Explore
For Enterprise
Join for Free
Log In

Arts and Humanities
Business
Computer Science
Data Science
Information Technology
Health
Math and Logic
Personal Development
Physical Science and Engineering
Social Sciences
Language Learning
Degrees and Certificates

Explore all of Coursera

Browse
Search
For Enterprise
Log In
Join for Free

Time, Motion, and Trajectories

Loading...

Robotics: Aerial Robotics

University of Pennsylvania

4.5 (1,970 ratings) | 78K Students Enrolled

Course 1 of 6 in the Robotics Specialization

How can we create agile micro aerial vehicles that are able to operate autonomously in cluttered indoor and outdoor environments? You will gain an introduction to the mechanics of flight and the design of quadrotor flying robots and will be able to develop dynamic models, derive controllers, and synthesize planners for operating in three dimensional environments. You will be exposed to the challenges of using noisy sensors for localization and maneuvering in complex, three-dimensional environments. Finally, you will gain insights through seeing real world examples of the possible applications and challenges for the rapidly-growing drone industry. Mathematical prerequisites: Students taking this course are expected to have some familiarity with linear algebra, single variable calculus, and differential equations. Programming prerequisites: Some experience programming with MATLAB or Octave is recommended (we will use MATLAB in this course.) MATLAB will require the use of a 64-bit computer.

Skills You'll Learn

Motion Planning, Robotics, Quadcoper, Matlab

Reviews

4.5 (1,970 ratings)

5 stars
1,342 ratings
4 stars
455 ratings
3 stars
94 ratings
2 stars
40 ratings
1 star
39 ratings

CG

Feb 12, 2016

Well balanced mix of theory and practical applicability. Explanation of the material is also very good.\n\nThe assignments are nicely built on the taught material to stimulate understanding.

AJ

Nov 22, 2017

Exceptional Material, not only are the concept are explained well, the supplementary material are provided for all possible requirement along side. Advance course with minimum prerequisite

From the lesson

Planning and Control

Welcome to Week 3! We have developed planar and three-dimensional dynamic models of the quadrotor. This week, you will learn more about how to develop linear controllers for these models. With this knowledge, you will be required to complete the second programming assignment of this course, which focuses on controlling the quadrotor in two dimensions. We encourage you to start working on the assignment soon. This week ends with a discussion on motion planning for quadrotors.

Time, Motion, and Trajectories9:08

Time, Motion, and Trajectories (continued)9:23

Motion Planning for Quadrotors12:04

Supplementary Material: Minimum Velocity Trajectories from the Euler-Lagrange Equations2:12

Supplementary Material: Solving for Coefficients of Minimum Jerk Trajectories5:57

Supplementary Material: Minimum Velocity Trajectories3:44

Supplementary Material: Linearization of Quadrotor Equations of Motion5:47

Taught By

Vijay Kumar
Nemirovsky Family Dean of Penn Engineering and Professor of Mechanical Engineering and Applied Mechanics

Until now, we've discussed how to control a quadrotor given a specified trajectory.

We now turn our attention to generating that specified trajectory.

Given a three-dimensional environment,

we want to be able to specify points in that three-dimensional environment and

have that quadrotor plan obstacle free trajectories in that environment.

Accordingly, we'll consider this very simple subproblem.

And this problem is as follows, imagine you have a rigid body or

a quadrotor that needs to go from a start position to a goal position.

Of course, we're also interested in orientations.

And in this problem, you may have intermediate positions that you want

the rigid body or the quadrotor to go through.

This is a very general problem, it arises in all contexts in robotics,

and it's particularly important for motion planning for quadrotors.

The general set up of this problem is as follows.

We're given start and goal positions, optionally orientations.

We might want the quadrotor to visit intermediate positions or

waypoints which can also include orientations.

In general, we'll insist that the trajectories that the quadrotors

follow are smooth because the quadrotor is a dynamical system,

it cannot follow arbitrary trajectories.

This generally translates to minimizing a rate of change of input,

as we will see soon.

We'll be particularly interested in the order of the dynamical system.

If you consider a robot with a kinematic model, in other words, you assume that you

can arbitrarily specify velocities, that's a first order system.

If you have a robot with second order dynamics,

that means that you can arbitrarily specify accelerations.

