Postgraduate Courses

Back

Expand All Print All

ROAS

Robotics and Autonomous Systems

ROAS 5100
Advanced Robot Mechanisms and Design
[3-0-0:3]
Description
The design of robotic mechanisms is a core topic in robotics, and nurturing creativity is a key issue in producing human resources. Through training in machine creation, students’ creativity could be expanded. This course provides students with an explanation of the creative design of robotic mechanisms to educate them on how to create a robotic machine. The content includes an overview of the mechanisms of robot manipulators and mobile robots and an explanation of some robotic mechanisms as examples of creative design. The goal is to understand basic robot mechanisms and learn the method of how to create machines to realize the given functions.
ROAS 5110
Data and Information Compression
[3-0-0:3]
Description
This course begins with an overview of some fundamental information theory related to data compression. Lossless techniques, including Huffman/ arithmetic coding, and their applications; and lossy techniques, including quantization (both scalar and vector), transform coding, predictive coding and their applications will be discussed. Several international standards (such as JPEG for image coding, and H26x, MPEGx and its variants) will be discussed. Programming demos and exercises on various image and speech codes will be an integral part of this course.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Understand the BIG background of audio-visual data compression.
- 2.
  Be familiar both mathematically and conceptually with the fundamental compression techniques.
- 3.
  Be knowledgeable to some popular image and video coding standards.
- 4.
  Understand why motion information is so important in video coding and how to perform motion estimation.
- 5.
  Apply the most suitable compression technique to a given audio or visual signal.
- 6.
  Use software tools, such as Matlab, to implement some practical coding systems.
ROAS 5111
Soft Robotics
[3-0-0:3]
Previous Course Code(s)
ROAS 6000K
Background
Students should have a basic understanding of mechanical design, materials, and manufacturing techniques. A background in engineering (e.g. mechanical engineering) or science (e.g. materials science) will help students follow the course content more easily.
Description
This course provides a comprehensive introduction to soft robotics, focusing on the design, fabrication, and application of compliant and flexible robots. Students will explore the fundamental principles of soft materials, actuation methods, and sensing technologies that enable soft robots to interact safely with complex, unstructured environments. It covers key topics including material selection, mechanical modeling, control strategies, and manufacturing techniques. Through detailed case studies, students will examine how specific materials and design choices improve the functionality and efficiency of these robots for applications in healthcare and exploration.
ROAS 5120
Formal Verification and Control of Dynamical Systems
[3-0-0:3]
Previous Course Code(s)
ROAS 6000L
Description
Formal methods originate from theoretical computer science and have been seamlessly integrated with dynamical systems. At its core, formal methods involve formulating specifications to form proof obligations, verifying that the systems indeed meet their specifications via algorithmic proof search, and designing systems to meet those obligations. This course bridges fundamental gaps between formal methods and control theory. It introduces fundamental theories and techniques of formal methods that apply broadly to various dynamical systems, such as robots, autonomous systems and cyber physical systems. Particularly, the following topics will be covered: symbolic system modeling, regular and omega-regular properties, linear temporal logic, model checking, system simulation and abstraction, game theoretic control synthesis and cutting-edge engineering applications.
ROAS 5200
Medical Microrobotics: Principles and Applications
[3-0-0:3]
Co-list with
BSBE 5700
Exclusion(s)
BSBE 5700
Description
This course introduces microscopic robots and covers the fundamentals of their design, fabrication, and control for healthcare applications. Students will explore the interdisciplinary fields of engineering, materials science, medicine, and nanotechnology, and understand how these robots are set to revolutionize diagnostics, targeted therapy, and minimally invasive surgery. This course aims to equip students with the theoretical knowledge and practical insights necessary to contribute to this rapidly advancing field.
ROAS 5210
Advanced Robotics
[3-0-0:3]
Previous Course Code(s)
ROAS 6000M
Description
Learning about the core principles and latest developments in Robotics is crucial for nurturing innovative students in this field. This course provides a comprehensive overview of the core concepts in robotics. Students will mainly explore the principles of robot kinematics, dynamics, planning and control. In addition, some latest advancements in robotics integrated with artificial intelligence will also be discussed, such as VLA and robot foundation models.
