Online Lecture Series for Control Theory of Mobile Robotics

I am a robotics engineer, focusing on control and planning of mobile robots. So, i looked for a lot of online resources to get deep insight in control engineering. Control engineering (for mobile robotics) can be sub-divided into following courses:-

1. Classical Control
2. Robust Control
3. Discrete Control
4. Advance (Modern) Control
5. Non-linear Control
6. Optimal Control
7. Adaptive Control
8. Predictive Control

Why control theory is difficult for few of us ? Control theory needs a strong knowledge of Applied Mathematics as well as physical significance of each topic. For instance, “Stability”, anyone should know the definition, criteria, plot and physical meaning of mathematical equation.

Classical Control is the base of all other courses. It help us to understand different transformation(Fourier, Laplace, z-transform) and their applications. What is LTI system ? Why we only study first order linear ODE ? Why we need transformation ? Time domain vs s-domain vs frequency domain vs z-domain ? What does Stability of system means ? Different input function like Impulse, Step, Ramp and how to use them for system analysis. How Closed loop system change the dynamics of system i.e. poles ? Plots like root locus, polar plot, bode plot, Nyquist plot help us in designing a stable plant. Practical validation in MATLAB give strong intuition of control system. Later on PID, Lead-Lag Compensator, Sensitivity can easy be understood.

Once you get strong background of classical control, everything will be easy for you. In robotics, modern control play important role because of it’s state space representation which is very easy in computation. Canonical Forms (Jordan, Controllable, Observable form), Stability, Solution of state space ODE, State Transition Matrix, Controllability (State and output controllability), (kalman and Gilbert Test), Observability (kalman and Gilbert Test), Pole placement, LQR, State Observer LQE (Kalman Filter), LQG are main highlight of modern control.

Currently Optimal Control, Adaptive Control and Predictive Control are mostly used in the applied robotics.

Now a day the term “intelligent control system” is widely focused in control engineering. It basically combine reinforcement learning, data-driven control, optimization and adaptive control and useful where we are unable to model our plant as a linear system or uncertainties are high. Also, fusion of Game theory and Control theory seems interesting for multi-agent systems.

Here are free youtube video lecture series which will help anyone to understand Classical (Robust and Discrete) and Modern Control.

1. Lecture Series on Control Engineering by Prof. Ramkrishna Pasumarthy, Department of Electrical Engineering, IIT Madras.
2. Classical Control Theory by Brian Douglas.
3. Advanced Linear Continuous Control Systems by Prof. Yogesh Vijay Hote Department of Electrical Engineering IITR.
4. Control Bootcamp by Steve Brunto.

First and Third lecture series are very long but detailed course of classical and modern control respectively. Second and Fourth lecture series are summarized but give strong intuition of classical and modern control respectively.

5. Applied Engineering Mathematics by Steve Brunto will help to understand ODE and their solution, Eigen values, Taylor Series, Linear Algebra, Fourier and Laplace Transformation, Numerical Solutions of ODE, SVD, diagonalisation and MATLAB examples for each.

6. Control of Mobile Robots by Dr Magnus Egerstedt is available on Coursera completely focused on mobile robotics. But this is not for beginners, who has no clue about classical and modern control.

7. John Rossiter and Brian Douglas. He covers practical application and control system design. Optional but go through quick review of his playlists once.

Note:- Applied Engineering Mathematics is core of Control Engineering. Use MATLAB and Python for practice.

Few other sub field of control are Fuzzy Control, Neural Network Control, Fractional Control, Multi Variable Control, Sliding Mode Control, Hybrid Control, Hierarchical Control, System Identification, Networked Control System, etc.