Mobile Robots
Introduction to Mobile Robotics: Defining mobile robots, exploring their diverse types, drawing inspiration from biology, understanding locomotion mechanisms, and examining their real-world applications.
Legged Robots: Analyzing various gaits like biped, quadruped, and hexapod locomotion, and understanding their unique advantages and challenges.
Wheeled Mobile Robots (WMR): Different wheel types, their kinematic constraints, and how these factors influence robot design and movement.
Wheeled Robot Structures: Mobile base structures, their impact on robot mobility, and the trade-offs involved in choosing the right configuration for specific tasks.
State-Space Models of WMRs: State-space models, a powerful mathematical tool for representing and analyzing the kinematics of wheeled mobile robots, including posture and configuration kinematic models.
Motion Control of WMRs: Motion control strategies for wheeled mobile robots, including path following, trajectory stabilization, and posture stabilization, and how feedback control systems play a crucial role.
Sensors for Mobile Robots: Different sensor types, understanding their performance characteristics, and how they enable robots to perceive and interact with their environment.
Simultaneous Localization and Mapping (SLAM): Definition of SLAM, its taxonomy.
Collision Avoidance: Obstacle avoidance techniques like Bug algorithms, Vector Field Histogram methods, and the Dynamic Window Approach.
Path Planning: Sampling-based algorithms (PRM, RRT, A*) and Artificial Potential Fields, enabling robots to navigate efficiently and autonomously.
Fundamental Concepts: Students will gain a solid understanding of the basic concepts in mobile robotics, different types of mobile robots, their unique properties, and their diverse applications across various domains.
Locomotion Mechanisms: Students will learn about the diverse locomotion mechanisms employed by mobile robots, with a particular focus on wheeled mobile bases, walking machines, and flying vehicles. They will understand how to describe their kinematics and motion constraints, crucial for designing and controlling these robots effectively.
Control Systems and Programming: The course will introduce students to the fundamental approaches and structures of control systems used in mobile robots. They will also gain insights into programming these robots, enabling them to execute desired behaviors and tasks.
Perception and Sensors: Students will delve into the world of robot perception, understanding how robots gather information about their environment. They will learn about the classification and properties of various sensors used in mobile robots, enabling them to make informed decisions about sensor selection and integration.
Autonomous Navigation: The course will introduce the challenges and solutions associated with autonomous navigation. Students will explore selected methods and techniques used to achieve various functions of a navigation system, including robot localization, map building, simultaneous localization and mapping (SLAM), collision avoidance, and path planning.
Modeling and Control of Manipulators
Task Priority Control
Introduction to Task-Based Control: Concept of tasks and priorities in robot control, advantages of task-based control over traditional methods, applications in various robotic domains (e.g., industrial manipulators, mobile robots, humanoid robots).
Mathematical Foundations: Review of linear algebra and matrix operations, geometric and kinematic fundamentals of robots, transformation matrices, Jacobians, and null spaces.
Task Specification and Prioritization: Defining tasks in terms of desired robot behavior, assigning priorities to tasks based on importance, handling equality and inequality constraints.
Task-Priority Inverse Kinematics: Solving the inverse kinematics problem with task priorities, classic and modern task-priority algorithms, handling underactuated systems and redundancy.
Control Architectures: Centralized vs. decentralized control for multi-robot systems, kinematic and dynamic control layers, integration of task-based control with mission planning.
Applications and Case Studies: Underwater floating manipulator systems (UVMS), cooperative transportation and manipulation, obstacle avoidance and navigation, teleoperation, and human-robot interaction.
Advanced Topics: Optimization techniques for task-based control, learning and adaptation in task-based frameworks, challenges and future directions in task-based control.
Understand the principles and advantages of task-based control for robotic systems: Students will understand the core concept of task-based control, where robot actions are defined in terms of tasks with assigned priorities. Students will learn how this approach differs from traditional control methods and the benefits it offers, such as flexibility, adaptability, and the ability to handle complex, multi-objective scenarios.
Formulate and prioritize control tasks for various robotic applications: Students will learn how to translate desired robot behaviours into specific control tasks and how to assign priorities to these tasks based on their relative importance and potential conflicts. This skill is crucial for designing effective control strategies for real-world robotic systems.
Implement task-priority inverse kinematics algorithms: Students will learn how to implement algorithms that solve the inverse kinematics problem, determining the required joint movements to achieve desired task objectives while respecting task priorities. This involves understanding both classic and modern task-priority algorithms and their trade-offs.
Design control architectures for single and multi-robot systems: Students will learn how to integrate task-based control into the overall control architecture of robotic systems, considering both single-robot and multi-robot scenarios. This includes understanding the interaction between kinematic and dynamic control layers and how task-based control can be combined with mission planning.
Analyze and evaluate the performance of task-based control systems: Students will learn how to analyze the performance of task-based control systems, identify potential issues or limitations, and propose solutions or improvements. This involves understanding the trade-offs between different control strategies and how to evaluate their effectiveness in achieving desired robot behaviours.
Robot Intelligence Course
Introduction to Cognitive Robotics: Definition of cognitive robotics and the challenges of creating robots with intelligent, context-aware behaviors. Key characteristics of a robot's cognitive architecture, including reactive execution, knowledge representation, reasoning, learning, memory models, predictive models, and interaction models.
Architectural Families: Three primary architectural families for robot cognition: sense-plan-act, reactive/behavior-based, and hybrid reactive-deliberative. Exploration of strengths and weaknesses of each approach and their suitability for different robotic tasks.
