Alla tentafrågor Flashcards

(26 cards)

1
Q

What can a CW radar measure?

A

It can measure relative speed using Doppler effect, and direction if rotating, but not distance.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Why can’t CW radar measure distance?

A

Because the signal is continuous and there is no timing reference to calculate distance.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What can a pulsed radar measure?

A

It can measure distance by using the time delay between sending and receiving a pulse.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What additional information can pulsed radar give?

A

By comparing multiple pulses, it can show if an object has moved (motion over time).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What can an FMCW radar measure?

A

It can measure both distance and speed using beat frequency and Doppler shift.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is forward kinematics?

A

It calculates the robot’s position from joint angles or motor inputs.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is inverse kinematics?

A

It finds the joint angles or inputs needed to reach a desired position.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is the differential drive model?

A

It models a robot with two independently driven wheels to calculate its movement.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

How can we estimate a robot hand’s pose without encoders?

A

By using motion capture, visual odometry, IMU, or GPS/RTK-GPS.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is the fourth variable in GPS estimation?

A

The receiver’s clock offset.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Why do we estimate the clock offset in GPS?

A

Because the receiver’s clock is not synchronized with the atomic clocks in satellites.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

How do we solve GPS equations?

A

By forming pseudo-range equations, linearizing them with Taylor expansion, and solving using least squares and Newton’s method.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What are sources of GPS error?

A

Multipath, atmospheric delay, clock errors, and signal blockage.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What does a perspective camera model do?

A

It projects a 3D point onto a 2D image using perspective projection.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What are intrinsic camera parameters?

A

Focal length, principal point, and skew – they describe the internal geometry.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What are extrinsic camera parameters?

A

Rotation and translation – they describe the camera’s pose in the world.

17
Q

Why is tomography with sparse angles difficult?

A

Because it is an ill-posed problem with many possible solutions.

18
Q

How does regularization help in tomography?

A

It adds constraints to stabilize the solution and reduce noise.

19
Q

What does Tikhonov regularization do?

A

It penalizes large values and makes the solution smooth.

20
Q

What does Total Variation (TV) regularization do?

A

It keeps edges sharp and favors flat areas.

21
Q

What does Non-Negative Least Squares (NNLS) do?

A

It constrains the solution to non-negative values.

22
Q

What is a IMU

A

IMU:n (Inertial Measurement Unit) sitter monterad inuti robotens kropp, ofta nära centrum av roboten (ex. nära tyngdpunkten), för att ge så stabila och representativa mätningar som möjligt av rörelsen.

Sensor som mäter acceleration (via accelerometer), rotationshastighet (via gyroskop), och ibland magnetfält (via magnetometer). Används för att uppskatta robotens rörelse och orientering. IMU-data är relativ och kan integreras för att få läge, men riskerar att driva över tid (ackumulerat fel). Vanligt i sensorfusion, t.ex. i Kalmanfilter.

23
Q

Visual Odometry

A

Teknik där roboten använder kamerabilder för att förstå hur den har rört sig. Den tittar på hur omgivningen förändras mellan bilder för att räkna ut sin rörelse (t.ex. hur långt den gått och i vilken riktning). Det är som att roboten ser världen och gissar sin förflyttning utifrån det. Ger relativ position, och används ofta tillsammans med IMU för bättre noggrannhet

24
Q

Dead reckoning

A

Dead reckoning innebär att roboten utgår från en känd startposition och beräknar sin nya position genom att summera sina egna rörelser över tid, till exempel med hjälp av hjulsensorer (encoders) och en IMU. Det är en intern metod som inte kräver någon information om omgivningen, men nackdelen är att små fel i mätningarna snabbt kan byggas upp till stora positionsfel (så kallad drift).

25
Motion capture
som är en extern metod där flera kameror runt om i ett rum spårar markörer på roboten och exakt mäter dess position i 3D. Det kräver att miljön är uppsatt med utrustning, men ger mycket noggranna och absoluta positionsdata – ofta använt i labb för att utvärdera andra navigeringsmetoder.
26