Chapter 13: Vector Functions

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement not replaced the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

Welcome to the deep dive, your shortcut to being well informed.

Today we're embarking on a journey into the very language of movement.

Imagine a plane banking elegantly through the sky.

Or maybe a planet's dead fastly circling its sun.

Or even, you know, the incredible unseen dance of DNA strands coiling up inside a cell.

Exactly.

All these motions, from the really big scale down to the microscopic, they trace out paths in space.

So the question is, how do we actually describe those paths mathematically?

And not just where something is, but how it's moving, how it's bending, maybe even twisting.

That's precisely what we're here to unpack.

Up until now, a lot of the math toolkit might focus on motion in a flat two dimensional plane.

Right, like on a piece of paper.

But well, the universe we live in is vibrantly three dimensional.

So this deep dive is all about equipping you with the crucial insights to describe objects navigating in full 3D space.

Our mission today is to pull out the most powerful concepts from the world of, well, vector functions.

You'll gain a concise, yet pretty comprehensive understanding of how these functions work, how calculus applies to them, and honestly, some of their astonishing real world applications.

Get ready for hopefully a few aha moments that can make even complex motion feel a bit more concrete and understandable.

Okay, so let's dive in.

To get a handle on motion in 3D, we first need to understand vector functions and what we call space curves.

Now you're probably familiar with functions where you put in a number, say x, and you get another number like fx equals Exactly.

Basic function stuff.

With vector functions, the input is still often a single number, let's call it t.

And t typically represents time in a lot of applications.

But here's the key difference.

The output isn't just a number.

It's a vector.

It's a vector, yeah.

So in three dimensions, this function, ret, might give us a vector with x, y, and z components.

We often write it like

retft, gtht.

Ah, so those fg and h bits.

Those are just regular functions, like the ones we already know.

Precisely.

They're just standard real -valued functions, each telling you how the x, y, or z coordinate is changing with respect to t.

So it's like three functions bundled together.

Okay, that makes sense.

So if you're given one of these vector functions, how do you figure out its domain?

Like where is it actually defined?

It's pretty logical, actually.

You just need to make sure that all of its component functions, the f, g, and h, are well -defined for a given t.

Meaning?

Well, you know, you can't take the logarithm of zero or negative number, right?

Or the square root of a negative number.

So if gt involves ln3t and ht involves square t, like in example one from the text.

Then t has to be less than three because of the logarithm, and t has to be zero or positive because of the square root.

Exactly.

So you combine those restrictions.

The domain for the whole vector function r key would be where all components are happy.

In that case, it's t values from zero but not including three.

So the interval is 0, 3.

Got it.

You find the domain where every single part works.

Yeah.

And does that same idea apply to limits and continuity,

component by component?

Absolutely.

It's wonderfully consistent.

If you want the limit of rt as t approaches some value, you just find the limit of ft, the limit of gt, and the limit of ht separately if they all exist.

Then the limit of the vector function exists, and it's just the vector made of those individual limits.

Like in example two, where they find the limit as t goes to zero.

Precisely.

You look at each component, find its limit like sin2 goes to one, and assemble the resulting vector.

And continuity works the same way.

If f, g, h are all continuous at a point, then the vector function rt is continuous thereto.

Okay, this naturally leads us to the idea of a space curve.

What is that exactly, and how does the vector function kind of draw it?

Great question.

Imagine rcress as representing a position vector.

So it's an arrow starting at the origin and pointing to a specific location in 3d space at a particular time state.

Okay.

Now as t changes, the tip of that vector moves.

It traces out a path, that continuous path traced by the tip of the position vector or crimp.

That's what we call a space curve, often labeled c.

Ah, so it's like leaving a trail, like a sparkler moving through the air at night.

Perfect analogy.

That trail of light is your space curve.

And often we describe these curves using parametric equations, x, ft, y, g, t, z equals ht.

It's the same information as the vector function, just written slightly differently.

So something pretty simple like rt equals one plus t, two plus five t, minus one plus six t.

That actually describes a straight line in 3d.

Exactly.

