AI in Robotics

Artificial Intelligence in robotics involves imbuing machines with advanced cognitive capabilities. This enables autonomous perception and complex decision-making, fostering greater efficiency and precision in sectors such as manufacturing and healthcare.

courses

• Artificial Intelligence for Robotics (opens in a new tab): This course covers programming aspects of robotic car systems. Topics include planning, search, localization, tracking, and control. The curriculum delves into Kalman Filters, Particle Filters, PID Control, SLAM, providing a comprehensive exploration of artificial intelligence applications in robotics.

• Underactuated Robotics (opens in a new tab): By MIT OpenCourseWare explores nonlinear dynamics and control of underactuated mechanical systems, emphasizing machine learning. Topics include nonlinear dynamics, motion planning, feedback linearization, energy-shaping control, optimal control, and reinforcement learning. Also discusses examples from biology and applications to legged locomotion, compliant manipulation, underwater robots, and flying machines.

• Robotics Specialization (opens in a new tab): By University of Pennsylvania, offers a thorough overview, including robot flight, environment perception, obstacle avoidance, navigation, and complex task completion. The series, comprising six courses, explores aerial robotics, motion planning, mobility, perception, and learning, culminating in a capstone project where students program a robot for tasks like flying and object manipulation. This specialization comprises six courses: Aerial Robotics, Computational Motion Planning, Mobility, Perception, Estimation and Learning and Capstone.

Articles

• Robots in the Manufacturing Industry (opens in a new tab): Explore the dynamic landscape of automation, robotics, and smart technologies reshaping modern manufacturing. Gain insights into how robots are enhancing efficiency, precision, and innovation across production lines, and discover the evolving role of AI-driven automation in this transformative field.

• Scaling up learning across many different robot types (opens in a new tab) (2023): DeepMind's new resources promote general-purpose robotics learning across diverse robot types. The Open X-Embodiment dataset compiles data from 22 robots, while the RT-1-X model transfers skills effectively among them. Tested in five research labs, it demonstrated a 50% improvement compared to specialized methods. (paper)

• Human-Computer Interaction (opens in a new tab): HCI explores the design and interaction between humans and computers. It emphasizes user-centered design, enhancing user experience through intuitive interfaces. HCI combines psychology, design, and technology to create effective, user-friendly systems, fostering seamless communication between users and machines.

Papers

• Deep Reinforcement Learning from Human Preferences (opens in a new tab) (2017): Exploration of reinforcement learning in gaming and robotics domains revealed its remarkable utility for Language Models.

• Habitat 3.0: A Co-Habitat for Humans, Avatars and Robots (opens in a new tab): AI Habitat, a 3D simulator, trains embodied AI agents, like virtual robots, in a photorealistic and efficient environment. Habitat 3.0 specializes in tasks such as house cleaning. In contrast, Habitat III refers to the 2016 UN Conference in Quito, Ecuador, aiming to promote sustainable urbanization and implement a New Urban Agenda. (paper)

Books

• Robotic Manipulation (opens in a new tab): MIT CSAIL's lecture notes on robotic manipulation, employed for an active course at MIT, comprehensively address perception, planning, and control in manipulation tasks. These notes prioritize high-quality implementations of essential tools for manipulation scientists, maintaining a balance between mathematical clarity and runtime performance.

• Robot-Proof: Higher Education in the Age of AI (opens in a new tab) by Joseph E. Aoun's book advocates for preparing students for the AI and robotics era by promoting a "robot-proof" education. This approach aims to cultivate a creative mindset and mental flexibility, enabling students to innovate in ways beyond the capabilities of sophisticated robots.

• Industrial robots and Cobots (opens in a new tab) by Michal Gurgul, presents a thorough exploration of industrial robotics and collaborative robots (cobots), including key subjects like their history, structure, motion, and programming. Tailored for non-technical readers, it aims to offer accessible insights into the growing influence of robots in today's industries.

• Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control (opens in a new tab): Combines data science and ML techniques for modeling, prediction, and control of complex systems. Geared towards advanced undergraduates and early graduate students in engineering and physical sciences, the book covers a wide range of applications, including turbulence, the brain, climate, finance, and robotics. It features in-depth examples, concise explanations, and comprehensive open-source code.

• Robot Building for Beginners (opens in a new tab) by David Cook, provides a comprehensive guide for individuals interested in building their own robots from common parts. The book covers essential topics such as the basics of wiring, electronics, mechanics, and tooling, enabling readers to acquire the necessary skills to create robots. It also includes an introduction to 3D printing, and the master project of the book is a simple, entertaining, line-following robot.

• Principles of Robot Motion: Theory, Algorithms, and Implementations (opens in a new tab): This book delves into crucial aspects of robot motion, covering sensor-based planning, probabilistic planning, localization, mapping, and motion planning for dynamic systems. It connects low-level implementation with high-level algorithmic concepts. Renowned for its coverage of recent advancements, the book's relevance extends to fields beyond robotics, encompassing areas like computer graphics and medicine.

• Introduction to Autonomous Mobile Robots (opens in a new tab): This book serves as a comprehensive introduction to all facets of mobile robotics, encompassing algorithms and mechanisms. It covers key topics, including mechanical, motor, sensory, perceptual, and cognitive layers, as well as locomotion, sensing, localization, and motion planning. The text explores techniques and technologies facilitating mobility, integrating diverse disciplines like kinematics, signal analysis, information theory, AI, and probability theory.

• Introduction to Robotics: Mechanics and Control (opens in a new tab) by Craig, a widely acclaimed textbook since its 1986 debut, this book is a key resource for robotics education. Covering topics like rigid-body transformations, kinematics, dynamics, control methodologies, mechanical design, and programming, it strikes a balance between practicality and theoretical foundations. Known for its real-world applicability and comprehensive coverage, it remains a go-to reference for essential robotics concepts.

Additional Reading

• Timeline of AI & Robotics (opens in a new tab): It provides a timeline of significant events in the history of AI and robotics, spanning from the 1939 introduction of the relay-based Elektro robot, which responded to voice commands, to the incorporation of Siri as a built-in feature on the Apple iPhone. Siri, a voice-activated personal assistant, demonstrated the ability to comprehend natural language requests.