cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

How Serious are Autodesk about Developing and Continually Improving Robot

How Serious are Autodesk about Developing and Continually Improving Robot

Morning All,

 

I have been using Robot since Version 2017.  I work a lot with older buildings and have spent a lot of time and resources developing workflows and methods that allows me to use Robot to effectively do this albeit some of it using workflows that are not optimal from a IT perspective but are reliable from a structural analysis/design perspective.

 

My overall experiece with it is that it is a powerful and useful tool but it is a real ugly duckling in terms of the workflows, user interface and supporting documentation.  Having used it regularly, it's impossible to not acknowledge that in that time, very little has changed particularly when you compare it to Revit and even AutoCAD when you see just how much those programs have evolved in that time.  The only major improvement I can recall was enabling of Dynamo in Robot 3 or so versions ago albeit with very limited capabilities and little has evolved since then. 

 

Recently, I have been working a lot with analytical modelling in Revit and interoperating it with Robot and though it technically can be described as 'working', the results are far from impressive or even remotely reliable.  I often have extreme difficulty charing even the most basic information with any degree of reliability.  Common themes are mapping member cross sections from Revit families to Robot databases, grid lines ending up on multiple different grid systems in Robot, multiple, duplicated load cases and recently discovered that for some unknown reason, the model generated in Robot comes in at a different location in space than the equivalent Revit model.  The reverse is even less reliable when shending Robot results back to Revit to the point where I simply don't use it and opt to make changes in Revit and then re-export to Robot rather than risk stuffing up the Revit model and even this oly goes so far as there are tasks that simply cannot be carried out in Revit.

 

I am getting really frustrated with the apparent lack of progress on the software front in terms of reliable, consistent, transparent and easy to use BIM workflows for structural engineering.  It now is nearly 17 years since I first heard the term BIM and beack as far as 2004 when I first worked as an engineer using AutoCAD in a design office to draw stick dragrams representing framing and wondered to myself how good it would be if I could also use it to calculate bending moment diagrams and carry out beam capacity checks and yet, in 2023, we still are in a position where two pieces of software from the same company are incapable of reliably sharing even the most basic information such as where stuff is, what is is, what size it is and what material it is made of let alone anything complex such as what loads are on it, how heavily stressed it is or wll it buckle.  I often encounter the accusation that I work in the most conservative industry in the world that relies on backwards workflows yet those who accuse this industry of that seem to have a massive disconnect as to just how hard it is to work with the products they put out and to get them to reliably do what needs to be done for us use them to solve the real world problems we face. 

 

I really want to continue to use Robot as I believe that inside of it along with the ease with which Revit can generate geometry, analytical models, structural loads and everything else it is capable of along with the powerful options Dynamo gives you that there is massive potential for it or some replacement product in the Autodesk lineup to become the go to in the industry the same way Revit is and AutoCAD was.  However, I would like to know that somewhere in Autodesk, there is someone working on this and taking its development seriously.  I remember there was massive excitment when the pre-release version of a product called React was released, however this project never seemed to evolve and eventually, all references to it disappeared from the Autodesk website.  It's also impossible to not notice that many of the most popular ideas on this form date from as far back as 2012 and appear to still have not been enacted upon, many of the design standards are superseeded or obsolete versions and when you post to the forums, if you get a response from Autodesk, it's usually one of about three people which indicates to me that few if anyone is actively working on developing and improving Robot.