That specific function describes a line that passes through the point one, two, negative one, one, t, zero, and moves parallel to the direction vector one, five, six.

Really fundamental.

And then there are more complex shapes, right?

Like the helix.

Oh yeah, the helix is a classic and really important example.

Think of a coiled spring or maybe a spiral staircase.

The function rt, two's t, sin tt, gives you this shape.

How does that work?

Well, look at the x and y components, sin two t and sin t.

You know, that cos two plus sin two equals one, right?

So bites two plus y two is one.

This means the curve always lies on the surface of the cylinder of radius one, centered along the z axis.

Okay, it's trapped on the cylinder.

And then the z component is just z o t.

So as p increases, the z value increases steadily.

It just climbs upwards as it goes around the cylinder.

Voila helix.

And this isn't just a mathematical toy.

You mentioned DNA earlier.

Absolutely.

The double helix structure of DNA is probably the most famous real world example.

That precise, repeating spiral geometry is fundamental to how it stores genetic information.

You also see helices and springs, screws, all sorts of things.

It's a really common form in nature and engineering.

Wow.

Okay.

And vector functions can even describe curves formed by, say, the intersection of two surfaces.

Yes, exactly.

Imagine you have a cylinder and maybe a plane slicing through it at an angle.

The curve where they meet can be described by a vector function.

Finding that function might take a bit more algebra, like in example six, but it's doable.

And for really complicated curves, like some of those knots or spirals shown in the figures.

That's where technology really helps.

Trying to visualize something like a tree foil knot or a toroidal spiral just from the equations is?

Well, it's tough.

Computer graphing software lets you rotate them, view them from different angles, see their projections onto the xi plane or yz plane.

It makes understanding their 3D structure much, much easier.

Okay.

So we've got a feel for what these vector functions are and the curves they trace out.

Now let's bring in the calculus.

How do derivatives and integrals work here?

It flows quite naturally from single variable calculus, actually.

The derivative of a vector function, R3t, is defined using the same limit definition as before, but the geometric meaning is super important.

Which is?

R3t is the tangent vector to the space curve C at the point corresponding to D.

Think of it as an arrow that's just kissing the curve at that point, pointing in the exact direction the curve is heading at that instant.

So if Rt is position, then Rt is velocity.

Precisely.

It tells you both the direction of motion and how fast you're going in that direction.

Its magnitude is the speed.

It's parallel to the tangent line to the curve.

And calculating it.

Is it complicated?

Actually, no.

That's the really nice part.

To find the derivative Rt, you just differentiate each component function f t, g t, and h t separately with respect to t.

So it's like doing three calc A derivative problems.

Pretty much.

You find f t, h t, and bundle them back into a vector, f t, a t, t, h t.

Example one shows this for a specific function.

Straightforward.

Okay.

And I saw something about a unit tangent vector.

Right.

The unit tangent vector, usually written tt.

You get this by taking the tangent vector Rt and dividing it by its own magnitude, Rt.

Why do that?

Because t then has a length of exactly one.

It purely tells you the direction of the curve at that point, stripping away any information about the speed.

It just points the way, essentially.

It's really useful for analyzing the curves geometry itself.

And the rules for differentiation, like the product rule, chain rule, do they still work?

Yes.

And that's another great thing.

Most of the familiar differentiation rules carry over beautifully to vector functions.

There are rules for sums, scalar multiples, dot products, cross products, and the chain rule.

It makes manipulating these functions feel quite natural if you're property where if the magnitude is constant, the derivative is orthogonal.

Ah, yes.

Theorem four, that's a neat one.

If the magnitude is constant, which means the tip of the position vector is always the same distance from the origin, so the curve lies on a sphere centered at the origin, then its derivative Rt, the tangent velocity vector, is always orthogonal perpendicular to the position vector Rt.

Why?

Think about it geometrically.

If you're moving on the surface of a sphere, your velocity at any point must be tangent to the sphere, and any line tangent to a sphere is perpendicular to the radius at that point.

Rt is like the radius, Rt is tangent.

It's a really elegant geometric insight revealed by the calculus.

Cool.