ROAS 5300
Multiple View Geometry for Mobile Robot Navigation
[3-0-0:3]
Description
Multiple view geometry techniques can provide robots with the ability to better perceive and reconstruct their surroundings, leading to improved navigation capabilities. This course aims to help students gain a deep understanding of the mathematical and computational methods used in multiple view geometry, including camera calibration, stereo vision, structure from motion, and 3D reconstruction. This knowledge is essential for developing algorithms and systems that can accurately interpret and manipulate visual data in mobile robot navigation and related applications.
ROAS 5400
Learning for 3D Vision
[3-0-0:3]
Previous Course Code(s)
ROAS 6000J
Background
Students are expected to have a basic knowledge of Machine Learning and Computer Vision.
Description
This course explores the forefront of computer vision through the lens of novel neural 3D representations. Students will investigate advanced techniques that leverage deep learning to enhance the understanding and interpretation of 3D data. Key topics include the development and application of innovative neural architectures for processing point clouds, voxel grids, and implicit surfaces, as well as the integration of these representations into real-world applications such as augmented reality, autonomous navigation, and robotics. Through a blend of theoretical insights and practical projects, students will engage with state-of the-art methodologies and contribute to ongoing research in the field.
ROAS 5500
Mechatronics Design
[3-0-0:3]
Previous Course Code(s)
ROAS 6000D
Description
This course introduces the essential fundamentals, including modeling, sensing, signal transmission and conversion, actuation, control, simulation, and implementation technologies used within the mechatronics design for robots and autonomous systems. It will give a holistic view of advanced automation technologies in industrial applications, taking electric vehicles and robots as examples. Also, this course will provide the essential skills to design intelligent mechatronics systems, and students will be allocated Arduino and STM32 development kits to participate in mechatronics design in the course project. Through this course, students can enhance their understanding of the cross-disciplinary integration and systematic optimization of mechatronics systems involving the knowledge of mechanics, electronics, control engineering, and computer science.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Have an overview of the basic components, functionalities, and applications of mechatronics systems.
- 2.
  Understand the core technologies in modeling, sensing, signal transmission and conversion, actuation, control, and simulation covered by mechatronics design.
- 3.
  Apply the learned knowledge to interpret, analyze and exemplify different areas of mechatronics design, such as legged robots, unmanned aerial vehicles, and autonomous-driving electric vehicles.
- 4.
  Contextualize and analyze mechatronics systems for engineering processes.
- 5.
  Develop a professional engineering sense and adequate maturity to design and implement real-life mechatronic systems.
- 6.
  Cultivate "5C" skills, namely creativity, critical thinking, communication, collaboration, cross-disciplinary capability.
- 7.
  Discover and learn the cutting-edge technologies and theories from the recent research papers and advanced mechatronics products in the related field.
- 8.
  Identify scientific and engineering significances, challenges, and new trends in mechatronics design.
ROAS 5600
Introduction to Discrete Event Systems
[3-0-0:3]
Description
This course aims to provide an introduction to the fundamental knowledge of physical systems modeled with discrete state space and event driven transitions. Discrete Event Systems (DES) arise in the modeling of many engineering domains, such as automated manufacturing systems, communication networks, software systems, process control systems, and transportation systems. This course will introduce a unified modeling framework and emphasize the analysis and control of DES. Basics of automata and language theory are presented first as mathematical preliminaries. Then comes a detailed treatment of state estimation, diagnosis, security and supervisory control theory of DES based on automata model. Topics of other DES models like Petri nets, timed and hybrid automata are also covered towards the end of the course.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Demonstrate mastery of mathematical essentials of discrete event systems.
- 2.
  Have a thorough understanding of the current development of discrete event systems.
- 3.
  Identify the significance of discrete event systems in modern robotics and autonomous systems.
- 4.
  Analyze and design control systems via knowledge of discrete event systems.
- 5.
  Build solid foundations for more advanced topics and courses in control and autonomous systems.
- 6.
  Emerge with a clear picture of both advantages and limitations of theory of discrete event systems.