For third order systems, you should be able to control or

specify or command the third order derivative which is called jerk.

And likewise, a fourth order system involves specifying or

commanding the fourth derivative, which is called snap.

The order of the system determines the input.

If it's an nth order system, in other words,

you're specifying or commanding the nth derivative of position,

you generally have boundary conditions on the first order,

the second order, the third order, all the way up to the n-1th derivative.

To solve this problem, we want to use tools from calculus of variations.

Calculus of variations refers to a body of literature in mathematics that

deals with a very simple problem.

In this equation, you're trying to minimize an integral.

You're trying to find the best function x of t, that minimizes a functional.

The functional is the integral of L, which depends on the function x of t,

its derivative, x dot of t, and of course, time.

The goal in calculus of variations is to find the best function, x star of t.

This problem arises in several settings.

If you want to find the shortest distance path, you will use a functional

which is the integral of x dot squared or the square of velocity.

If you're trying to find the shortest time path,

a problem that arises in optics, you will let L be the simple functional 1.

Likewise, in mechanics, you usually refer to a functional

that's obtained from the kinetic energy T and the potential energy V.

In our case, we'll be interested in connecting two given points,

x specified at time 0 and x specified at time T.

We're looking for a trajectory, in fact, the best trajectory,

and we restrict ourselves to differentiable curves, and

there are, of course, many such differentiable curves.

And again, we want to find something that minimizes a suitable functional.

Independent of the functional, the optimal curve,

must satisfy what are called Euler Lagrange equations.

These are necessary conditions that must be satisfied by this optimal function.

The Euler Lagrange equations involve partial derivatives and total derivatives.

So you take the partial derivative of L with respect to x dot and

the partial derivative of L with respect to x.

You also need the total derivative with respect to time.

Let's look at the Euler Lagrange equations for the simplest setting.

Again, you're specified x at time 0, and you're specified x at time T.

Let's take the functional to be the integral of x dot squared.

In other words, the input is simply the velocity.

We're going to try to minimize the square of this input, or

the square of the velocity integrated through the entire trajectory.

If we take the Euler Lagrange equations and let the functional

L be equal to the square of x dot, you quickly find out that this

equation reduces to a very simple differential equation,

x double dot equal to 0, which you can then solve.

The resulting equation is simply the equation of a straight line.

Going back to the boundary conditions that are specified, if we substitute

the fact that you want the straight line to start at x0 at time 0, and

end at xT at time T, you can get the straight line equation.

This is the case when you have a first order system with the input

being the velocity, and it works for robots that have a kinematic model.

Again, a kinematic model is one where you can specify the velocity,

you can command the velocity.

In general, the system might be characterized by higher order dynamics.

So for an nth order system,

you might not be able to specify the velocity or command the velocity.

The system has inertia, and

you can't willy-nilly specify velocities.

For an nth order system, the input is the nth derivative of position.

In this setting, the Euler Lagrange equation looks a little more complicated.

But once again, it involves partial and total derivatives of L.

If you take n=1, as we've seen,

we get the shortest distance curve, the straight line.

N=2 corresponds to minimizing a functional that involves

the square of the acceleration.

N=3 involves minimizing the square of the jerk.

N=4 involves minimizing the snap.

Accordingly, we get the minimum velocity trajectory, the minimum acceleration

trajectory, the minimum jerk trajectory, and the minimum snap trajectory.

One thing you might ask yourself is the discrepancy

between n=1 and the other values of n.

And n=1 really corresponds to the minimum velocity trajectory,

and yet it yields us the shortest distance trajectory.

Why is that?

Explore our Catalog

Join for free and get personalized recommendations, updates and offers.

Coursera provides universal access to the world’s best education, partnering with top universities and organizations to offer courses online.

© 2019 Coursera Inc. All rights reserved.

Download on the App Store

Get it on Google Play

Coursera
About
Leadership
Careers
Catalog
Certificates
MasterTrack™ Certificates
Degrees
For Enterprise
For Government

Community
Learners
Partners
Developers
Beta Testers
Translators

Connect
Blog
Facebook
LinkedIn
Twitter
Instagram
YouTube
Tech Blog

More
Terms
Privacy
Help
Accessibility
Press
Contact
Directory
Affiliates