited access to IoT-capable network. In such situation, robots need to rely on available on-board embedded computing to perform recognition tasks. This project will explore techniques to develop a lightweight recognition model such that the model can perform recognition with plausible accuracy and speed while being deployed in a small computing platform. The model will then be evaluated on autonomous robots such as delivery robot and USV.1. To develop techniques to convert existing deep learning (DL) model to a lightweight version based on quantization method
2. To train the lightweight DL model for autonomous robot navigation
3. To evaluate the performance of the lightweight DL model on autonomous robot navigation
1. Laptop
2. Raspberry pi (available in CUTe)
3. Rasberry pi cam/ webcam (available)
4. AI server (available)
4. A robot platform (available)
Intelligent control
20
3D Face Transformer for Authentication SystemRecently, vision transformer has shown promising performance for 2D image understanding compared to other machine learning models. However, the application in 3D recognition has been relatively limited, especially in face authentication domain. This project tends to explore the feasibility of the use of vision transformer in solving 3D face recognition task. The 3D face model will then be deployed in a face authentication pipeline.1. To develop a vision transformer (ViT)-based model for 3D face recognition
2. To compare the performance of the 3D Face ViT with state-of-the-art CNN-based 3D face models.
3. To incorporate the 3D Face ViT model in a face authentication pipeline on Jetson Nano
1. Laptop
2. Zed stereo camera (available)
3. Jetson Nano (available)
Intelligent control
21
E-nose for rapid pesticide detection Use of pesticides has been linked to neurobehavioral deficits among exposed workers. In Malaysia, organophosphate and pyrethroid pesticides are commonly used to control mosquito-borne diseases. Currently, there is no suitable method to measure the gas onsite and real time exposed to the worker. We are developing an e-nose to mitigate this issues. (Co with Dr Hasmawati)1-To construct a portable and handheld device that can detect the presence of pesticides.
2-To evaluate the performance of the electronic nose development.
3-To apply Machine Learning algorithm to successfully train a developed portable
device.
Gas sensor, Arduino, LCD Display, Buzzer Interfacing lab
22
Stress Level Detection Using a Piezoelectric-Based Sensory SystemPrevious project has successfully identified the difference between Relax and Stress condition using Pulse Rate acquired via piezoelectric sensors.
For this project, student is expected to extend the research by developing an algorithm to identify the different level of stress using the same piezoelectric-based sensory system.
To develop classification algorithm to identify different stress levels based on Pulse Rate (PR) acquired using piezoelectric-based sensory system. Arduino microcontroller, WEKA software, excel, SPSSSignal processing
23
Analysis of Photocurrent from Photoelectrochemical Cell (PEC)-based Resistive Sensor for Water Quality Monitoring System Real time electrochemical (EC) detection for water quality monitoring faces challenges because of the need for self-powered and highly sensitive sensor for wide area inspection as well as portable data acquisition system for rapid analysis and prediction. Therefore, fundamental studies on how reliable the developed sensor and monitoring mechanism reacts to the most common pollutant in water streams are essentially needed. 1. To design and fabricate a resistive sensor.
2. To design the water quality detection system using the prepared sensor.
3. To analyze the photocurrent generation from the developed sensor with the reaction of water contaminant in the detection system
raspberry pi and python -
24
Radar Scene Generator for FMCW Marine RadarDevelopment and tests of a marine radar takes a very long time. Radar scene generator can be used to shorten this long development time. By using this scene generator we can simulate many real signal conditions captured by the radar scanner. In this final year project the student must develop an algorithm for generating a series of radar echoes. The developed algorithm will be implemented and tested on a real marine radar system.(1) To develop algorithm for generating radar echoes (2) To test the performance of the developed algorithm on a real systemMATLAB/SciLab/GNU-Octave, C/C++ compiler, Quantum2 marine radar(1) Signal and Systems, (2) Programming skills (3) Interfacing Lab
25
Development of PC-based Plan Position Indicator for Marine RadarA plan position indicator (PPI) is a type of radar display that represents the radar antenna in the center of the display, with the distance from it and height above ground drawn as concentric circles. As the radar antenna rotates, a radial trace on the PPI sweeps in unison with it about the center point. It is the most common type of radar display. In this project the student will develop a PC-based PPI for a marine radar that can be integrated with the real system.1. To develop the software-based PPI, 2. To integrate the developed PPI in the real marine radar, 3. To evaluate the performance of the developed PPIGNU-Octave/MATLAB/SciLab, C/C++ compiler, Quantum2 radarInterfacing Lab, C/C++ programming skills, Signals and Systems
26
Identification of the correct Tajweed in a Quranic Words.The work will start with collecting suitable database from the expert that represents the correct pronunciation of a letter. The suitable features combination will be extracted from each of the letter and later classification of the letters will be conducted. The task will end with classification of the group of articulation point and characteristics using new testing data.This project aims to identify the correctness of the tajweed that consist of makhraj and sifaat in a word of Quran. Signal processing, AIAI, Machine Learning
27
Virtual Reality-based Indoor Mapping Robot for Disabled StudentsStudent will explore the use of virtual reality-based cooperative and communicator mechanism which could be used among the disabled students such as visually impaired, which is called S-CORT. S-CORT is an assistive robot equipped with multiple sensors which are ultrasonic sensor, line following sensor and gyrometer sensor. The body of the S-CORT is designed with modular connection which is for detachable sensors or equipments. S-CORT is easily programmed using Arduino IDE and the holonomic X-Drive motor configuration provide a high mobility for S-CORT movement. 1. To formulate an efficient VR-based indoor mapping strategy for disabled students.
