David Rosen Goals Overview of some of the big ideas in autonomous systems Theme: Dynamical and...

42
Dynamical Systems, Stochastic Processes, and Probabilistic Robotics David Rosen

Transcript of David Rosen Goals Overview of some of the big ideas in autonomous systems Theme: Dynamical and...

Page 1: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Dynamical Systems, Stochastic Processes,

and Probabilistic Robotics

David Rosen

Page 2: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Goals Overview of some of

the big ideas in autonomous systems

Theme: Dynamical and stochastic systems lie at the intersection of mathematics and engineering

ZOMG ROBOTS!!!

Page 3: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Actually, no universally accepted definition For this talk:

◦ Sensing (what’s going on?)◦ Decision (what to do?)◦ Planning (how to do it?)◦ Actuation & control (follow the plan)

What is a robot?

Page 4: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Dynamical SystemsHow do we begin to think about this problem?

Page 5: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

General Mathematical Framework: Dynamical Systems

Page 6: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

General Mathematical Framework: Dynamical Systems

Page 7: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Sensing and Estimation

What’s going on?

Page 8: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Sensing and Estimation How do we know

what the state is at a given time?

Generally, we have some sensors:

◦ Laser rangefinders◦ GPS◦ Vision systems◦ etc…

Page 9: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Great!

Well, not quite…◦ In general, can’t measure all state variables

directly. Instead, an observation function H : M → O maps the current state x to some manifold O of outputs that can be directly measured

◦ Usually, dim O < dim M ◦ Given some observation z = H(x), can’t determine x

!

Sensing and Estimation

Page 10: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Sensing and Estimation Maybe we can use the

system dynamics (f ) together with multiple observations?

Observability: Is it possible to determine the state of the system given a finite-time sequence of observations?◦ “Virtual” sensors!

Detectability (weaker): Are all of the unobservable modes of the system stable?

Page 11: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Sensing and Estimation What about noise?

In general, uncorrected/unmodeled error accumulates over time.

Stochastic processes: nondeterministic dynamical systems that evolve according to probability distributions.

Page 12: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

New model:

for randomly distributed variables wt and vt .

We assume that xt conditionally depends only upon xt-1 and the control ut (completeness):

Stochastic processes that satisfy this condition are called Markov chains.

Sensing and Estimation

Page 13: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Similarly, we assume that the measurement zt conditionally depends only upon the current state xt :

Sensing and Estimation

Page 14: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Thus, we get a sequence of states and observations like this:

This is called the hidden Markov model (HMM).

Sensing and Estimation

Page 15: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

How can we estimate the state of a HMM at a given time?

Any ideas?

Sensing and Estimation

Page 16: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Hint: How might we obtainfrom ?

Sensing and Estimation

Page 17: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Bayes’ Rule

Page 18: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Bayes’ Rule

Page 19: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Bayes’ Rule

Punchline: If we regard probabilities in the Bayesian sense, then Bayes’ Rule provides a way to optimally update beliefs in response to new data. This is called Bayesian inference.

It also leads to recursive Bayesian estimation.

Page 20: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Define

Then by conditional independence in the Markov chain:

and by Bayes’ rule:

Recursive Bayesian Estimation: The Bayes Filter

Page 21: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Recursive Bayesian Estimation: The Bayes Filter

This shows how to compute given only and the control input .

Recursive filter!

Page 22: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Initialize the filter with initial belief

Recursion step:

◦ Propagate:

◦ Update:

Recursive Bayesian Estimation: The Bayes Filter

Page 23: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Recursive Bayesian Estimation: The Bayes Filter

Benefits of recursion:◦ Don’t need to

remember observations◦ Online implementation◦ Efficient!

Applications:◦ Guidance◦ Aerospace tracking◦ Autonomous mapping

(e.g., SLAM)◦ System identification◦ etc…

Page 24: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Example: Missile guidance This clip was

reportedly sampled from an Air Force training video on missile guidance, circa 1955.

It is factually correct.

See also:◦ Turboencabulator◦ Unobtainium

Rudolf Kalman

Page 25: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Trajectory GenerationHow do we identify trajectories of the system with desirable properties?

Page 26: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Recap: Control Systems

Page 27: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Controllability: given two arbitrary specified states p and q, does there exist a finite-time admissible control u that can drive the system from p to q ?

Reachability: Given an initial state p, what other states can be reached from p along system trajectories in a given length of time?

Stabilizability: Given an arbitrary state p, does there exist an admissible control u that can stabilize the system at p ?

Key questions for trajectory generation

Page 28: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Several methods for generating trajectories ◦ Rote playback◦ Online synthesis from libraries of moves◦ etc…

Optimal control: Minimize a cost functional

amongst all controls whose trajectories have prescribed initial and final states x0 and x1.

Trajectory Generation

Page 29: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Provides a set of necessary conditions satisfied by any optimal trajectory.

Can often be used to identify optimal controls of a system.

The Pontryagin Maximum Principle

Lev Pontryagin

Page 30: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

The Pontryagin Maximum Principle

Page 31: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

The Pontryagin Maximum Principle

Page 32: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Can also derive versions of the PMP for:

State-constrained control

Non-autonomous (i.e., time-dependent) dynamics.

etc…

The Pontryagin Maximum Principle

Page 33: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Nothing Could Possibly Go Wrong…

Page 34: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Trajectory FollowingHow can we regulate autonomous systems?

Page 35: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

The problem

Real-world systems suffer from noise, perturbations

If the underlying system is unstable, even small perturbations can drive the system off of the desired trajectory.

Page 36: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Example: Pendulum on a Cart

Page 37: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

We have a desired trajectory that we would like to follow, called the reference.

At each time t, we can estimate the actual state of the system .

In general there is some nonzero error

at each time t.

What to do?

Page 38: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

What to do? Maybe we can find some

rule for setting the control input u (t ) at each time t as a function of the error e (t ) such that the system is stabilized?

In that case, we have a feedback control law:

Page 39: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Many varieties of feedback controllers:

Proportional-integral-derivative (PID) control

Fuzzy logic control Machine learning Model adaptive control Robust control H∞ control etc…

Trajectory Following: Feedback Control

Page 40: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

We started with what (at least conceptually) were very basic problems from engineeringe.g.,

make do this this

The Moral of the Story

Page 41: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

and ended up investigating all of this:

Dynamical systems Stochastic processes Markov chains The hidden Markov model Bayesian inference Recursive Bayesian estimation The Pontryagin Maximum Principle Feedback stabilization

and this is just the introduction!

The Moral of the Story

Page 42: David Rosen Goals  Overview of some of the big ideas in autonomous systems  Theme: Dynamical and stochastic systems lie at the intersection of mathematics.

Questions?