Okay, so derivatives give us tangent, vectors, velocity.

What about integrals?

Same.

Nice pattern.

To integrate a vector function, whether it's an indefinite integral or a definite integral, you just integrate each component function separately.

So integrate fd, integrate gt, integrate ht, and put them back in a vector.

Exactly.

Just like differentiation, but in reverse.

Example four shows finding both indefinite and definite integrals this way.

It's very consistent.

All right, so we can do calculus with these things.

Now, how do we use that to really understand the behavior of the curves?

Things like how long they are or how sharp they bend?

Perfect transition.

That brings us to arc length and curvature.

Arc length seems pretty straightforward, just the length of the curve between two points, like measuring a piece of string.

Exactly like measuring a piece of string, even if it's all twisted up in 3D space.

The formula to calculate the length l of a curve traced by Rt from ta to tdb involves an integral, specifically l integral from a to b of rt dt.

Let's unpack that.

Rt is the velocity vector, so the magnitude.

The magnitude of the velocity vector, which is just the speed.

Ah, so you integrate the speed over the time interval to get the total distance traveled along the curve.

That makes perfect sense.

It does, right.

To compute distance, integrate speed.

It's a familiar idea, now applied rigorously to curves in space.

Example one calculates the length of one turn of the helix.

And there's also an arc length function.

Yes, it's t.

This function measures the length of the curve starting from some fixed point, say at ta, up to the point corresponding to t.

So st integral from a 2t of ru du.

Okay, why is that useful?

It gives us a way to reparametrize the curve.

Instead of describing your position based on time t, you can describe it based on how far you've traveled along the curve, s.

This is considered a more intrinsic way to describe the curve shape, independent of how fast you happen to be moving along it.

Example two shows reparametrizing the helix using arc length.

Interesting.

Okay, moving from length to bending.

Curvature.

What exactly is that measuring?

Curvature, usually written with the Greek letter kappa puk, measures how quickly the curve changes direction at a specific point.

So a really sharp turn would have high curvature.

Exactly.

Think of a hairpin turn on a mountain road.

Very high curvature.

A long, straight stretch of highway has essentially zero curvature.

It's not changing direction at all.

And circles.

A small, tight circle bends very sharply, so it has large curvature.

A huge circle, like the Earth's orbit, feels almost flat locally, so it has very small curvature.

The formula DTD captures this.

It's the magnitude of the rate of change of the unit tangent vector t with respect to arc length s.

DTD is change in direction per unit of length traveled.

Okay.

Are there easier ways to calculate it?

Yes, thankfully.

There are formulas involving the derivatives with respect to t.

The two most common are TTR, and often more practical, key TRT3.

That second one uses the cross product of velocity and acceleration.

Okay, formulas exist.

And you mentioned the curvature of the circle with radius a is just 1a.

Yep, that's a fundamental result shown in example 3.

Makes perfect sense.

Bigger radius a, smaller curvature 1a.

Example 5 calculates it for a parabola, showing how the curvature decreases as you move away from the vertex, the parabola gets flatter.

So t tells us the direction of the curve.

What tells us the direction the curve is turning?

Excellent question.

That's the job of the principal unit normal vector, usually just called the normal vector nt.

It's defined by taking the derivative of the unit tangent vector pt and then normalizing it, dividing by its magnitude.

So nt, Tt, and which way does it point?

It always points towards the inside of the curve in the direction the curve is bending.

If you're driving around a curve, n points towards the center of the turn.

It's always orthogonal to t.

Okay, so we have t forward and n

inward turn direction.

Is there a third direction?

There is.

It completes the 3D picture.

It's the binormal vector bt, and it's defined very simply as the cross product of the first two.

bt equals txnt.

Cross product.

So b must be orthogonal to both t and n.

Exactly.

t, n, and b form a set of three mutually perpendicular unit vectors at every point on the curve.

This is called the t and b frame or frenet frame.

Think of it as a local coordinate system that travels along the curve with you.

Like an airplane's roll, pitch, and yaw axis.

Sort of analogous, yeah.

It provides a complete orientation in space relative to the curve itself.

