Original subtitles

Hi and welcome to the Supplementary Materials for

the Aerial Robotics course of robotics specialization.

This series of videos will introduce you to topics in mathematics,

dynamics and controls that we'll use throughout this course.

First, I wanna briefly introduce myself.

My name is Sarah Tang and I'm a PhD student at the University of Pennsylvania.

My research interests are in developing motion planning algorithms

that will allow autonomous robots to perform tasks in cluttered environments.

Specifically, I'm interested in three types of problems.

First, I'm interested in developing algorithm for

a specific class of system called hybrid dynamical systems.

These are systems that have multiple modes of operation, each with their own sets of

dynamics and discrete events trigger transitions between these different modes.

The challenge in planning for these systems is in planning not only

the behavior of a system within a specific mode, but also in planning when and

how the system will transition.

An example of this type of system is a quadrotor with a cable suspended payload.

In this case, pick-ups and releases of the payload are the discreet events that

trigger transitions between the with and without payload modes.

In our lab, we've equipped a quad rotor with an electromagnet,

allowing it to pick up magnetic payloads and our algorithm is able to generate

trajectories that pick-up objects and release them to drop at desired locations.

Second, I'm interested in planning for higher order under actuated systems.

These systems have many degrees of freedom, but

cannot directly control all these degrees of freedom to arbitrary configurations.

Again, a quadrotor with a cable suspended payload is an example of this type

of system.

Both the quadrotor and the payload can rotate and translate in 3D space.

However, they cannot do this independently.

In fact, the trajectory of the payload determines a required configuration for

the quadrotor.

Our goal is to plan aggressive motions for the system.

For example, suppose the quadrotor comes across a window obstacle where the height

of the window is smaller than the length of the cable.

Our algorithm can generate a trajectory that contains that necessary swing to

bring both the quadrotor and the payload through this narrow gap.

Finally, I'm interested in the multi-robot planning problem.

In this problem, we are given a two more robots at sets start positions.

These robots must navigate to a set of goal locations.

However, the goal that each robot must navigate to is specified beforehand and

cannot be changed.

The goal of this problem is to generate trajectories that will bring each robot to

its goal while avoiding collisions with each other.

Throughout this course, I'll be delivering supplementary videos.

These videos will serve three purposes.

First, we hope to provide you with some technical background on concepts used

throughout the course.

Second, we'll work through some example problems that will help you with

the quizzes and the exercises.

And finally, there will be MATLAB demonstrations that will help you as you

work through your programming assignments.

I'm glad to see that you're taking this course and

I look forward to working with you.