Subashish
Home
Research
Software
Courses
Students
Workshop
Presentations and Talks
PG Orientation
Reading Materials

Robotics and Automation (ELL406)

Time and Venue (Semester-II, 2024-25)

  • 11:00 AM – 12:00 Noon, On Tuesday, Thursday and Friday at LH410 in LHC.

TAs: Suman, Anindya and Dhruva

Syllabus

  • Introduction to Robotics: Type of robots, Robot manipulators, Kinematic chains, Degree of Freedom, Manipulator workspace, Manipulator classification: Cartesian, Cylindrical, Spherical, SCARA and Anthropomorphic (LN 1)

  • Rigid Motion: Rotation matrix, Rotational transformation, Composition of rotations, rigid motion, Homogeneous transformation (LN 2, LN 3, LN 4)

  • Robot Kinematics: Forward kinematices, Assignment of coordinate frames: DH Convention, Inverse kinematics, Velocity kinematics: angular and linear velocities, Skew symmetric matrices, Derivative of rotation matrix, Velocity transformation, Forward velocity kinematics: Jacobian matrix, singular manipulator configuration, Inverse velocity kinematics (LN 5, LN 6, LN 7, LN 8).

  • Path and Trajectory Planning: Path planning with obstacle avoidance: Attractive and repulsive potential field, Force and torque computation, Gradient descent algorithm; Trajectory generation, polynomial trajectory, trapezoidal velocity trajectory and minimum time trajectory (LN 9, LN 10).

  • Robot Dynamics: Euler-Lagrange equation, Constrained systems, Holonomic and Nonholonomic constraints, Principle of virtual work, Euler-Lagrange equation for constrained systems, D'Alembert's Principle (LN 11, LN 12)

  • Robot Control: Independent joint control: PID Controller, Static state feedback control, Trajectory tracking (LN 13)

Robotics System Toolbox (in MATLAB)

  • Robotics System Toolbox has following features:

    • It provides tools and algorithms for designing, simulating, testing, and deploying manipulator and mobile robot applications.

    • For manipulators, the toolbox includes algorithms for collision checking, path planning, trajectory generation, forward and inverse kinematics, and dynamics using a rigid body tree representation.

    • For mobile robots, it includes algorithms for mapping, localization, path planning, path following, and motion control.

    • You can develop a functional robot prototype by combining the kinematic and dynamic models provided. The toolbox allows you to simulate your robot by connecting directly to the Gazebo robotics simulator.

  • This toolbox contains the following:

    • Robot Manipulator Modeling: You can build and model a manipulator robot (web links: Link-1, Link-2) using the function: rigidBodyTree. A rigid body tree (RBT) model is made up of rigid bodies as objects. Each rigid body has a RBT joint object associated with it that defines how it can move relative to its parent body. Please go through the different web-links in this section to know the other features.

    • Inverse Kinematics solver: Inverse kinematics (IK) determine joint configurations of a robot model to achieve a desired end-effect position. Robot kinematic constraints are specified in the rigidBodyTree robot model based on the transformation between joints. Several transformation matrices can be obtained, which is available in this web-link.

    • Motion and Path Planning: You can plan paths using different planners such as: i) rapidly exploring random tree (RRT), ii) covariant Hamiltonian optimization (CHOMP), and iii) Deep-Learning-Based CHOMP Optimizer for motion planning algorithms for manipulators. One can also generate smooth trajectories using the feature: trajectory generation.

    • Dynamics and Control: Robot dynamics brings the relationship between the forces acting on a robot and the resulting motion of the robot. This web-link covers the different elements, properties, and equations of rigid body robot dynamics. Robot manipulators are typically position-controlled devices. For joint-space control, you specify the vector of joint angles or positions, q, that tracks a reference configuration, qref. You can do this by using closed-loop control on the robot joints and using the motion model to simulate the behavior of a robot under this control.

    • Robot Simulation with Gazebo: The algorithms that you design through Robotics System Toolbox in MATLAB and Simulink for robotic manipulator can be simulated, and observed its performance in a virtual simulation environment, which uses the Gazebo Simulator. The working details can be found in this web-link.  

Robotics Toolbox (in Python)

Evaluation

  • Homework/Assignment (submission and discussion): 25 %

  • Mid-sem Exam: 35%

  • End-sem Exam: 40%

  • For Audit: You have to obtain “C” Grade for NP.

Suggested References

  • Robot Modeling and Control by M. W. Spong, S. Hutchinson and M. Vidyasagar

  • Robotics: Modeling, Planning & Control by B. Siciliano, L. Sciavicco, L. Villani and G. Oriolo