2. To develop the convenient, cheap and compact robot with capability to robustly detect and predict indoor movements and interactions.
3. To analyse both hardware and VR mapping strategy and perform simulation for evaluation.
1.hardware components needed
-Arduino Nano
-Gyro Sensor
-Motor Driver
-High Torque DC Motor

2.software used
-Arduino IDE
-MBlock
Intelligent System
28
IIUM: Smart IoT Energy Optimisation and Localisation Monitoring for E-Bike Sharing
1. E-bike will be equipped with GPS and communication module to collect location and speed data, as well as an elevation profile of the chosen route from Google Earth [5].
2. The user can monitor energy availability of e-bikes at current locations to travel the desired distance and two cases are considered:
a. When e-bikes are parked, the user can check their status before picking them up and predict their reliability to travel specific distances based on state of charge (SoC).
b. While moving, the elevation profile information will be used to control the maximum motor power according to geographical characteristics, battery level and physical strain of the user in order to optimize the e-bike energy and travel distance.

To design a communication module strategy for location and speed data collection for e-bike sharing
To develop an IOT-based Energy optimization for E-bike sharing localization monitoring
To evaluate energy profile information with travel distance optimisation
1.hardware components needed
-GPS and LORAWAN
-Arduino Nano
-Gyro Sensor
-Motor Driver
-High Torque DC Motor

2.software used
-Arduino IDE
-MBlock
Intelligent System
29
Seismic response control using electromagnetic vibration absorberIn this work, vibration absorbers using electromagnets will be designed. The absorber will then be applied to a prototype model of a three-storey building to reduce earthquake response.To design electromagnetic vibration absorbers to control vibrations of structures subjected to earthquakesNI DAQ device-
30
Vibration control of a suspension system using magnetorheological fluid damperin this work, To evaluate the performance of a single-degree-of-freedom suspension system with an MR fluid damper NI DAQ device-
31
Shape memory alloy based artificial muscle for application in soft gripperThis project will improve the SMA soft gripper developed from last time by integrating shape memory polymer (SMP) to create variable stiffness at the hinges of the gripper. This will create a more compliant gripping mechanism as we all as reversible deformation.1. To investigate different SMA-SMP finger configuration which can produce the largest bending deformation
2. To fabricate the SMA-SMP soft gripper complete with actuation system.
3. To test the performance of the SMA-SMP soft gripper in terms of gripping compliance of various objects in different orientation.
shape memory alloy, shape memory polymer, 3D printer, CAD, ArduinoNone
32
Design and development of tethered wall-climbing robotTo design and develop a tethered wall-climbing robot with circuit and wiring optimization1. To design, develop and revise the developed tethered wall-climbing robot
2. To optimize the circuit and wiring of the robot
3. To analyze the performance of the developed robot
raspberry pi, arduino, python etc.Interfacing lab
33
Development of haptic and audio tactile globe for blind and visually impaired people
Tactile graphics is a notoriously difficult area in the education of congenitally blind children, and yet, since so much information vital for the correct functioning in this, predominantly sighted world, is available as two-dimensional visual displays, tackling the problem of making this information accessible to blind people becomes an urgent task. Tactile globe is a great aid for people who are blind or partially sighted. Up to date there are only high contrast raised surfaces identifying regions on the globe. The aim of this research is to make 'Globe Learning' more interactive and informativeTo investigate the accessibility of BVI students to globe in special education for the blind

To design a tangible globe embedded with assitive system that involve haptic and audio feedback features for BVI students

To evaluate the proposed design through a pilot study among BVI students and special education teachers for the blind
Microcontroller
Sensors and Actuators
Interfacing Lab
Instrumentation and Measurement
DLD and MicroP
34
Local active noise cancellation in car cabinOne of the noise source inside a moving car is from the engine. This project will develop an active noise cancellation system to reduce noise perceived by the driver. This can be done by installing a small set of speakers at the headrest location (i.e. near driver's ears) and generating a sound wave which is 180 degree of phase of the engine noise, with the help of set of microphones and a microcontroller. 1) deriving a mathematical model of the active noise cancellation system.
