Chevron Left
Robotics: Computational Motion Planning に戻る

ペンシルベニア大学(University of Pennsylvania) による Robotics: Computational Motion Planning の受講者のレビューおよびフィードバック

4.3
991件の評価
253件のレビュー

コースについて

Robotic systems typically include three components: a mechanism which is capable of exerting forces and torques on the environment, a perception system for sensing the world and a decision and control system which modulates the robot's behavior to achieve the desired ends. In this course we will consider the problem of how a robot decides what to do to achieve its goals. This problem is often referred to as Motion Planning and it has been formulated in various ways to model different situations. You will learn some of the most common approaches to addressing this problem including graph-based methods, randomized planners and artificial potential fields. Throughout the course, we will discuss the aspects of the problem that make planning challenging....

人気のレビュー

FC

2018年11月27日

The course was challenging, but fulfilling. Thank you Coursera and University of Pennsylvania for giving this wonderful experience and opportunity that I might not experience in our local community!

AD

2018年7月2日

The topic was very interesting, and the assignments weren't overly complicated. Overall, the lesson was fun and informative , despite the bugs in the learning tool(especially, the last assignment.)

フィルター:

Robotics: Computational Motion Planning: 176 - 200 / 248 レビュー

by Prabin K R

2018年8月13日

awesome!!!

by Deep P

2021年7月13日

I would like to thank Coursera team, university of Penn and Prof. CJ Taylor for providing this course. Please take this as a constructive feedback and not a complain. I personally felt that lectures were too short and didn't do justice to the topics for all 4 weeks. Even though lectures were crisp and to the point for learning the algorithm, still I feel that more comprehensive knowledge about the topics should be shared. For ex- applications of these algorithms. Also a little more focus on implementation part please. It seemed that Prof. Taylor was screen reading the lecture content. I was very disappointed when I realized this (in week1 only). Unlike other courses where instructor engages with students as if they are really talking to us, this felt plain. As for the assignments, for week1 and 3 the pseudo code displayed in the lecture video wasn't tested in the assignment. It was more like complete the code and make it working rather than program the core steps of pseudo code. To conclude, this course needs some improvement but crucial ones.

by Matthew D

2021年8月22日

The lecture material is excellent, and Dr Taylor's passion really shines through when presenting the material. Although I was familiar with most of the topics presented, and well versed in Matlab, the assignments were a struggle as there is no access to Matlab online as there was with the aerial robotics module. The trouble is that some of the code is not copy-paste compatible with octave, and without a proper debugger it ended up taking much longer than it should have ironing out very basic bugs. Referring to the week 4 assignment, It is not reasonable to expect one to complete this assignment without even being able to plot the output to see what is going wrong. The assignments are structured such that a unique solution is required, and this does not encourage one to really delve deeply into the material. I suggest the course staff review the material in 'Modern Robotics', as the assignments such as the RRT method were in my opinion done exceptionally well there.

by Sj

2016年3月13日

Overall decent course.

This course focused less on the theory aspects in the course videos, which bothered me a lot considering I am paying for it. But the explanations were still good for those algorithms.

The assignments were good as well. I liked how they made us work on them instead of the first course where we were mostly tuning parameters. Hopefully MOOCs start having challenging assignments too.

The instructor explained really well too!

I didn't really end up visiting the Discussion Forums for this course at all. So can't comment on the participation from other students or TAs.

Future Advice -

Considering how other courses offer about 1-2 hours of course videos, I think this course could offer a lot more. One assignment problem focusing on one algorithm, while having other challenging algorithms taught in those videos to be left for our own implementation would help students a lot more i believe.

by Glenn B

2016年3月8日

The material is interesting, however there is not enough information provided by the course to effectively implement the algorithms in the allotted time of each week's assignments. It relies on deferring to external reading materials as primary sources, and these resources were not specified in advance to secure copies in a timely manner.

Additionally, there is a big disconnect between the knowledge provided by the weekly material and what is required to easily do the programming assignments in the suggested time of 3 hours.

Overall the course material needs to provide more background material to be more effective in delivering the knowledge expected each week. This may be an artifact of trying to cram what other online course provide in 7-10 weeks down into 4 weeks. If the intention is to give a "flavor" in 4 weeks, then the material needs to be distilled down into more of a cookbook format.

by Manoj R

2018年6月2日

Very good overview of basic topics in Computational Motion Planning. The material is nicely and intuitively presented in short video lectures and is a rapid overview of the first 5-6 chapters in the book by Choset et. al.

Some of the assignments were too simple and required us to work on the non-critical parts of the problem. For example, only focusing on descending along gradients of artificial potential fields, instead of constructing them and seeing the effect of different types of potentials.