 

 

12 Comments
andi4G4Z4
Enthusiast

@anthonymctigue Even I faced difficulties in using analytical lines. It's not a smooth process like the Autodesk post in releases. And I always prefer one way integration ie from REVIT to ROBOT and if you're using Roundtrip integration it completely mess up your REVIT model templates etc

Advantages of using analytical one way integration I noticed are

  1. Grid from REVIT can be exported
  2. Geometry can be exported ( But skeptical)
  3. Supports and releases can be assigned in REVIT ( Time consuming than in ROBOT)

Disadvantages of using analytical one way integration are

  1. Rolled steel profiles are transferring as welded profile in ROBOT, which means the if you alter the Rolled profile names in REVIT, say HE360B to some other named then ROBOT won't map it as HEB 360. This will be unnoticed until you check your steel design clauses.
  2. Unconnected analytical lines in REVIT, say pitched roof with different Beam sections are connecting then its hard to create "Reference plane" in REVIT just for analytical connectivity
  3. Duplication of loads (DL etc) during "direct integration" during model changes.
  4. Missing of temperature loads
  5. Unable to assign support conditions in REVIT with one go. Instead always needs to pick one node at a time to assign support (very time consuming)
  6. Grid exported from REVIT are unable to edit
  7. There is no relation regarding the member number between REVIT and ROBOT or atleast Autodesk should provide some " compliance list" between REVIT and ROBOT integrated members. It is required for  quality analysis in projects

I think there should be some "unique ID" capturing through Dynamo or Python scripts atleast in next revision when you integrate with REVIT-ROBOT

anthonymctigue
Advocate

@andi4G4Z4,

 

Thanks for your input.  This is exactly the type of response I would like to see a lot of here coming from ordinary engineers expressing in a constructive way their current, day to day problems of working with the Revit/Robot workflows in the hope that Autodesk will pay attention and come to the conclusion that while it technically speaking satisfies the bare minimum definition of 'interoperability' and it is hypothetically possible with enough stuffing around to eventually to get it to do what you want, the process is often so painful and unreliable that the time savings are at best questionsable and at worse, slower than simply remodelling the entire structure a second time in Robot.

 

I have had all these difficulties and then some though I did get a killer tip from Autodesk Helpdesk on how to avoid one of those problems which is mapping sections from Revit to Robot.  It involves a rather tedious and highly untransparent process of editing the FamilyMap.xml file and paying very careful attention to how you name things in both the Robot database and your Revit families but you can get it to work reliably!  I spent a lot of time working with developing Revit families and Robot databases for discontinued steel sections as we deal with a lot of old buildings so there are no Revit families or Robot databases for a lot of the steel sections we work with.  I have managed to not only get it to map the sections correctly and reliably but also carry out the design checks to AS 4100.  See below screenshots of Revit and matching Robot model below.

 

REVIT 01.PNG

 

REVIT 02.PNG

 

ROBOT 01.PNG

 

ROBOT 02.PNG

 

Happy to provide further information as to how I got this to work if it is of interest to you. 

 

As for the other items you alluded to, one of my biggest gripes is that the mapped geometry in Robot is NOT in the same location in space as it is in Revit.  I recently discovered this when I had a Robot model directly exported from Revit which covered only part of the building I was trying to model.  I then realised I wanted to add a bit more of the building into the Robot model and rather than using the reather unpredictable direct export option to 'update' the Robot model with additional geometry from Revit, I thought I could be smart and export the additional geometry using a Dynamo script only to discover that the additional geometry appeared in Robot in the incorrect location.  I have been intermittently posting rants about this sort of problem on LinkedIn  (as well as posting stories of successes when they happen) and tagging Autodesk into it in the hope that they eventually notice so more info on this can be found on this particular problem at the link below.

 

https://www.linkedin.com/posts/anthony-mctigue-73620248_after-half-a-day-trying-to-figure-out-why-ac...

 

andi4G4Z4
Enthusiast

Hi @anthonymctigue Thank you for the tip. However I'm unable to get the FamilyMap.xml in REVIT database or the window of REVIT showing the section properties and I don't want to change the section type name in REVIT to map with ROBOT database.
Say for example below the " Beam_steel_HE400A is mapped as welded profile in ROBOT database. It should map as HEA 400 in ROBOT (Rolled steel profile).I don't know how to correct it automatically while mapping. If not this will be mistake in steel design as the parameters for the welded profile buckling classes will be taken in RSA steel design module.

andi4G4Z4_0-1694693188941.png

 

The solution I tried to solve above was to export the sections from ROBOT to excel and change the names of bars to HEA 400 in excel and paste back to RSA tables. This is time consuming and may cause manual errors.