Unified Modeling Language (UML): UML as a tool for designing and visualizing robot cognitive architectures. Various UML diagrams, including class diagrams, object diagrams, component diagrams, state machine diagrams, activity diagrams, use case diagrams, and sequence diagrams.
Design Patterns: Design patterns as reusable solutions to common problems in robot architecture design. Various design patterns, including adapter, computational, publish-subscribe, request-process-reply, and sensor-device patterns.
Bio-Inspired Architectures: Bio-inspired architectures for robot cognition, reflexes, taxes, fixed-action patterns, schemas, and innate releasing mechanisms. The subsumption architecture as a prominent example of a bio-inspired approach.
Human-Robot Collaboration: Challenges and opportunities in human-robot collaboration (HRC), the need for robots to adapt to human variability, exhibit intelligibility and naturalness in their interactions, and make decisions in partially well-defined tasks. The concept of AND/OR graphs as a tool for representing and reasoning about cooperative tasks.
Planning with Discrete and Continuous Quantities: Classical planning concepts to handle both discrete and continuous quantities, enabling robots to reason about the physical world more effectively. PDDL+ as an extension of PDDL to model continuous processes, events, and durative actions.
Task-Motion Planning (TMP): Challenges of integrating task planning and motion planning, considering uncertainties in perception, modelling, and execution. Approaches to TMP that leverage belief space planning and probabilistic reasoning.
Search-Based Algorithms: Various search-based algorithms for robot planning and decision-making: explicit search algorithms, SAT procedures, and hierarchical task networks (HTNs). The trade-offs between satisficing and optimal planning approaches.
AND/OR Graphs, HTNs, and Behavior Trees: Advanced planning and execution frameworks, including AND/OR graphs, HTNs, and behavior trees, and their applications in various robotic scenarios such as human-robot collaboration, task decomposition, and reactive behaviors.
Foundational Models & End-to-End Learning Architectures: The role of foundational models and end-to-end learning architectures in robot intelligence. Discussion on how these approaches can enable robots to learn from large-scale datasets and generalize to new tasks and environments.
Understanding Robot Cognition: Students will develop a deep understanding of the fundamental concepts and challenges in robot cognition, including knowledge representation, reasoning, planning, learning, and interaction.
Designing Cognitive Architectures: The course will equip students with the skills to design and implement cognitive architectures for robots using various approaches, including symbolic planning, reactive behaviors, and hybrid methods.
Applying AI Techniques: Students will learn how to apply various artificial intelligence techniques, such as search algorithms, probabilistic reasoning, and machine learning, to solve problems in robot cognition and behavior generation.
Analyzing and Evaluating Architectures: Students will develop the ability to critically analyze and evaluate existing robot cognitive architectures. They will learn how to identify strengths and weaknesses, compare different approaches, and make informed decisions about the most suitable architecture for a given robotic task.
Addressing Real-World Challenges: The course will emphasize the practical challenges of deploying robot intelligence in real-world scenarios, such as human-robot collaboration, task-motion planning, and handling uncertainties in perception and action.
Visual Perception
This course explores the fascinating world of visual perception, examining how we see and interpret the world around us. We will delve into the biological, psychological, and computational mechanisms that underlie our ability to perceive shape, color, motion, depth, and objects. Through lectures, readings, discussions, and demonstrations, students will gain a comprehensive understanding of the perceptual processes that shape our visual experience.
Upon successful completion of this course, students will be able to:
Introduction to Visual Perception:
The Visual System:
Early Visual Processing:
Object Recognition:
Scene Perception:
Attention and Awareness:
Visual Illusions:
Applications of Visual Perception:
Perception and Manipulation
This course explores the intertwined processes of perception and manipulation, focusing on how robots and intelligent systems perceive their environment and interact with it. We will delve into the fundamental principles of computer vision, tactile sensing, and robot kinematics, dynamics, and control. Through lectures, readings, discussions, and hands-on projects, students will gain a comprehensive understanding of the challenges and solutions in building robots that can perceive and manipulate objects in the real world.
Upon successful completion of this course, students will be able to:
Introduction to Perception and Manipulation:
Computer Vision for Robotics:
Tactile Sensing and Haptics:
Robot Kinematics and Dynamics:
Motion Planning and Control:
Grasping and Manipulation:
Perception-Based Manipulation:
Applications and Case Studies:
Computer-Aided Design
This course provides a comprehensive introduction to Computer-Aided Design (CAD) with a focus on both theoretical concepts and practical applications. Students will learn to use industry-standard CAD software to create 2D drawings and 3D models, applying these skills to solve design problems in various engineering and design disciplines. The course will cover fundamental concepts such as geometric modeling, projections, dimensioning, and tolerancing, while also exploring advanced topics like assemblies, simulations, and manufacturing processes.
Upon successful completion of this course, students will be able to:
Introduction to CAD:
2D Design Fundamentals:
3D Modeling:
Advanced CAD Techniques:
Design for Manufacturing:
Applications and Case Studies:
Product Design and Development
This course provides a comprehensive overview of the product design and development process, from initial concept to market launch. Students will learn about the integration of user needs, market trends, engineering principles, and business considerations in creating successful products. The course will cover a range of topics including design thinking, user research, ideation, prototyping, manufacturing, and product lifecycle management. Through lectures, case studies, and hands-on projects, students will gain practical experience in developing innovative and marketable products.
Upon successful completion of this course, students will be able to:
Introduction to Product Design and Development:
Design Thinking and User-Centered Design:
User Research and Market Analysis:
Concept Generation and Selection:
Prototyping and Testing:
Manufacturing and Materials:
Product Lifecycle Management:
Case Studies and Industry Examples: