I am glad Mr. Stachniss is doing this kind of videos on RUclips. Sometimes I am missing a bit of background but the videos give a very good general idea of the conveyed concepts. Please continue!
Came across him back in college. His remote sensing knowledge is astounding. Please keep the content coming! Specifically some Lidar object classification. Please and thank you for all.
Thank you for a great lecture! Can this filtering algorithm be used to build a local map of current environment around a robot by noisy data of range sensor? For use in tasks such as collision avoidance or local path planning
Hello Cyrill, I wanted to know what values are reasonable for p_occ and p_free. Can you tell an example of values that you used in the past? Just to get an idea. Thank you very much for the lecture and your work in general.
One questions Prof Stachniss, how should we update the map on loop closure? Because at that point, all the wrong cells will be marked with the occupancy probabilities based on the pose-landmark graph, and as we know, Loop closure can change the pose-landmark graph dramatically. Another question, say we have a occupancy grid based on two sensors say a LIdar, and a Sonar, how do we puse both of them? Should we just multiply the probabilities, or look for something like a Kalman Filter?
Nice video. am new to this topic, where do i get know meaning (details ) for mapping, grid. local grid , uncertainty ...? is there any book or link for beginners with all the basic details to expert level
Great video Sir it really helps. I have a question if you can help me, what is the difference between and occupancy grid map and reflection or counting based mapping.
As the name says, one approach estimates the probability of occupancy and the other one estimate the probability if a beam gets reflected by cell or not. These are two related concepts but not the same things.
@@CyrillStachniss Thank you so much Prof. for the reply. Another question to that how this makes a difference when considering windows/glasses in the map.
The slides or similar sliedes can be found on the "mobile robotics lecture" of the Uni Freiburg. Just google "uni freiburg mobile robotics". The slides are a bit different because their are from Prof. Burgard, but their almost the same. Be careful, the order of the slides their are not the same order as this playlist. You have to look at the topics to find the right one.
![Mary J. Blige - Breathing (feat. Fabolous) [Official Video]](http://i.ytimg.com/vi/snMHObq-6E0/mqdefault.jpg)
I am glad Mr. Stachniss is doing this kind of videos on RUclips. Sometimes I am missing a bit of background but the videos give a very good general idea of the conveyed concepts. Please continue!
Came across him back in college. His remote sensing knowledge is astounding. Please keep the content coming! Specifically some Lidar object classification. Please and thank you for all.
I must say this was 100 times better explained then by Prof. Burgards mobile robotics lectures, really nice and clean.
Thank you Cyrill for clear explanation on robot map this help me to build my map to learn on robot and autonomous cars !!
Extremely helpful and well explained. Thank you for this video!
Great explanation!
very well explained
Thank you for a great lecture!
Can this filtering algorithm be used to build a local map of current environment around a robot by noisy data of range sensor? For use in tasks such as collision avoidance or local path planning
Yes, that’s what it is made for
Es wäre sehr hilfreich, wenn Sie die ppt slides für diese informative Vorlesung teilen könnten.
Can you guy recommend me some coding books for these lectures, thanks?
Hello Cyrill, I wanted to know what values are reasonable for p_occ and p_free.
Can you tell an example of values that you used in the past? Just to get an idea.
Thank you very much for the lecture and your work in general.
Starting probability is 0.5 since it’s unknown
One questions Prof Stachniss, how should we update the map on loop closure? Because at that point, all the wrong cells will be marked with the occupancy probabilities based on the pose-landmark graph, and as we know, Loop closure can change the pose-landmark graph dramatically.
Another question, say we have a occupancy grid based on two sensors say a LIdar, and a Sonar, how do we puse both of them? Should we just multiply the probabilities, or look for something like a Kalman Filter?
Regenerate it with new poses…
Nice video. am new to this topic, where do i get know meaning (details ) for mapping, grid. local grid , uncertainty ...? is there any book or link for beginners with all the basic details to expert level
If you don’t mind reading, look into the probabilistic robotic book by Thrun, Burgard, Fox
which is the best software to deal with lidar data & occupancy grid generation purposes?
Great video Sir it really helps. I have a question if you can help me, what is the difference between and occupancy grid map and reflection or counting based mapping.
As the name says, one approach estimates the probability of occupancy and the other one estimate the probability if a beam gets reflected by cell or not. These are two related concepts but not the same things.
@@CyrillStachniss Thank you so much Prof. for the reply. Another question to that how this makes a difference when considering windows/glasses in the map.
will you please provide slides ?
The slides or similar sliedes can be found on the "mobile robotics lecture" of the Uni Freiburg. Just google "uni freiburg mobile robotics". The slides are a bit different because their are from Prof. Burgard, but their almost the same. Be careful, the order of the slides their are not the same order as this playlist. You have to look at the topics to find the right one.
Great explanation and very helpful information. Thank you!