0% found this document useful (0 votes)
3 views19 pages

Module-V

The document provides an introduction to robotics, covering key concepts, types of robots, and their historical development. It discusses robot architecture, kinematics, trajectory planning, and methods for motion planning. The content emphasizes the interaction between robots and humans, as well as the importance of optimizing robot movements for efficiency and safety.

Uploaded by

deepagerestrict
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
3 views19 pages

Module-V

The document provides an introduction to robotics, covering key concepts, types of robots, and their historical development. It discusses robot architecture, kinematics, trajectory planning, and methods for motion planning. The content emphasizes the interaction between robots and humans, as well as the importance of optimizing robot movements for efficiency and safety.

Uploaded by

deepagerestrict
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Artificial Intelligence

Dr Abhishek Mukhopadhyay
Assistant Professor, Amity University
(Ex Manager, Ashok Leyland || PostDoc, IISc)
Copyright: Dr Abhishek Mukhopadhyay
Module V : Introduction to
Robotics

Copyright: Dr Abhishek Mukhopadhyay


Introduction

• Robotics is the interdisciplinary field focused


on the design, construction, operation, and
use of robots for performing tasks
autonomously or semi-autonomously
• Key Concepts
• Robot = Mechanical structure + Sensors +
Actuators + Control System
• Autonomous Robot: Operates with
minimal human intervention
• Industrial Robot: Repetitive and precise
work in factories
• Service Robot: Performs human-assisting
tasks (e.g., healthcare, delivery)

Copyright: Dr Abhishek Mukhopadhyay


Type of Robots

Collaborative Robots

Slime/Soft Robot

Self-driven Robot
Copyright: Dr Abhishek Mukhopadhyay
Historical Mention of Robots
• Heron of Alexandria built many remarkable practical inventions
including the Wind-powered organ

• During First BCE, middle eastern civilization excelled in in


making automated mechanical and hydraulic systems
• One of first verified humanoid automaton was designed by
Leonardo Da Vinci

Copyright: Dr Abhishek Mukhopadhyay


Human and
Robots
• Co-existence of robots and human
in terms of enhancing safety and
productivity (CoBot)
• “An apparatus and method for direct
physical interaction between a
person and a general purpose
manipulator controlled by a
computer” - J. Edward Colgate and
Michael Peshkin, professors at
Northwestern University, 1996

Copyright: Dr Abhishek Mukhopadhyay


Architecture of Robot
Component Description

Links, joints, and actuators


Mechanical Unit
(e.g., motors, gears)

Vision, proximity, force


Sensor Unit
sensors

Microcontroller, CPU,
Control Unit
decision logic

Power Unit Battery, power supply

Programming, AI models,
Software Unit
feedback control
Copyright: Dr Abhishek Mukhopadhyay
Robot Kinematics: Position Analysis

• Kinematics
• Study of motion without considering
forces
• Forward Kinematics (FK)
• Computes end-effector position from
known joint parameters
• Example: Given angles of a robotic arm’s
joints, calculate hand location
• Inverse Kinematics (IK)
• Computes joint parameters to reach a
desired position
• Often complex and may have multiple
solutions or none

Copyright: Dr Abhishek Mukhopadhyay


Robot Kinematics: Position Analysis
Parameters Forward Kinematic Model Inverse Kinematic Model
Known Link and Joint variables Position and Orientation of End
effector
Unknown Position and Orientation of End Link and Joint variable
effector

Copyright: Dr Abhishek Mukhopadhyay


Copyright: Dr Abhishek Mukhopadhyay

Forward
Kinematics (Using • Please check the link:
Homogeneous [Link]
ARx1n8EFa9CikRVrDosJH1OdIl/view?us
Transformation) p=sharing
Trajectory Planning

1 2 3 4
Task planning – Path planning – Trajectory planning – Trajectory following –
Designing a set of high- Generating a feasible Generating a time Once the entire
level goals, such as “go path from a start point schedule for how to trajectory is planned,
pick up the object in to a goal point. A path follow a path given there needs to be a
front of you” usually consists of a constraints such as control system that can
set of connected position, velocity, and execute the trajectory
waypoints acceleration, obstacles in a sufficiently
accurate manner

Copyright: Dr Abhishek Mukhopadhyay


Task Space vs. Joint Space
• Task space means the waypoints and
interpolation are on the Cartesian pose
(position and orientation) of a specific
location on the manipulator – usually the
end effector
• Joint space means the waypoints and
interpolation are directly on the joint
positions (angles or displacements,
depending on the type of joint)

Task space vs. Joint Space Trajectory

Source: MathWorks
Trajectory Planning in Robotics

• Trajectory Planning is the process of


determining the path a robot should
follow to reach a goal while
optimizing criteria such as speed,
smoothness, and obstacle avoidance
• Types
• Joint space trajectory:
Movement in terms of joint
angles
• Cartesian space trajectory:
Movement in terms of position
and orientation

Copyright: Dr Abhishek Mukhopadhyay


Copyright: Dr Abhishek Mukhopadhyay

Trajectory Planning for Robot Manipulators


Tasks

Task
planning
• Part of motion planning
• Path: A set of points that links a
start and goal (pose only) Path
vs planning
• Trajectory : Scheduled motion to
follow, including time information
(pose, speed, acceleration) Trajectory
planning

Low-level
control
Joint Space vs Task Space
Trajectories

• Task Space
• Trajectory interpolated in task space
• Inverse kinematics solved every time step

• Joint Space
• Inverse kinematics solved only at
waypoint
• Joint positions interpolated in between

Copyright: Dr Abhishek Mukhopadhyay


Methods

• Pre-programming: Blind specification in task space


• Robot can easily collide with its surroundings
• Teach points: Jog robot using pendant, save points in memory
• Robot should take same path each time
• Motion planning algorithms: Use joint space with potential fields or
sample-based methods
• Robot can avoid or approach objects in environment

Copyright: Dr Abhishek Mukhopadhyay


Procedures

1. Identify the obstacles


2. Identify end effector poses
3. Connect points smoothly

• If objects to avoid/approach are static, then poses can be hard-coded


• If objects move, then the poses must be calculated / updated using
potential fields

Copyright: Dr Abhishek Mukhopadhyay


Connect Points smoothly

• Methods
• Quintic polynomials – a polynomial of fifth degree (task or joint space)
• Roll, pitch, yaw and translation
• Rotation interpolation (task space)
• SLERP: Spherical liner interpolation (joint space)

• All can be used; just switch between task and joint space using transformations

Copyright: Dr Abhishek Mukhopadhyay


• Hands on experience on training and
evaluating on numerical and image
datasets
• Basics of NLP
• Steps in NLP
• Lexical analysis
Take Away • Hands on
• Syntactic Analysis
Points • Parts of Speech Tagging
• Phrase Structuring Rules
• Dependency Parsing
• Semantic Analysis
• Transition Networks
• Type of Transition Networks

Copyright: Dr Abhishek Mukhopadhyay

You might also like