It's incredibly important in differential geometry, physics, computer graphics, understanding how things like spacecraft maneuver.

Example six finds t, n, and b for the helix.

Okay, t, n, b.

Now what about this osculating stuff?

Sounds fancy.

It does.

Osculating comes from the Latin word for kissing.

We have the osculating plane and the osculating circle.

Kissing plane.

The osculating plane at a point p on the curve is the plane determined by the tangent vector t and the normal vector n at that point.

It's the plane that best fits the curve right near p.

It sort of kisses the curve, containing its immediate direction in the direction it's starting to bend.

And the osculating circle.

That's the circle that lies within the osculating plane and best approximates the curve at that point p.

It shares the same tangent vector t, the same normal vector n, and crucially, the same curvature as the curve at that point.

So it's the circle that fits the bend perfectly right there.

Exactly.

And its radius is simply the reciprocal of the curvature.

Radius world one.

A high curvature, sharp bend, means a small osculating circle.

A low curvature means a huge osculating circle.

Example nine finds it for a parabola at its vertex.

Okay, one more geometric property.

Torsion.

If curvature is about bending,

torsion is about.

Twisting.

Torsion, often denoted by the Greek letter tau, measures how much the curve twists out of its osculating plane.

How does it leave that kissing plane?

Precisely.

Think about a flat curve, like a circle drawn on paper.

It bends, has curvature, but it stays entirely within its plane, the paper.

So its torsion is zero.

But the helix.

The helix is constantly twisting upwards as it spirals.

It doesn't stay flat.

It has non -zero torsion.

In fact, the helix we looked at, cos t, sin t, t, has constant torsion, as shown in example 10.

Torsion quantifies that out -of -plane twisting motion.

It's the final piece needed to fully describe the 3D shape of the curve locally.

Wow, okay.

Courpture for bend, torsion for twist.

Got it.

So now, let's connect all this geometry back to actual motion.

Velocity and acceleration.

Yes.

This is where vector functions really shine in physics and engineering.

The connection is incredibly direct and powerful.

If r t is the position vector of a particle at time t.

Then r t is its velocity vector, v t.

Right.

And the magnitude of that velocity, v t, is the particle speed.

And the derivative of the line.

Is the acceleration vector, a t.

So a t, v, it's the second derivative of the position vector.

It's important to remember though, right, that velocity and acceleration depend on how you travel the curve, not just the curve itself.

Absolutely crucial distinction.

The arc length, the curvature, the torsion, those are geometric properties of the curve itself, intrinsic to its shape.

But velocity and acceleration depend on the specific parameterization, how fast you're moving along that curve.

Whether you're speeding up or slowing down.

Like driving on a road.

The road's bends are fixed, but your speed and acceleration change depending on how you press the pedals.

Perfect analogy.

The math reflects that distinction clearly.

And since integration reverses differentiation, if we know the We can find the velocity by integrating a t.

And if we know the velocity, we can find the position r t by integrating v t.

You'll need initial conditions, like starting position and starting velocity, to pin down the constants of integration, like in example three.

Which leads us straight to the heart of classical mechanics, Newton's second law.

In its full vector glory, f t equals m a t.

The net force vector acting on an object equals its

acceleration vector.

This is huge.

It connects the geometry of motion acceleration to the physical causes of motion forces.

And it lets us analyze things like circular motion.

Definitely.

Example four uses ephema to show that an object moving in a circle at a constant angular speed requires a force directed towards the center of the circle, this intripidal force.

The math clearly shows f points opposite to r.

Okay, let's talk about another classic application, projectile motion.

Throwing a ball, firing a cannon.

Assuming we ignore air resistance, the only significant force acting is gravity, which pulls straight down.

Using ephema, where f equals zero zero zero m h m g, zero if y is vertical, or maybe f equals zero zero zero, this is a vertical, and integrating twice.

You can derive the exact position function r t.

And what shape does it trace?

It always traces a parabola.

Vector calculus proves this fundamental result.

It also lets us calculate things like the maximum height, the range, how far it travels horizontally, and interestingly, that the maximum range for a given initial speed occurs when the launch angle is 45 degrees.

