Introduction to Robotics

• Robotics covers basic concepts

Definition of a Robot

• The word "robot" was created in 1920 by Czech novelist Karel Capek in his play "Rossum's Universal Robots" (RUR).
• The Czech word "robota" translates to "worker" or "servant".
• RIA defines a robot as a reprogrammable, multifunctional manipulator designed to move material, parts, tools, or specialized devices through variable programmed motions to perform diverse tasks.
• JARA describes a robot as a device for movement in a flexible way similar to living organisms, with or without intelligent functions, enabling operations based on human commands.
• DRAE defines a robot as a programmable electronic machine or device able to manipulate objects and perform various operations.
• Keywords include: Manipulator, Reprogrammable, Intelligent functions and Movement

History and Classification

• Areas covered can include; what a robot is?, what a robot can do?, its history, key components, applications etc
• The development of robots can be traced via automatons and a modern timeline
• Robots can be classified by: generation, application area, actuators, number of axes, configuration, kinematic configuration, etc.
• First generation robots are reprogrammable, arm-type devices capable of memorizing repetitive movements using internal sensors,
• Second generation robots (1980s) possess external sensors and feedback, enabling them to make limited decisions and respond to their environment.
• Third generation robots (late 90s and beyond) use artificial intelligence to recognize, learn, think, and adapt to new situations.

Key Robot Components

• Some key robot components are links, actuators, sensors, end-effectors, communication systems and controllers
• Links act as rigid bodies
• Actuators connect and move links
• Sensors help perceiving the world

Robot Construction Limitations

• Physical constraints
• Sensor and actuator technology limits
• Power consumption
• Design issues (general structure, and interaction modality)
• Level of articulation

Robotics Areas

• Mechanical manipulation
• Locomotion
• Computer vision
• Artificial intelligence

Manipulation

• Robotics can move, arrange, operate, or control using hands/mechanical means with skill.
• Examples of manipulation include moving, joining, grinding, bending, and reshaping objects.
• Using a model of how the world works, robots can be programmed to choose actions to accomplish a given goal, searching through action sequences until a suitable plan is found.
• Techniques from rigid body mechanics, kinematic constraint, Coulomb friction, gravity, and impact are used to develop solutions for manipulation problems.

Locomotion

• In biology, locomotion is the self-powered, patterned motion of limbs/anatomical parts by which an individual moves.
• Forms of locomotion include walking, running, crawling, climbing, swimming, and flying.
• Apprenticeship learning allows the control algorithm to accept isolated advice at different hierarchical levels of the control task
• Experimental quadruped locomotion can accomplish tasks with complete trajectories and 3D transformations.

Computer Vision

• Building artificial systems to obtain information from images or multi-dimensional data.
• Perception is extracting information from sensory signals; and computer vision investigates artificial systems for perception from images/multidimensional data.

Artificial Intelligence

• AI is the science of enabling machines to act intelligently with intelligent behavior, learning, and adaptation.
• The seven areas of AI; knowledge representation, understanding natural language, learning, planning and problem solving, inference, search, and vision

Collaborative Robots (Cobots)

• Robots that work alongside humans, and the NASA Valkyrie robot

TED Talks

• TED talks are; Meet ‘Spot, the robot dog that can run, hop and open doors’ | Marc Raibert
• And Meet ‘the dazzling flying machines of the future’ Raffaello D'Andrea

Robot Action Paradigm

• Robot primitives include sense, plan and act
• The hierarchical paradigm flows from sensor data to sensed information, then directives followed by actuator commands
• The reactive paradigm flows from sensor data to sensed information then directly to actuator commands
• The hybrid deliberative/reactive paradigm flows from sensor data to actuator commands, or from cognitive data to directives