 

anthonymctigue
Advocate

@andi4G4Z4,

 

I know that pain you are referring to with editing the Robot database (the pro.xml file) and using pasted data from Excel to do it.  It's a completely untransparent process with little to no documentation from Autodesk describing how to do it other than it is theoretically possible to copy and paste information between databses but as you probably have now found it, this is fraught with difficulty.  I am willing to bet you got that error message about attempting to edit and unbound field or something like that more than a few times at this point!

 

The method that I have found that works best is that you need to set up your Excel spreadsheet in exactly the same format as the Robot database.  However, even this is fraught with problems in that you will find that the pro.xml database files that come with Robot, including those for standard cross sections, do not all follow the same column arrangement or format.  This took me ages to work out and the way I eventually worked around it was as follows (bear in mind that for the most part I was making entirely new databases for obsolete steel sections):

 

  1. Copy one of the built in pro.xml database files (in my instance, the Australian sizes), clear out the contents and save it on the server as a template file for all new Robot databases.
  2. Use the 'Developer' add-in on Excel to link directly to the blank database file into Excel such that it displays in a table format (see screenshot below).
  3. EXCEL 01.PNG
  4. Open a second Excel window with a new blank workbook and copy and paste Columns NAME10 to last column from the linked database to the blank workbook (these are the columns you will need to make sure are the same between your Excel and Robot database).  There will be other columns visible in the linked database  that you won't need.
  5. Populate the new Excel table with the information you require.
  6. Copy and paste the blank pro.xml file to create a new database and copy and paste your finished excel table into it. 

If this sounds confusing and far fetched, it's because it is.  There are also quite a few other steps in there that are very hard to describe in a message in terms of defining basic section geometry if you are using entirely new cross sections that don't come built in to Robot but getting your Excel template format to match exactly the Robot database format is one of the first important steps you need to master to get this to work. 

 

My Excel template that I have been working on has a macro inside it that allows me to create the Revit family types .txt file and the FamilyMapping.xml file entries for any given dataset.  It's more than a bit rough around the edges and I am still trying to work out how to make it easier to use so I don't give it out, even within our own office as it often involves having to make slight edits to the code but if you want to direct message me and send me your Robot database and Revit family, I am happy to see if I can get this to work for you and/or send you some examples of some of my Revit families/Robot databases/FamilyMapping.xml file entries or if you want me to walk you through it step by step, happy to do a Teams call sometime.

 

Anthony

 

 

andi4G4Z4
Enthusiast

@anthonymctigue  Thank you for the tips and your time . I completely understood your effort behind it. But the point of using these addin is reduce time and improve quality. So I guess Support center will take this in their "bucket list" to improve quality of this addins.

 

Thank you for the suggestions as well  however I'm not using the REVIT-ROBOT integration much after initial integration and I don't want to spend time to look into these disadvantages of using the process . But I recommend other to look into the process rather than simply go with the conventional RSA modelling

 

BR 

Arun 

anthonymctigue
Advocate

@andi4G4Z4.  No worries!  my preference too is for these workflows to work better so I don't have to spend days upon days working this stuff out or at worst, if there is something unusual that I need to do with it that couldn't have reasonably been forseen by the developers as something that it might get used for, at least have the process open, easy to understand and with good documentation such that one can develop a workflow that does work.

arodrigues9E874
Contributor

Hi anthony, I am not as nearly as knowledgeable as you but rather a novice structural engineering worker starting in the industry and aspiring to be a structural engineer someday. That being said, I completely agree with you, I taught myself RSA, I have completed two full online courses on it, and it is still a like I don't know anything. It is one of the least intuitive software's I have ever used. I really admire you for having used it for so long, most people dont even bother to create workflows like you did. Would you be willing to share your workflows with me?