ROAS 5700
Robot Motion Planning and Control
[3-0-0:3]
Previous Course Code(s)
ROAS 6000C
Description
This course introduces the advanced methodologies in the context of motion planning and control for robotics and autonomous systems. Various methodologies are introduced, including search-based methods, grid-based methods, sampling-based methods, optimization-based methods, learning-based methods, etc. In general, this course covers modern approaches, deep theory, and good practice envisions. In addition to the fundamental knowledge in motion planning and control, the students will also have the opportunity to discover and learn cutting-edge methodologies in the related field, aligning with the substantial developments in robotics, autonomous driving, UAVs, etc.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Have an overview of the basic components, functionalities, and applications of motion planning and control.
- 2.
  Understand the core algorithmic approaches in motion planning and control of robotics and autonomous systems.
- 3.
  Apply learned methodologies to motion planning and control tasks in specific applications (with simulations), such as mobile robots, autonomous driving, UAVs, etc.
- 4.
  Develop presentation skills and scientific writing skills.
- 5.
  Discover and learn cutting-edge methodologies from the recent research papers in the related field.
- 6.
  Identify scientific and engineering significances, challenges, and new trends in motion planning and control.
ROAS 5710
Optimal Control and Reinforcement Learning
[3-0-0:3]
Previous Course Code(s)
ROAS 6000I
Description
Optimal control and reinforcement learning is an effective approach that has been widely used in robotics and autonomous systems. This course aims to provide students with a firm foundation in optimality principles in modern control systems. Fundamental key concepts in optimal control are introduced, including Hamiltonian, Pontryagin's minimum principle, Bellman equation, dynamic programming, etc., which bring the prospect of formal linkage to reinforcement learning techniques. Different optimal control and reinforcement learning methods are also covered in this course, such as LQR, Kalman filter, LQG, MPC, Actor-Critic, PPO, etc. Additionally, the students will have the opportunity to discover and learn cutting-edge methodologies in the related field and develop the expertise in optimal control and reinforcement learning.
ROAS 5800
Physical-based Vision for Robot and Autonomous Driving
[3-0-0:3]
Previous Course Code(s)
ROAS 6000B
Description
Light traveling in the 3D world interacts with the scene through intricate processes before being captured by a camera. These processes result in the dazzling effects like color and shading, complex surface and material appearance, different weathering, just to name a few. Physics based vision aims to invert the processes to recover the scene properties, such as shape, reflectance, light distribution, medium properties, etc., from the images by modelling and analyzing the imaging process to extract desired features or information. This course introduces the advanced methodologies in the context of physical-based vision for robotics and autonomous systems. We will introduce diverse techniques, covering from traditional methods based on hand-crafted features to recent deep learning methods. Apart from the fundamental knowledge in physical-based vision, the students will also have opportunities to discover and learn cutting-edge methodologies in popular physical-based vision topics (i.e., bad-weather restoration, shadow detection and removal, specular highlight detection and removal, intrinsic image decomposition, reflection detection and removal, and so on) of the physical-based vision, aligning with the substantial developments in robotics, autonomous driving, UAVs, etc.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Have an overview of the basic components, popular topics, and applications of physical-based vision.
- 2.
  Understand the core algorithmic approaches in physical-based vision for robotics and autonomous systems.
- 3.
  Apply advanced deep learning techniques to address physical-based vision task for robots, autonomous driving, UAVs, etc.
- 4.
  Develop presentation skills and scientific writing skills.
- 5.
  Discover and learn cutting-edge methodologies from the recent research papers in the related field, such as bad-weather restoration, shadow detection and removal, specular highlight detection and removal, intrinsic image decomposition, reflection detection and removal.
- 6.
  Identify scientific and engineering significances, challenges, and new trends in physical-based vision.
- 7.
  Possess research capabilities in computer vision and medical imaging, robots, and autonomous driving.
ROAS 5810
Advanced Learning Techniques for Medical Robotics Systems
[3-0-0:3]
Previous Course Code(s)
ROAS 6000G
Description
This course will present how to utilize many advanced learning techniques and modules to analyze and interpret diverse medical image modalities (such as ultrasound, CT, MRI, X-ray, and so on) in recent years. The techniques to be covered include supervised learning, video understanding, multi-modal learning, transformer, self supervised learning, semi-supervised learning, domain adaptation, generative models, large models, and so on. The course will cover an overview of medical image modalities, traditional medical image processing methods, deep learning techniques and their applications in medical image analysis, and advanced data-efficient learning on medical imaging.
