Robot Dynamics and Control
Equations of motion and the controllers that make robots move accurately: rigid-body and Lagrangian dynamics, actuators, gearing and friction, PID design and anti-windup, Routh, root locus and Nyquist stability, state feedback and observers, LQR, trajectory generation, computed torque control, impedance, admittance and hybrid force control, robust, adaptive and cascaded digital control, and a simulated two-link arm capstone.
ROB-300 Year 3 3 credits 12 lessons · 7 h 10 min
Offered by HSM University · Robotics
What you'll learn
- Compute masses, centres of mass, moments of inertia and inertia tensors of robot links, and apply the parallel axis theorem and the Newton–Euler equations
- Derive the equations of motion of manipulators with the Lagrangian method and interpret the inertia matrix, Coriolis, centrifugal and gravity terms and their properties
- Model joint actuators, transmissions, friction and flexibility, and size motors against torque–speed envelopes and RMS duty-cycle torque
- Design P, PD and PID joint controllers by pole placement and classical tuning, and handle derivative filtering, saturation and integrator windup
- Analyse closed-loop stability with Routh–Hurwitz, root locus and Nyquist methods, and evaluate gain, phase and delay margins
- Design state-feedback controllers by pole placement, test controllability and observability, and build Luenberger observers using the separation principle
- Design linear quadratic regulators, choose weights with Bryson's rule, and compute discrete LQR gains by iterating the Riccati recursion
- Generate cubic, quintic, trapezoidal and jerk-limited trajectories, synchronize joints and scale time to respect velocity, acceleration and torque limits
- Apply gravity compensation, inverse-dynamics feedforward and computed torque control, and predict the effect of model error on tracking
- Design impedance, admittance, direct force and hybrid position/force controllers, and reason about contact stability and passivity
- Improve robustness with sliding mode control, disturbance observers, adaptation and parameter identification, and implement cascaded digital control loops
- Take a robot controller from requirements through modelling, trajectory planning, control design and realistic simulation to a verified design
Course content
12 lessons · 7 h 10 min total · final exam
- 1 Rigid-body dynamics for robots 42 min · quiz
- 2 Lagrangian dynamics of manipulators 34 min · quiz
- 3 Actuator dynamics and joint modelling 36 min · quiz
- 4 PID control of a robot joint 34 min · quiz
- 5 Stability analysis of control loops 33 min · quiz
- 6 State feedback, pole placement and observers 35 min · quiz
- 7 Optimal control: the linear quadratic regulator 34 min · quiz
- 8 Trajectory generation and time scaling 37 min · quiz
- 9 Trajectory tracking and computed torque control 35 min · quiz
- 10 Force control: impedance, admittance and hybrid control 38 min · quiz
- 11 Robust, adaptive and digital control in practice 37 min · quiz
- 12 Capstone: designing and simulating a controller for a two-link arm 35 min · quiz
- E Final exam 20 questions
Requirements
Recommended background. You can start any course at any time.
Description
Equations of motion and the controllers that make robots move accurately: rigid-body and Lagrangian dynamics, actuators, gearing and friction, PID design and anti-windup, Routh, root locus and Nyquist stability, state feedback and observers, LQR, trajectory generation, computed torque control, impedance, admittance and hybrid force control, robust, adaptive and cascaded digital control, and a simulated two-link arm capstone.
Part of the Science stream, typically taken in year 3.