Control of Mobile Robots is a course that focuses on the application of modern control theory to the problem of making robots move around in safe and effective ways. The structure of this class is somewhat unusual since it involves many moving parts - to do robotics right, one has to go from basic theory all the way to an actual robot moving around in the real world, which is the challenge we have set out to address through the different pieces in the course.
Hybrid Automata
Loading...
From the course by Georgia Institute of Technology
Control of Mobile Robots
1019 ratings
From the lesson
Hybrid Systems
Meet the Instructors
Dr. Magnus EgerstedtProfessor
School of Electrical and Computer Engineering
So in the previous lecture, we saw that switches show up very naturally,
especially in robotics where we're going to switch between different modes of
operation. And what I want to do today is talk a
little bit about the models that we're going to use to describe these phenomena.
And the models that we're going to use are called Hybrid Automata.
And these are models that contain both the continuous dynamics.
Meaning to x dot part and the discrete switching logic that says oh, if you see
an obstacle you should switch to an obstacle avoidance behavior, or something
like that. Now, what this means is that the discrete
logic will be modeled as a finite state machine that moves between different
discrete states, so that's the discrete logic.
But just a finite state machine on static steroids in the sense that inside each
state we have a continuous dynamics loraking so.
Having said that, what we need first of all is the notion of a state.
So x, as before, is the continuous state of the system.
This is the physical state typically of what a robot is doing.
But then what we're going to do is we're going to add a discrete state that I'm
going to call q. And q is going to tell me which different
continuous mode I am in. So my dynamics now can be encoded not As
xdot=f(x,u) but xdot=fsubq(x,u). Where this q now tells me which mode am I
actually in. And what we're going to do is transition
between different discrete modes in a state machine where here I'm having mode
q. So really, xdot inside that mode.
Is going to, be fsubq and xdot inside this mode is going to be xdot is, fsubq
prime. Now, these are called transitions right,
so when you jump in between different discrete modes you're making transitions
between different states in the finite state machine.
Now what we need to understand is, when do we actually make these Jumps, well for
that we need something called a guard and the guard is something that checks
whether or not you should jump. And then we're also, [COUGH], excuse me,
also going to add something else which is called a reset which is going to tell you
not when you jump but where you end up after you made the transition.
So the guard condition is a condition that tells you when it is time to jump.
So let's say that I am in mode q and I want to jump to mode q prime.
Well, what I have, is I have a guard that says, if.
x belongs to this guard, let's say that this is, I'm going to switch from one
gear to another gear if my rpm is above 3,000 then I'm going to switch.
This is a guard condition that encodes when it is time to jump.
Now, when I'm jumping I may actually reset the value of my state.
For instance, if I'm dropping a ball, whoop, onto this desk, and it bounces up,
it loses energy in, in the bounce so what we're doing is we're subtracting a piece
away from the state. At each bounce.
This way of messing with the state at the moment the transitions occur is called a
reset. So, if I am going, I'm here in mode q.
[SOUND]. And then my guard is true, so I am moving
to q prime. Well, what's happening here is that x
changes for instance, x is now lets say, the old x/5.
That would be a guard that says, oh I'm getting only a fifth of the energy
somehow after this jump. So, what we need are dynamics, individual
dynamics which are called modes. We need transitions which tell us, tells
us which discrete states are we moving in between.
We need guards which tell us When we're going to make these jumps and we're, need
resets to tell us how these jumps end up effecting the state.
So if we put all of this together, we get a rather arguably messy looking thing.
But it's very rich. This is the hybrid automaton.
Singular, or hybrid automata, plural model. And the way this works is, let's
see here, we start with x equal to x not, we end up in this mode, which is q=1,
where we are evolving according to this differential equation.
Right? Okay, if in this case, x belongs to G12, then x, we're going to move, over
into this mode. And on our way. We're going to reset the state.
Now the state is here. This is my dynamics.
I'm in q2 mode. Well if, G21 becomes true, I jump this
way. But if G23 becomes true, I jump that way.
So let's say G23 became true, I jump this way.
