Special Topics Course Descriptions - page 2

Operations

OIE 598E. ENGINEERING ECONOMICS

The intention of this course is to aid all engineering students in understanding economics and business constraints on engineering decision making. Topics may include but will not be limited to: evaluation of alternative; the six time-value-of-money factors; present worth, annual cash flow and rate-of-return analysis; incremental analysis; depreciation and income taxes; replacement analysis; inflation; handling probabilistic events; public economy; break-even and minimum cost points; and foreign exchange.

OIE 598. ST: SUSTAINABLE SUPPLY CHAIN & OPERATIONS MANAGEMENT

This course is intended to provide students with understanding the intra- and inter-organizational implications of environmental practices and policies.  The role of organizational operational and supply chain management functions, activities, tools and methods and their relationship to the natural environment will be introduced and discussed. At the end of the course a successful student should be able to: grasp the scope of general operations and supply chain management and environmental sustainability as they relate to the firm, be able to relate to the manners in which management may respond and collaborate/cooperate with suppliers, customers, and various other stakeholders influencing and influenced by operational and supply chain activities from practical and theoretical case studies and able to evaluate various factors and understand trade-offs in management decisions as they pertain to environmental operations and supply chain management.

Robotics Engineering

RBE 521 ST. LEGGED ROBOTICS

Foundations and principles of parallel and walking mechanisms. Topics include advanced spatial/3D kinematics and dynamics of parallel manipulators and legged/walking mechanisms including workspace analysis, inverse and forward kinematics and dynamics, gait analysis of walking mechanisms, motion analysis of parallel mechanisms as well as legged and walking mechanisms, stability/balance analysis of walking mechanisms, and control of parallel manipulators and walking mechanisms. The course will be useful for solving problems dealing with parallel manipulators as well as multi-legged walking mechanisms including humanoid robots, quadruped robots, hexapod robots and all other types of legged walking mechanisms. A final term project would allow students to apply all this information to design, analyze, and simulate parallel and walking mechanisms. Students taking this course are expected to have a background in kinematics and dynamics.

RBE/CS 526. HUMAN-ROBOT INTERACTION

This course will introduce human-robot interaction on the topics of (1) shared autonomy for medical robots, (2) robot operation interface, and (3) robot learning from a human teacher. Topic 1 will cover the shared autonomous motion planning for upper limb exoskeleton and mobile humanoid nursing robot; Topic 2 will cover the EMG/IMU interface for wearable robots and the multi-modal interface for robot teleoperation; Topic 3 will cover the modeling and learning of human motor skills and action decision-making intelligence, in order to reproduce and facilitate physical human-robot interaction. The course projects will focus on the application of tele-nursing robots (physical platform) and predictive biomechanical human driver model (simulation). The course work will include but not limited to paper reading and review, implementation of algorithms, human movement data collection, and analysis, as well as user study for robot autonomy design evaluation. 

RBE 595 ST. ADVANCED ROBOT NAVIGATION

In recent years, robots have become part of our everyday lives. Leaving the research labs to be part of the common tools of a household, tools such as robotic vacuum cleaners (iRobot Roomba, Kalorik), pool cleaners (Polaris, Maytronics), Lawn mowers (Landroid, LawnBott) and more abound. For navigating safely, these robots need the ability to localize themselves autonomously using their onboard sensors. 
Potential applications of such systems include the automatic 3D reconstruction, 3D reconstruction of buildings, inspection and simple maintenance tasks, metric exploitation, surveillance of public places as well as in search and rescue systems. In this course, we will dive deep into the current techniques for 3D localization, mapping and navigation that are suitable for robotic applications. Required prerequisites: RBE 500 - Foundations of Robotics, RBE 501 – Robot Dynamics, RBE 502 – Robot Control

RBE 595 ST. FUNDAMENTALS OF ROBOTICS AND ARTIFICIAL INTELLIGENCE FOR AUTONOMOUS VEHICLE APPLICATIONS

This course will provide an introduction to the fundamentals of robotics and artificial intelligence, with a focus on their applications in autonomous vehicles. Topics covered will include:

  • The history of robotics, artificial intelligence, and autonomous vehicles
  • The fundamentals of robotics, including sensors, actuators, and control systems
  • The fundamentals of artificial intelligence, including machine learning and computer vision
  • The application of robotics and artificial intelligence to autonomous vehicles, including object detection and classification, obstacle avoidance, localization & mapping, and motion planning & control

Prerequisites:
Strong background in mathematics and physics is required. We will use Python and Jupyter-lab to go through examples and homework. If you have never used Python and Jupyter-lab programming before the class, we will cover the basics of them in the first week of the class.

RBE 595 ST. SOFT ROBOTICS (2 CREDITS)

Soft robotics studies ‘intelligent’ machines and devices that incorporate some form of compliance in their mechanics. Elasticity is not a byproduct but an integral part of these systems, responsible for inherent safety, adaptation and part of the computation in this class of robots. This course will cover a number of major topics of soft robotics including but not limited to design and fabrication of soft systems, elastic actuation, embedded intelligence, soft robotic modeling and control, and fluidic power. Students will implement new design and fabrication methodologies of soft robots, read recent literature in the field, and complete a project to supplement the course material. Required Background: Differential equations, linear algebra, stress analysis, kinematics.