anthonymctigue
Advocate

@arodrigues9E874, I would be happy to show you what I know but there is not a lot to what I have figured out!  A lot of it is very haphazard and adjusted on the fly to suit the specifics of a given project but happy to share what I do have in the hope it can be useful.

ismaeelm
Participant

I agree with this post. I have been a loyal Robot user for a long time. I tried SAP2000 recently and the software feels so much more modern than Robot. Autodesk needs to update the interface and GUI. Modelling needs to be smoother. The analytical model in Revit helps for initial modelling, but once imported into Robot I spend 95% of my project time in Robot, especially when the project scope changes which happens on most projects. Once a model is complete with all loads and inputs, it’s just not possible to get it back into Revit for updates without losing critical aspects. Autodesk has the modelling technology in Autocad and Revit. It would be amazing to see this being implemented into Robot. Imagine drawing your elements with snaps as smooth as Revit. 

steinryan28
Participant

Autodesk is known for its commitment to providing a wide range of software solutions for various industries, including architecture, engineering, and construction. Robot Structural Analysis is a structural analysis and design software that is widely used in the engineering and construction industry for tasks like finite element analysis and structural design.

 

To assess how serious Autodesk is about developing and improving Robot Structural Analysis, you should read this informative blog for latest update and guide: Robot Structural Analysis: Advanced Design Modeling Solutions

anthonymctigue
Advocate

@steinryan28,

 

Thanks for your contribution.  I mean no disrespect, but this response is nothing but bog standard, copied and pasted marketing spin.  I have used Autodesk products since Autocad R14 and I know that they continually improve them, at least most of them.  This topic is specifically relating to Robot which wasn't always an Autodesk product much in the same way Revit wasn't but Revit comes along in leaps and bounds with every new version while Robot has never shown anything that anyone will ever convince me in my time of regularly using it that compromises a massive improvement.  I also had limited exposure to it nearly 20 years ago before it ever was an Autodesk product and the current 2024 version is still very similar in how used to operate in that version I saw 20 years ago. 

 

It seems to be the forgotten child in the Autodesk product family and this is based on a good solid 6 years of continuous use and giving interoperability with Revit in particular a red hot go and continually coming up against roadblocks to use it effectively in my line of work.  Robot doesn't even have access to services such as cloud model sharing that most of the other Autodesk products have access to, it's support tools and resources are hopeless other than to show you how to set up very basic models, certain features such as phased construction are all but unusable and many of the national standards (speaking specifically of AS 3600) are either not fully implemented or the versions that are implemented are several version out of date (AS 4100:1998 for example when the current version is AS 4100:2021). 

 

As a user, this screams that there is a lack of interest from Autodesk on what to do with it which is not necessarily a critisim of them as they may have other priorities which is fine, I am not going to sit here and dictate to them as to how to run their business.  The aim of this post is not to badmouth Robot or Autodesk per say but to try and gauge what Autodesk's attitude is towards further development and what their long term plan for Robot is.  For example, if I know with reasonably certainty that in the next two years, there will be a massive push to improve the interface, bring natioal standards up to date, properly enable bidirectional workflow with Revit, enable cloud services and dramatically expand Dynamo capabilities, then I will happily sit here and plough away with the current version until then.  However, what I want to avoid is another 5 years of the same forgotten child approach that I have observed over the last 6 with still no noticeable improvements and/or Autodesk either abandoning the product or selling it off, leaving me with having to start again with something new.  I want to know if I should shortcut the latter option by finding a new product now and spending the next 5 years working that out as opposed to spending those 5 years on a road to nowhere.

 

Anthony

pawelpiechnik
Autodesk
Status changed to: Gathering Support

Anthony, Please see how Autodesk sees the future of its Structural Analysis and adjacent workflows explained in the Public Roadmap here: https://blogs.autodesk.com/aec/roadmap/revit-structure-roadmap/

regards

Can't find what you're looking for? Ask the community or share your knowledge.

Submit Idea  

Autodesk Design & Make Report