I change my state, possibly according to this reset map.
I have a third dynamics, q=3. And then when x enters G31 guard, it
jumps back into q=1, with a new possible reset.
So this is the general model that we're going to work with.
Now, it looks a little bit messy, but it's actually really not.
In most cases, these models are going to look rather, innocent.
So, here's for example, a very very simple Hybrid system.
it's a thermostat, or it's my idea of a thermostat.
Typically, the way a thermostat works, at least cheap ones, is you set some
temperature you would like. The thermostat is on the heater is on
until you reach that temperature, or typically, go a little bit above it.
And then, it turns off, and then you cool down, and then, when you're too low, you
turn it on again. And you've heard this all.
In buildings with, with heaters, or that turn on and off.
So, what we're doing is, we're heating it up until we're comfortable and warm.
And in fact, this is my desired temperature and you typically want to add
a little slack that says that I'm going to, shut this thing off when I'm at,
whatever I want to be 70 degrees. Plus a little slack, let's say.
By the way, this is 70 degrees Fahrenheit.
For you Celsius people out there, let's say it's 20.
so now, we're above 20 Celsius, and now we're going to cool it down.
And then we're cooling it down until we're,in this case, below, well, 20
degrees Celsius, or whatever it is in Fahrenheit, - epsilon.
And the reason why we need this epsilon is.
If we didn't have it, then this guard is true.
And this is guard is immediately true. 'Cuz if t is = to t desired, we're just
going to start spinning around indefinitely here.
So, the little epsilons are important. So, this would be 1 way.
In which we will model a thermostat. There are no resets here just current
conditions because we're not magically making the temperature jump just because
we're changing the heaters from, from on to off.
Well, here's a gear shifter, right Well, we have some dynamics.
v is the velocity of the car. We're now in first gear.
This is our dynamics. u is our smashing the break of the gas
pedal down. So this is how we are driving.
When we have the RPM above some threshold then we change gears.
We switch up to second gear which is here, right.
So now we are having the second gear. If we keep pressing the gas.
So that the RPM goes up beyond, beyond some c2, then we switch to third gear.
Apparently this car only has three gears. but, we go to the third gear, and the
same thing here with a downshift. And note this here that I have c2 prime
and c1 prime. And c1 and c2 here.
This is again, because you want to build in a little bit of slack.
This is my way of hiding an epsilon inside there so we don't immediately
transition between, between gears. So this would be a hybrid atomaton gear
shifter. And I'm not writing out these dynamics.
Because the particulars aren't that important right now.
The important thing is, the switching logic and the way the guards operate.
Well finally, let's look at the behavior-based robotics system.
What I'm doing is I'm running the robot. So let's say at x is the position of the
robot. According to a go to goal behavior.
So f go to goal of x. This is me having to assign that.
When the distance between where I am and where the closest obstacle is when that's
less than some sum d, meaning I'm too close to an obstacle, then I'm going to
switch to another mode or another behavior which is Avoid obstacles.
So now I have decide my avoid obstacle behavior and it's safely taking me away
from the obstacle and then when the position of the robot is greater than
some other distance, the prime away from the obstacle then I switched to goal to
goal again. And the way this would look is, let's say
here's the obstacle. Here's the goal.
Here's my robot. [SOUND].
My robot is doing fine. Here's the ball of radius d.
Once I get in here, I'm going to switch to a goal to goal behavior.
That's going to take me away from the robot.
Let's say here is a larger circle of radius d prime and when I'm here, I'm
going to be safe and I'm going to switch to goal to goal behavior.
So this is what would happen in practice if you ran this hybrid atomaton as a
behavior based robotic system. Now, unfortunately, things aren't all
rosy in the hybrid world. And what we'll see in the next lecture is
that things may actually go wrong even the we are meaning very well when we
start building our systems. So the next lecture is a Danger:Beware
sign that we need to put up whenever we are designing hybrid systems, as opposed
to standard standalone linear time and variant systems.
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.
© 2018 Coursera Inc. All rights reserved.