Also, a dominant portion of my time was spent fighting the autograder. There are tips on the forums to help deal with this but sometimes an almost-complete solution is presented by some of the earlier students in a frustrated attempt to get help with the autograder.

Many of these autograder related problems have not been addressed for many months.

by Ajinkya K

2016年3月6日

Although the course covers interesting subject areas, I feel like the various topics should have been explored to a greater depth. I understand that someone with lesser background in the relevant areas might not agree with me. But overall, I felt slightly underwhelmed by the course.

Also, the skeleton of code provided for the assignments had minor errors and the instructions for assignments were sometimes ambiguous or even incorrect as compared to what was actually required of the code. But these minor issues will most likely get resolved in subsequent offerings of the course.

by Eduardo K d S

2016年8月3日

The course is ok, it touches on some interesting topics and it serves its purpose as an introductory course. Unfortunately more interesting topics are only briefly mentioned at the end of the last video. I also think the assignments can be improved, some assignments lack documentation, one of them had a coordinate system swapped from what was shown on screen and the evaluation of some assignments are quite tight, even if you have it working, unless you deliver exactly as it is expected you will fail, not to mention what is expected is sometimes blurry.

by Benjamin K

2017年12月6日

If the course had the same information and effort as week1 over all 4 weeks i would gave 5 stars, although the assignments are pretty good and I learnt something new, however the assignments are fun but the grader is annoying as the single error output is..... something is wrong... try again?!

by Keng-Hui W

2016年5月8日

三顆星都是給老師

兩顆星都是扣在作業和TA上面,TA完全隱形狀態。

作業提供skeleton的code非常的糟糕,不直覺的實作方式加上詭異的coding style

讓這門課程的表面上難度不高,但即使你已經課堂講解的演算法,你仍舊可能會浪費很多生命花在理解code。

再加上那有跟沒有差不多的測試回饋,你幾乎無法從中獲得任何修改你code的提示,就像是跟你在寫C時遇到沒有compile錯誤的bug一樣,痛苦萬分。

就像:

你跑了10個測試,通通都錯喔,錯在哪裡自己思考吧!

不過念別人的code還是有好處啦,避免自己以後也寫出這種code...

by Daniel W

2016年6月5日

First course of this specialization was really GREAT, byt this course disappoints.

Of course, there are some interesting topics, but the form of the course is way lazier. Videos are short, there is small amount of additional materials,

by Emiliano J B O

2016年2月26日

I think that the theory was very poor in sense of the videos were very short and with little content. The topics that we've seen were difficult to learn by itself, and a better explanation could be very useful in practice.

by Rafay A K

2016年3月14日

The first two assignments certainly tested knowledge of the subject however the last two assignmnets were lacking. Good course that does what it is supposed to do. More feedback from test cases would be very helpful.

by Антон Л

2017年10月30日

Lectures are small, assignments are poor quality, you will probably solve matlab coding problems and try to adapt your working (at least your visual inspection says so) solution bad autograder without ANY FEEDBACK

by Julius S

2016年6月6日

Great course! but there was too little content!!!! Double it !! Or double the coursework! make us do more work! Also, tell people to use 'parfor' to speed up the computations.

Otherwise, great course!

by Mike Z

2016年5月23日

Very good introduction course for motion planning. Could be better if there is more interactions with the TAs. Also the matlab assignments have some minor mistakes which takes time to figure it out.

by Emeka E

2016年3月11日

I think there is need to provide clearer instructions on how to get the programming assignments done. The course content is good, but doing the programming assignments needs to be more clarified.

by Alex M

2016年3月13日

Most of the homework assignments aren't graded correctly out of the box and have errors. Also, only specific solutions are selected. Otherwise it's great material at a good pace.

by Lucas H C S

2017年9月23日

Matlab online makes this course activities expensive in time, and some algorithms are not explained on the classe or texts, so you need to search a lot.

by Taimoor D K

2018年8月27日

Course content is very good however topics should be covered in much detail. Frequent bugs in programming assignments is also a concern.

by 李晨曦

2017年6月24日

Too few details of the algorithms are provided. The assignment are too simplified to help students develop a good grasp of the contents.

by Luke J

2016年9月13日

Not much content covered in course, especially compared to Aerial Robotics. No real great sense of achievement on completion.

by Unnat A

2019年7月1日

The lectures should cover more in depth theory to better explain the concepts before giving such challenging assignments.

by Rayad K

2016年3月9日

In comparison to the first course this one lacks a lot of organization and debugging before sending it to the public

by Marthinus J ( N

2020年4月8日

There was not enough examples or supplementary readings. Also the mentors and teachers dont reply on the forum.