RBE 595 ST. REINFORCEMENT LEARNING

This course will provide a solid introduction to the field of Reinforcement Learning (RL). Students will learn about the core challenges and approaches including Markov decision processes, model based, model free RL, on-policy and off-policy learning, and approximate solution techniques. Through a combination of lectures and coding assignments, students will become well versed in key ideas and techniques in RL and its application in robotic systems. To get students familiarized with the state-of-the-art RL algorithms in robotics, research papers are provided, and students are required to give a presentation about the papers. In addition, an end of the term team project would allow the students to apply mastery of the subject to a real-world robotics application.

Prerequisites: A probability course is required, as well as proficiency in Python. RBE 500/Foundations of Robotics and basic knowledge of neural networks preferred, but not required.

RBE 595 ST. DEEP LEARNING FOR ADVANCED ROBOT PERCEPTION

This course will cover deep learning and its applications to perception in many modalities, focusing on those relevant for robotics (images (RGB and RGB-D), videos, and audio). Deep learning is a sub-field of machine learning that deals with learning hierarchical features representations in a data-driven manner, representing the input data in increasing levels of abstraction.

The course will cover the fundamental theory behind these techniques, with topics ranging from sparse coding/filtering, autoencoders, convolutional neural networks, deep belief nets, and Deep reinforcement networks. We will cover both supervised and unsupervised variants of these algorithms, and we will work with real-world examples in perception-related tasks, including robot perception (object recognition/classification, activity recognition, loop closure, etc.), robot behavior (obstacle avoidance, grasping, navigation, etc.), and more.

The course will involve a project where students will be able to take relevant research problems in their particular field, apply the techniques and principles learned in the course to develop an approach, and implement it to investigate how these techniques are applicable.

RBE 595 ST. ARTIFICIAL INTELLIGENCE FOR AUTONOMOUS VEHICLES

Autonomous vehicles or self-driving vehicles represent one of the most significant advances in technology. Their impact will go beyond technology, beyond transportation, beyond urban planning to change our daily lives in ways we have yet to imagine. 

 This course is an introduction to machine learning and deep neural networks and its application in the domain of self-driving cars. Deep learning algorithms are becoming the disruptor agent that is allowing the rapid developments of autonomy in its multidimensional level. Currently we can find research of deep learning algorithms trying to improve driving performance, detecting pedestrians, detecting traffic, detecting and taking actions while navigating in a highway, reading the traffic signs and evaluating in real-time the readiness of the driver while he/she is on control and more.

Students who enroll in this course will master state-of-the-art technologies that are shaping the future of the field. Students will be exposed to interactive projects in vehicle perception, vehicle cognition, vehicle controls, localization, motion planning, and more. 

The course will involve the development of a project where students will be able to take relevant research problems in the field of autonomous vehicles, apply the techniques and principles learned in the course to develop an approach, and implement it to investigate how these techniques are applicable.   Prerequisites: Undergraduate or graduate level course in Linear Algebra, RBE 500  Foundations of Robotics,  RBE 550 Motion Planning,  Proficiency in Python.

RBE 595 ST. ROBOT PERCEPTION

This course will cover deep learning and its applications to perception in many modalities, focusing on those relevant for robotics (images (RGB and RGB-D), videos, and audio). Deep learning is a sub-field of machine learning that deals with learning hierarchical features representations in a data-driven manner, representing the input data in increasing levels of abstraction. 

The course will cover the fundamental theory behind these techniques, with topics ranging from sparse coding/filtering, autoencoders, convolutional neural networks, deep belief nets, and Deep reinforcement networks. We will cover both supervised and unsupervised variants of these algorithms, and we will work with real-world examples in perception-related tasks, including robot perception (object recognition/classification, activity recognition, loop closure, etc.), robot behavior (obstacle avoidance, grasping, navigation, etc.), and more. 

The course will involve a project where students will be able to take relevant research problems in their particular field, apply the techniques and principles learned in the course to develop an approach, and implement it to investigate how these techniques are applicable.

RBE 595 ST. SENSORS FUSION AND PERCEPTION FOR AUTONOMOUS VEHICLES

This course focuses in Sensor Fusion, Image Processing and Computer Vision techniques for Autonomous Vehicles. The class covers four topics: Image Processing (Image Enhancement, Filtering, Advanced Edge and Texture), 2D/3D Vision (3D Geometry from Multiple view geometry, Motion Processing and Stereo) Sensor fusion (homogeneous fusion, heterogeneous fusion and sensor integration) and Image Segmentation and Object Recognition. Students will be introduced to several existing software toolboxes from Vision and Robotics, and will implement a number of smaller projects 

Moreover, this course presents a variety of tools and approaches for solving fundamental problems involving sensor fusion and perception. Topics to be covered include the mathematical formulation of fusion algorithms, the use of sensor fusion to solve visual perception degeneratives, time domain discrepancies, and accurate reconstruction, and the design and implementation of heterogeneous sensor fusion approaches.  Prerequisite: RBE 500.

