Chevron Left
Back to Robotics: Estimation and Learning

Learner Reviews & Feedback for Robotics: Estimation and Learning by University of Pennsylvania

4.3
stars
500 ratings

About the Course

How can robots determine their state and properties of the surrounding environment from noisy sensor measurements in time? In this module you will learn how to get robots to incorporate uncertainty into estimating and learning from a dynamic and changing world. Specific topics that will be covered include probabilistic generative models, Bayesian filtering for localization and mapping....

Top reviews

SS

Apr 6, 2017

Leanring of mechanism and implementation of Kalman filter and particle filter from experiment is very interesting for me. And these method let me know more about map building in SLAM framework.

VG

Feb 15, 2017

The material is clearly presented. The Matlab exercises complement and reinforce the subject, the level of difficulty is well balanced, thanks for this great course.

Filter by:

1 - 25 of 110 Reviews for Robotics: Estimation and Learning

By Eduardo K d S

Oct 26, 2016

I wouldn't recommend this course to my worst enemy. There is 0 commitment from the TA and mentor staff. After 4 weeks of course, not a single reply from any of them in the forums.

To make matters worse, the material is very superficial and lacking, the biggest proof of that is that each module is composed of about 4 videos of 5 minutes each! How can you learn anything in 5 minutes? The topics are so complex, there is simply no way to convey their message in just about 5 minutes. I had to search a lot outside of this course to grasp something of the topics covered. Actually, I found way better explained videos on youtube for free.

The assignments of this course are poorly developed and don't reflect what is discussed in the videos well. I repeat myself, the study material was very lacking and consequently not enough for the assignments themselves. I had to spend days coding, debugging and reverse engineering the assignment files to finally be able to pass because they were wrong! They were incomplete and had wrong information. This is not a reverse engineering course... I shouldn't need to do that. Anyway, I did it because there was also no help from any TA to guide me in the right direction.

Long story short, you are way better off checking the syllabus of this course and checking videos on youtube to learn about them, you'll learn way more than with this money grabber course.

By Niju M N

Jun 20, 2016

This is course is really helpful for beginners to understand how probability is useful in Robotics.Assignments are bit tough but worth the time .

By cyril m

Jun 10, 2016

Great content but dome speakers really need to be more involved... Some videos are perfect if you want to sleep...

Also there was some mistakes in the programming assignments (more or less fixed now) and maybe a bit more details or example code would help. It was hard to make the assignments without extra help and many students had difficulties finding what to do, hopefully the forum was a real life saving tool.

The MATLAB introduction is just ridiculous: you'll need 10x more knowledge of it to do the assignments. it's like taking your first driving lesson on Monday and do Indianapolis 500 on Tuesday...

For someone like me that was not very familiar with this tool, I spend 90% of the time googling "matlab doing this or that" to find example code and making the code work instead of working on the algorithm itself.

The tutorial in the Aerial Robotics is more complete.

By Ben A

May 11, 2020

This class does a decent job of describing the theoretical / mathematical aspects of Gaussian modelling, kalman filters, occupancy grid mapping, and particle filters. However, it doesn't do such a good job with the practical / implementation details of these topics. The videos are very short, distilled down to only the essential information. I had to seek external sources for further reading to complete two of the programming assignments. Also, TA help on the forums could be quicker / more responsive.

By David A

Feb 6, 2021

This course was interesting but I think the video material was too shallow and not detailed enough. The assignment for Week 4 was extremely challenging!

By Wang

Oct 8, 2020

The course is too difficult and the class is too short to understand, I have to spend a lot of this learn the knowledge needed in other place.

By Xiaoyuan C

Dec 28, 2019

A very good introductory course about the topic. But the lecture material is not enriched enough, and the gap between lecture material and programming assignment, which in itself is very good, is a bit large, and rather difficult to handle.