Example six even works through a specific scenario, firing something off a cliff, finding where it lands and how fast it's going on impact.

Exactly.

It demonstrates the practical predictive power of these equations.

Now, you mentioned acceleration.

Can we break it down somehow relative to the curve?

Yes, this is a really insightful decomposition.

The total acceleration vector a a can be split into two orthogonal components,

one tangent to the curve and one normal to the curve.

So a equals a t t plus a n n, using those unit vectors t and n we talked about.

Precisely.

A t is the tangential component of acceleration and n is the normal component of acceleration.

What do they represent physically?

A t tells you the rate at which the object's speed is changing.

It's simply v, the derivative of the speed v, v equals v.

If a t is positive, you're speeding up.

If negative, slowing down.

If zero,

constant speed.

Ending?

A n is responsible for the change in the object's direction.

It turns out that an n equals uv v2, where two is the curvature and v is the speed.

So the normal acceleration depends on how sharp the curve is and how fast you're going v2.

Exactly.

Think about taking a turn in a car.

If the turn is very sharp, large call, or if you take a very fast large v, you feel a stronger push towards the curve.

That feeling is related to the normal acceleration needed to change your direction.

And importantly, the acceleration vector a always lies in the plane formed by t and n, the osculating plane.

Yes.

Since a t t plus a n, there's no component in the b by normal direction.

Acceleration only involves changing speed along the path or changing the direction within the plane of the curve's immediate bend.

It doesn't inherently cause motion out of that plane.

That's Kepler's laws of planetary motion.

This is really one of the crowning achievements demonstrating the power of calculus and vector functions.

Johannes Kepler, back in the early 1600s, painstakingly analyzed astronomical data.

Just observations, right?

No calculus yet.

Right.

Purely empirical.

He deduced three laws describing how planets orbit the sun, but it was Sir Isaac Newton, later in that century, armed with his laws of motion, his law of universal gravitation, and the tools of calculus, including vector concepts.

He was able to prove Kepler's laws mathematically.

Exactly.

Starting from FMA and the inverse square law of gravity, Newton could derive Kepler's observational laws.

It was a monumental validation of his theories and showed that the same physics governs falling apples and orbiting planets.

So what are the laws?

First, a planet revolves around the sun in an elliptical orbit with the sun at one focus of not necessarily a circle.

Second,

the line joining the sun to a planet sweeps out equal areas in equal times.

This means the planet moves faster when it's closer to the sun and slower when it's farther away.

Third,

the square of the orbital period, the time for one full revolution,

is proportional to the cube of the length of the semi -major axis of the elliptical orbit.

Basically, planets farther out take much longer to orbit.

And the proofs rely heavily on the calculus of vector functions we've been discussing, analyzing RT, VT.

Conservation laws derive from them.

Absolutely.

The text outlines the proof for the first law, showing how the elliptical path emerges directly from the mathematics of gravity and vector calculus.

It's a beautiful demonstration of how these concepts unlock the workings of the cosmos.

And there you have it.

We've really journeyed through the world of vector functions, seeing how they give us this incredibly elegant language to describe everything from simple lines and helices to the arc of a baseball and even the orbits of planets.

These tools let us precisely quantify how curves bend, how they twist, and how objects actually move through three -dimensional space.

It's a powerful framework.

You've now got the essential building blocks, hopefully, to analyze motion and curves in a whole new dimension.

So the next time you see a roller coaster doing a loop -the -loop, maybe you hear about a satellite's trajectory, or even think about that DNA double helix.

You'll have a deeper appreciation for the mathematics underneath it, all the vector functions, the derivatives, the curvature, all working together.

So here's a final thought.

Consider how these mathematical descriptions, which really grew from just extending the idea of functions to output vectors,

end up revealing such deep underlying order and predictable beauty in the universe.

From the tiniest biological structures to the grandest celestial dances,

what other hidden patterns, what other secrets of nature, might these elegant mathematical tools eventually help us uncover?

A warm thank you from the Last Minute Lecture Team for joining us on this deep dive.