2) to verify the performance of rhe proposed noise cancellation method experimentally.
Matlab/scilab, labview, Single Board/ compact RIO, micrpohone, loudspeakers etc. Control system1, DSP.
35
Vibration supression of multistorey building using adaptive tuned mass damperThe project is aimed to control vibration of the multi storey building during seismic vibration such as earthquake. A small prototype model of the proposed project will be developed at the end of the course.1) to derive a mathematical/ simulation model of multistorey building with adaptive tuned mass damper.
2) to verify the proposed model experimentally.
Matlab/scilab, labview/arduino IDE, arduino microcontroller/SBRio, dc motor, accelerometer, tuned mass damper. EtcControl system 1, DSP, system dynamics/modelling
36
Smart Urban Ageing Building (roof) investigationThe project investigates the different maintenance requirements for an urban building and propose different technique in improving the maintenance of an ageing urban building.1. Investigate different maintenance requirement in ageing urban building
2. Propose maintenance management for urban building
3. Build smart management system for the urban building
NASoft computing
37
Low Resolution Image AI Traffic Flow Control At Ramp and InterchangeThe project propose a traffic flow control at ramp and interchange using AI. The project propose that a low resolution image is used to identify different vehicle types amd speeed during travel, and their predicted flow to ramp and interchange.1. Develop dataset for different vehicle and speed
2. Train suitable AI backbone network for traffic flow control
3. Validate the trained network
NASot computing
38
Upper Limb Tele- Rehabilitation for Pediatric Patients with Cognitive ImpairmentsThis study focuses on the development of an upper limb tele-rehabilitation system for pediatric patients with cognitive impairments. The system is targeted to be implemented as an online/ tele-rehabilitation treatment where the pediatric patients can perform the exercises at home and their progress can be monitored by the doctors in the hospitals. This reduces the time and save the parents from coming to the hospitals regularly as most of them have work commitment and need to take care of other children. Sensors will be used to measure the patient's movement, a related game will be provided for the children's treatment, IOT set up for the system will be established and machine learning algorithm will be used to measure the patients' progress. This project is based on the request from a doctor in IIUM Hospital in Kuantan. It is targeted that the project will contribute towards the improvement of rehabilitation treatment of the pediatric patients.1. To develop the data collection set up for the upper limb rehabilitation system for pediatric patients with cognitive impairments.
2. To develop the IOT settings for the upper limb rehabilitation system for pediatric patients with cognitive impairments.
3. To develop the Machine Learning algorithm in recognizing the pediatric patients' recovery level.
4. To verify the system by simulation and hardware experimental tests.


sensors, pythonSensors and actuators, mechatronics workshop
39
Home-based rehabilitation system for knee and ankle injuries following surgeryIt is vital for patients with knee and ankle injuries to perform repetitive rehabilitations therapy after their surgery. This is important so that they can fully recover their lower limb functions and avoid the injury to further escalate to a more serious illness. However, there are many patients who refuse to come to the hospitals for the repeated long therapy. This project proposes a home-based rehabilitation system for knee and ankle injuries following surgery. Sensors will be used to measure the knee and/ or ankle movements progress during the exercise and the information will be transferred to the doctors in the hospitals for monitoring purpose. Statistical methods will be used to determine the patients' recovery level and relate to the clinical assessment. This project is based on the request by a doctor in IIUM Hospital Kuantan. It is targeted that the project will contribute to their patient's recovery treatment.1. To develop a data collection set up for a home-based rehabilitation system for knee and ankle injuries following surgery
2. To develop the automatic assessment related to the clinical assessment based on statistical methods.
3. To develop a GUI for the home-based rehabilitation system for knee and ankle injuries following surgery
4. To test the developed home-based rehabilitation system for knee and ankle injuries following surgery
Sensors, arduinosensors and actuators, instruments, mechatronics workshop
40
Identification of Autism in Children Using 3D facial features
with Deep Neural Networks
Autism spectrum disorder (ASD) is a complicated neurological developmental disorder that manifests itself in a variety of ways.. The applicability of static features extracted from autistic children’s face photographs as a biomarker to distinguish them from typically developing children is investigated in this research. Available dataset will be used a publicly to train the suggested models, which consisted of face pictures of children diagnosed with autism and controls classed as autistic and non-autistic.1. Study on Autism spectrum disorder (ASD)
2. Conduct experimental study on different attentional tasks for few children and explored the limits of the face-based attention recognition model for participant and task differences.