ROAS 5900
Analytical Methods in Human Factors Research
[3-0-0:3]
Co-list with
INTR 5330
Exclusion(s)
INTR 5330
Background
Previous coursework in Probability and Statistics, including knowledge of estimation, confidence intervals, and hypothesis testing and its use in at least one and two sample problems. Some familiarity with Calculus and Linear Algebra.
Mode of Delivery
[BLD] Blended learning
Description
The course will cover a wide range of analytical methods used in human factors research domain. The students will gain an understanding of the procedures, objectives and limitations of different research methods. The course will also include four case studies so that students would gain first-hand experience in applying the methods in real projects. These contents are required for research investigating users’ behaviors.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Understand commonly used models of human beings
- 2.
  Acquire techniques in analyzing humans’ behaviors
- 3.
  Understand commonly used statistical models for analyzing experimental data
- 4.
  Gain first-hand experience in designing experiments in human factors area and applying the techniques in research projects
- 5.
  Understand how to present research processes and results (either orally or in writing) in a scientific way
ROAS 5910
Engineering Psychology and Transportation Applications
[3-0-0:3]
Co-list with
INTR 5260
Exclusion(s)
INTR 5260
Description
The course will cover a wide range of engineering psychology topics as well as how the research in these directions can affect policies and regulations in vehicle design and surface transportation. The students will gain an understanding of the characteristics and limitations of human beings from engineering psychology perspectives of view and how the design of traffic control devices, the roadway, the in-vehicle devices, regulations and traffic rules can be affected by the research in these directions.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Understand basic characteristics and limitations of human beings.
- 2.
  Acquire techniques in analyzing road/traffic control design and machine design from human factors point of view.
- 3.
  Understand common collision types on road and corresponding commonly used countermeasures.
- 4.
  Understand characteristics and causes of driver impairments and countermeasures.
- 5.
  Understanding the human factors challenges in the context of driving automation and connected vehicles technologies.
- 6.
  Design and conduct experiments to collect data from human subjects and analyze the data.
- 7.
  Write written reports and orally present the findings in the experiments.
ROAS 6000
Special Topics in Robotics
[3-0-0:3]
Description
Selected topics in robotics of current interest in emerging areas and not covered by existing courses. May be repeated for credit if different topics are covered.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Identify the knowledge of the selected topics in the field of robotics.
ROAS 6800
Seminar in Robotics and Autonomous Systems
[0-1-0:0]
Description
Seminar topics presented by students, faculty and guest speakers. Students are expected to attend regularly and demonstrate proficiency in presentation in accordance with the program requirements. Graded P or F.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Demonstrate effective presentation skills in seminar presentations.
- 2.
  Understand current research trends and applications.
- 3.
  Engage argumentative and critical thinking skills.
ROAS 6900
Independent Study
[1-3 credit(s)]
Description
An independent study on selected topics carried out under the supervision of a faculty member.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Demonstrate mastery of the knowledge and skills in the selected topics related to Robotics and Autonomous Systems.
- 2.
  Apply an interdisciplinary approach in examining the selected topics.
- 3.
  Critically evaluate different aspects of the selected topics.
- 4.
  Communicate findings effectively in written reports.
ROAS 6990
MPhil Thesis Research
Description
Master's thesis research supervised by co-advisors from different disciplines. A successful defense of the thesis leads to the grade Pass. No course credit is assigned.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Design, develop and conduct crossdisciplinary research in Robotics
  and Autonomous Systems.
- 2.
  Communicate research findings effectively in written and oral
  presentations.
ROAS 7990
Doctoral Thesis Research
Description
Original and independent doctoral thesis research supervised by co-advisors from different disciplines. A successful defense of the thesis leads to the grade Pass. No course credit is assigned.
Intended Learning Outcomes
On successful completion of the course, students will be able to:
- 1.
  Design, develop and conduct crossdisciplinary research in Robotics and Autonomous Systems.
- 2.
  Communicate research findings effectively in written and oral presentations.

Back