WR 593. ROBOT FUTURES: DESIGN, ETHICS, AND COMMUNICATIONS

This course introduces the theory and practice for the motion control of human-compatible robotic systems. Ideally, the motion of a wearable robot system should be dynamically transparent to its operator, sensitively responsible to the voluntary and involuntary motions of its operator. When used for robot-assisted stroke rehabilitation, a wearable robot system is expected to assist to the operator’s motor skills and correct abnormal arm motions resulting from motor disabilities. In this course, students will study the biomechanics of human motion, the theories of human motion control, and the methods for controlling biologically-compatible robots. Students will also experimentally investigate human motions using a Vicon motion capture system, propose and test their hypothesis on human motion control strategies, and implement motion control algorithms on wearable robots and/or arm-like robotic manipulators.

Systems Engineering

SYS 579. SOFTWARE SYSTEMS ENGINEERING

Software has become the primary engineering mechanism for implementation. And while Systems Engineers may not write code (20% of work) , every other aspect of "software engineering" (80% of work) is vital to their success.
This hands-on course teaches the critical elements of software engineering through a class project. Requirements (using user stories), Design (using UML), Testing (using inspection) and Evaluation (using ODC). The scope is ambitious. You will learn, through working with a team applying the theory you learn in class on the work product that you are responsible to deliver. This is as close as it gets to an industry project with a coach who can teach theory and help you execute your tasks.

On completion of this course, students will:

  • Identify the pros and cons of each software process model.
  • Build an application as a team class project
  • Capture requirements in natural language, User Stories, and UML
  • Use UML to translate detailed requirements into Use Cases
  • Implement Class diagrams and Sequence diagrams
  • Inspect the work product, open defects and track closure
  • Learn and apply ODC: Orthogonal Defect Classification
  • Analyze data from cross-team artifacts
  • Understand the goals of Integration and the current practice of CICD
  • Contrast classical project management with current best practice

SYS 579C. COMPLEX DECISION MAKING

One of the biggest ways that you can influence the quality of your life is by improving the quality of your decisions. Complex Decision Making is intended for professionals in management positions and/ or those individuals, regardless of industry, who seek to enhance both their career potential and their overall quality of life. Based on logical principles, and informed by what we know about the limitations of human judgment and decision-making in complex situations, the course trains managers how to think about and structure decisions. These decisions incorporate both their everyday decisions as well as the tough, complex decisions that involve uncertainty, risk, several possible perspectives, and multiple competing objectives, thus improving the quality of the resulting decisions. In addition to teaching formal decision theory and application, we will explore cognitive biases that prevent us from being completely rational in our thinking and deciding. Exit this course able to define the right decision problem, clearly specify your objectives, create imaginative alternatives, understand consequences, grapple with trade-offs, clarify uncertainties, and think hard about your individual values and risk tolerance.

SYS 579D. ENGINEERING DEPENDABLE AND SECURE SYSTEMS

This course considers all facets of engineering dependable and secure systems, i.e., systems that are reliable, available, secure, and can be depended upon to deliver their intended capabilities despite hardware failures, software failures, network failures, external attack, and unexpected behavior. Topics include building dependable system architectures; resilience; security and quality of service of networks; dependability assessment; and software reliability. The class will consist of lectures, case studies, and a class project. (Prerequisite: SYS 501.)

SYS 579R. SYSTEM RELIABILITY ENGINEERING

This course will present reliability, maintainability, and related topics with the breadth of techniques and depth of detail that will benefit the systems engineer by allowing him/her to understand how they relate to the specification, development, testing, and fielding of reliable systems. The reliability of electronics, mechanical equipment, and software will be covered from the component level through their application at the system level. Other key topics will be: reliability prediction; failure modes, effects, and criticality analysis; stress testing; accelerated life testing; and reliability management. In addition, a series of relevant case studies will be studied and discussed.

SYS 579S. SYSTEM OF SYSTEMS ENGINEERING

An innovative approach to engineering complex systems of systems is developed. This approach relies heavily on case studies to drive the discovery of effective techniques. We will discuss complex systems of systems characteristics and behaviors, enterprises, the principled engineering of systems of systems, and distinctions between these forms and conventional approaches. A forward-looking, people-focused approach will be developed, with emphasis on systems thinking; posing a guiding architecture (not just architectural views) up-front that does not change much as the system evolves; balancing competing factors rather than subsystem optimization; pursuing opportunities as opposed to just mitigating risks; sharing information to build interpersonal trust; and communicating individual perspectives to collectively garner better views of the underlying reality. The overall goal is to revisit and broaden one’s “mindsight” in order to build more effective, resilient, scalable, and durable systems. Prerequisites: SYS 501 and SYS 510.