By Francesca G

Jun 16, 2020

I think that this course has a great potential because, through a practical approach, it offers a nice introduction on the fundamentals of the robotic estimation. Regarding the lessons it would be very useful to have detailed references to study in deep the weekly topics and, in particular for week 2 and 4, I suggest to review the materials giving more care to key aspects and removing typos. I appreciate the focus on the assignments (that always gave me the chance to really understand the lessons) but more care in avoiding possible confusion about frames or other details related to the input data can help people to focus on the solution of the problem. I appreciate in week3 assignment the possibility to know if the solution is ok before submitting the result file.

By Matthew P

May 14, 2018

This course covers some very important techniques in modern robotics including Kalman filters, mapping, and Particle filters. However, the way that these topics are presented in this course is not very clear. The later lectures especially lack the necessary content to provide a clear understanding of advanced topics. The final assignment in particular is very poorly documented and the included instructions are a bit misleading. In addition to that, the forums seem to have been abandoned by the course instructors and are full of unanswered questions from struggling students, some of them more than a year old. This course needs some serious attention and revision. Definitely the lowest quality course of this series.

By karthik r

Oct 31, 2017

Although the course is structured properly, the lectures are horrible, explanation for kalman filter lasts couple of minutes,while in universities the topic is studied and implemented as thesis over 6 months, week 4 also throws very poor insight on particle filter, week1 and week3 were better explained. I've learnt more from youtube , The lecturers should see how Andrew Ng teaches his courses, he works through the algorithms step by step. I had to painfully finish this course to unlock the capstone project. I do not recommend this course if you are new to robotics.

By Wilmer A R

Jun 9, 2016

A lot of things to improve, specially thr learning courve is from 1 to 100 and a lot of pre knowledge need, your future public is the hobby robotics people who want to expand their knowledge, a litlle more weeks maybe two can increase the likes for the course. Check this one Control of Mobile Robots you can get an example of a good learning curve

By Barak R

Jan 22, 2021

the assignments in this course are impossible, full of errors and poorly explained.

this is a really interesting topic and most lectures are shallow and unrelated to the assignments.

errors and unexplained parts of the assignment wasted ALOT of my time for no reason

By Wahyu G

Mar 24, 2018

Pretty short course but it is really worth it if you want to learn about SLAM. Just like any other courses in this specialization, help in the forums is really minimum and the course is pretty though, so you have to spend more time to complete the course. Overall it is a great course, at least for me. Thank you for all lecturers.

By Janzaib M

Apr 4, 2017

Here I learnt all the building blocks of Probabilistic Robotics and the significance of statistical methods etc to deal with the the non-linear world.

The course content is very concise and to the point. And, the knowledge transferred is well structured.

By Louis B

Jun 30, 2019

This lecture is very useful from the perspective of approaching robotics for the first time. I recommend it!

There was a lot of effort to get back to the normal distribution I had studied before, but it was very good.

By Shuang S

Apr 6, 2017

Leanring of mechanism and implementation of Kalman filter and particle filter from experiment is very interesting for me. And these method let me know more about map building in SLAM framework.

By vincent g

Feb 16, 2017

The material is clearly presented. The Matlab exercises complement and reinforce the subject, the level of difficulty is well balanced, thanks for this great course.

By Abhilash

Jun 25, 2016

A tough course with few hours of lecture material and some good programming assignments.You will be satisfied by those assignments however .

By Adi S

Nov 23, 2019

Really good course. Engaging and relevant content. The assignments push you but test your fundamentals and you end up learning a lot.

By Vu N M

Sep 19, 2018

This is a really comprehensive course which gave me a good knowledge about Gaussian Model and Kalman Filter ...

By SHAO G

Dec 13, 2016

It's a great course. Although the assignment is little tough, you will gain a lot after completing it.

By Shubham G

Mar 3, 2018

Very succinct lectures which provides necessary foundation to learn advanced localization algorithms.

By 李鹏飞

Aug 8, 2017

It's a really great course and I learn a lot of things which helps me get started with this subject!

By Sai P K K

Aug 29, 2019

week 2 and 4 needs more information. Yet great learning experience at affordable price.

By bertal m a

Mar 5, 2020

i loved it , especially the particle filtre , i'll start using that on ros , thank you