3. Analysis data using deep neural network method to detect ASD.
Camera
Software: MATLAB, Python
Signal processingWP3 - Depth of analysis required, WP5 - Extent of applicable codesEA2 - Level of interactions, EA4 - Consequences to society and the environmentNot suitable for FYP
41
Localization for IUM Medibot via ROS, SV to confirm Current Medibot have limited localization via SLAM. This project will incorporate indoor localization via vision or beaconing system- Investigate low cost different localization mechanisms
-incorporate precise indoor localization via vision or beaconing system
Medibot, Realsense Depth CameraInterfacing, Machine Vision
42
Smart Control and Monitoring of mulitple devices in Industrial Automation Lab using IoTProject to control and monitor the motor friction using different type of loads using IoT systemTo help student learn how to control and monitor the motor operation remotely using IoT systemInternet of Things (IoT) components, Nodered, Cloud and PLCIndustrial Automation, System Reliability, Instrumentation, Control System
43
Automated robotic sealing process for aircraft components using UR10e.Student will explore the approach and technique available in sealing 3D components of an aircraft structure. There will be three major components that need to be integrated in the system; UR robot, dispensing system and sealant applicator. The UR robot is an e-series comes with TCP and Remote TCP in which the student need to compare and select which is suitable for the sealing application. Meanwhile, the dispensing system used is a programmable dispenser controller based on time and pressure. As for the sealant applicator, the student has to define the best applicator position and approach to ensure the sealant is applied uniformly on the components based on the requirement. This automated sealant dispensing system is expected to replace the manual application as well as reduce the process cycle time and material wastage.Objective:
1) To design an optimum robot trajectory for 3D path of an aircraft components.
2)To develop an automated sealant dispensing system using UR10e and dispensing system.
3) To analyse the performance developed system based on sealant thickness,width & appearance.
Sensors, actuators, microcontroller, Signal and system analysis, IDPWP1 - Depth of Knowledge Required, WP3 - Depth of analysis requiredEA1 - Range of Resources, EA3 - Innovation
44
Correlations between tilted inclination angles and pressure distribution in wheelchair during dental treatmentDue to budget constraint, patients with physical impairments who cannot afford to buy proper seating equipment will use their wheelchairs while under dental treatments. This can cause posture discomfort where patients will suffer from strenuous neck and will lead to back pain. This project will investigate the relation between tilted inclination angles and pressure distribution in wheelchair during dental treatment.1. To investigate the optimal configuration of pressure sensor on wheelchair during dental treatment
2. To investigate the tilted inclination angles provide comfortable pressure distribution to the patient during dental treatment.
pressure sensor, arduinoinstrumentation
45
A wearable mask to support rehabilitation of facial paralysisThis aim of this project is to develop a wearable mask that can support the physiotherapy session for facial paralysis patients especially the elderly. This system can assist the patient to undergo dental treatment properly.1) to investigate the facial area which is important during physiotherapy session
2) to develop a wearable mask that can support the rehabilitation of facial paralysis
actuator, arduinoinstrumentation
46
Eddy current NDT for welded structureTo develop the system and algorithm for detecting and sizing crack in welded structures. In collaboration with Malaysia NuclearTo develop the system and algorithm for detecting and sizing crack in welded structures
To evaluate its performance
eddy current system. LabVIEW and matlabn/a
47
Pixel level segmentation for sizing pavement cracks by using deep learningTo develop the deep learning model and enhance the system with geolocational data from a GPS systemTo develop the deep learning model
To develop the dataset
To integrate with a GPS system
To evaluate
GoPro Camera, PythonAI
48
Formulation of Logic-based Reinforcement Learning Reward Functions for Self-learning Robotic Navigation.Autonomous robotic navigation has become hotspot research, particularly in complex environments, where inefficient exploration can lead to inefficient navigation. Previous approaches often had a wide range of assumptions and prior knowledge. Adaptations of machine learning (ML) approaches, especially deep learning, play a vital role in the applications of navigation, detection, and prediction about robotic analysis. Further development is needed due to the fast growth of urban megacities. This work proposes a neuro-symbolic approach to get AI to reason. Deep reinforcement learning (DRL) is proposed to develop a self-learning robot that interacts with an unknown environment with embedded evaluative Fuzzy logic function to empower the robot with reasoning capabilities while navigation as well as to optimize DRL computations. The reinforcement learning approach is proposed to develop a model-free and self-learning robot that can be deployed in real scenarios to interact with an environment.-To formulate reinforcement learning reward function via Fuzzy Logic Approach
- To model and simulate the robotic environment navigation through the developed reward function.
-To validate the proposed model
Good Laptop, Unity3D, sensors, raspberry pi.Intelligent COntrol
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100