Your peers in the course This is your first subject at the TU München create and share summaries, flashcards, study plans and other learning materials with the intelligent StudySmarter learning app.

Get started now!

This is your first subject

Name at least 5 constraints of **trajectories in c-space**

– finite length

– Bounded curvature

– Smoothness

– Minimum length

– minimum time

– minimum energy

This is your first subject

What happens if you do weight initialization with all weights to zero?

1. There is no symmetry breaking

2. All units are going to compute the same function

3. Gradients are going to be the same

This is your first subject

Explain the difference between a **free path **and a **semi-free path**

A **free path** lies entirely in the free space F and a **semi-free path** touches the boundary of one or more obstacles.

This is your first subject

How do we know whether a configuration s in the free space?

Compute the position of the robot at that configuration in the workspace. Explicitly check for collisions with any obstacle at that position:

– if colliding, the configuration is within C-space obstacle

– Otherwise, it is in the free space

This is your first subject

Name 7 assumptions for the bug algorithm

1. Bounded world

2. Known global goal

3. Measurable distance from point x to any point y: d(x,y)

4. Unknown obstacles

5. Robot has local sensing

6. Robot is tactile (bump sensor)

7. Robot can measure travel distance

This is your first subject

What is the problem with **complete algorithms **and **heuristic algorithms ** in path planning tasks?

Complete Algorithms are **slow**

Heuristic Algorithms are **unreliable**

This is your first subject

What is the biggest challenge with attractive and repulsive fields?

You can create local minima which would stop the robot

This is your first subject

What does the Kalman gain calculate?

The Kalman gain calculates the proportion of measuret variables and predicted variables for the new prediction

This is your first subject

What are the three assumptions of CNNs?

1. nearby pixels are correlated

2. weight sharing -> interesting features all over the place

3. Translational invariance

This is your first subject

Name 3 classic path planning approaches

Roadmap

Cell decomposition

Potential field

This is your first subject

What is the difficulty with classic roadmap approaches

Running time increases exponentially with the dimensions of the c-space

several variants of the path planning problem have been proven to be PSPACE-hard

This is your first subject

What happens if you do weight initialization with big random numbers? (10 layers with 500 neurons each, tanh and gaussian input)

Output saturates quickly because the tanh function saturated for big numbers (either -1 or 1)

For your degree program Robotics, Cognition, Intelligence at the TU München there are already many courses on StudySmarter, waiting for you to join them. Get access to flashcards, summaries, and much more.

Back to TU München overview pageStudySmarter is an intelligent learning tool for students. With StudySmarter you can easily and efficiently create flashcards, summaries, mind maps, study plans and more. Create your own flashcards e.g. for This is your first subject at the TU München or access thousands of learning materials created by your fellow students. Whether at your own university or at other universities. Hundreds of thousands of students use StudySmarter to efficiently prepare for their exams. Available on the Web, Android & iOS. It’s completely free.

Best EdTech Startup in Europe

1## Learning Plan

2## Flashcards

3## Summaries

4## Teamwork

5## Feedback