Implementation of 1873-2015 - IEEE Standard for Robot Map Data Representation for Navigation
Hi,
Is there an implementation of IEEE Standard for Robot Map Data Representation for Navigation (1873-2015) available in ROS or any other platforms to convert grid map to/from IEEE standard?
Edit2:
another update link
Thank you.
Asked by bvbdort on 2019-03-11 08:00:17 UTC
Answers
I don't think that there's an implementation. I'm not familiar with the standard, and cannot access the document due to the paywall. That will also make it hard to implement and verify an open source implementation if the reference cannot be redistributed.
Update:
Thanks for the updated link to the article about the standard. Reading it through it appears that the format defined is an xml format. Specifically to your question I have not heard of anyone using an xml format to represent grid maps.
Overall I'm not sure the xml is the right data format to represent dense data such as grid maps.
Asked by tfoote on 2019-03-11 12:40:26 UTC
Comments
I have added a new link which is accessible.
Asked by bvbdort on 2019-03-11 17:20:49 UTC
I can't access that link either it looks like it's missing the hostname
http://publication/322549175_A_Standard_for_Map_Data_Representation_IEEE_1873-2015_Facilitates_Interoperability_Between_Robots
Asked by tfoote on 2019-03-11 17:26:27 UTC
sorry, I have just updated the correct one.
Asked by bvbdort on 2019-03-12 03:16:19 UTC
Interesting to see this discussion coming up here. I agree that I have never seen an xml format for gridmap data, and that it is not the right format for a grid map being used actively on a robot.
They work around the "dense" part a little bit by allowing to cluster aligned portions in one single tag by using the width and height portions, so one could emulate something a little bit more generic than a quadtree. But as this is a non-unique representation in the end, I'm not sure what this would mean for any implementation...
Looking into the Scope of the standard, it seems to be targeted mainly at exchanging data between different systems, and not (necessarily) for active use withing e.g. a localization algorithm...
Will be interesting to see if this will be adopted anywhere in the near future...
Asked by mgruhler on 2019-03-12 03:56:55 UTC
I was involved in the early stages of the creation of that standard. It is indeed targeted at information exchange, rather than run time use. The use case we were working with was provided by a group from Korea working on a demonstration smart city, and was about distributing map data for sections of the city to individual robots working in specific areas.
The standard was published three and a half years ago now. I don't know how much adoption it's had by industry since then, but if it hasn't had any yet then I wouldn't expect it to gain much now. In fact during those early discussions a comment we often heard industry people making was "how can we shut this effort down?" I think the fear was that it would be a standard produced by academics that didn't meet a real-life need of industry, but despite that institutions such as the US military would require it simply because it exists.
Asked by Geoff on 2019-03-12 19:04:18 UTC
Comments