Missing Plugins/Commands in Rhino

[mc_rash]

Hey everybody,

During my study I got interested in Arduino projects, started coding and after a while I got the oppertunity to follow a Python Course at my university. Maybe the biggest project of my Ardunio hobbywork is a lidar-based 3D scanner I made to scan boats/ships and get a 3D shape as a base for refit etc. It's now laying arround waiting for real world tests with a hull from which the lines are known. For the Python Course I invented an application which finds the fastest route for sailboats depending on weather forecast and the boatspecific VPP. This application works but it is just a prototype and I'm not planning to work further on it.

For me as naval architecture student Rhino is probably the favorite 3D design and CAD software. It's affordable, easy to use and offers plenty of functionality. This functionality can be extented with scripts, commands, plugins etc.

I would like to write some plugins for Rhino which simplifies the life of the maritime engineer. The issue is - I'm lacking ideas for plugins which could be actually usefull.

Do you have any ideas or suggestions what I could 'invent'? What are you missing in Rhino? Let me know.

 

[Remmlinger]

I am interested in your 3D-scan. Which hull did you scan? Any hull, where the lines plan is published by the class-society (e.g. H-boat) can be used for your validation. Are you able to import your scan into Rhino and create smooth surfaces?
Regards, Uli

 

[mc_rash]

Hi Remmlinger,

I would like to satisfy your interest in a 3D scan of a hull, unfortunately there is none as is I said the scanner is waiting for 'real world' tests with a hull from which the lines are known.
Until now I didn't have the chance to test the scanner in a maritime environment.

I am able to import the data in Rhino as a 3D pointcloud, see the attached file with a top view of a room, the outline is clearly visible.

There are still some improvements necessary to get from a 'simple' prototype to a proper working device:

Better accuracy, in a later stage a solution would be a more costly but better LIDAR sensor. Now there are tolerances up to 10 cm.

Code improvements, software is stable but gets unstable/ data get a little unreliable if much sensor data are collected. Edit: code is a bit slow and could be speed up for a faster measuring experience.

Hardware: Once I blowed up the SD memory card in smoke.. probably a software issue as mentioned above or a simple short circuit. Didn't use it since then.

To answer your last question, data-postprocessing might be the main subject which should be focused on. Pointclouds are nice to view but (smooth) surfaces should be the final resultat as you can do nearly nothing with just points.

For boat/ship hulls: scanning the inside should be no issue if the space falls within the range of the LIDAR ( roughly 12 m), to scan the outside several scans from different points of view have to be done as hulls are curved and the scanner can't look around the whole hull. Edit: multiple scans needs to be combined in postprocessing, requires reference points etc.

 

[ludesign]

Hi Remmlinger,

I would like to satisfy your interest in a 3D scan of a hull, unfortunately there is none as is I said the scanner is waiting for 'real world' tests with a hull from which the lines are known.
Until now I didn't have the chance to test the scanner in a maritime environment.

I am able to import the data in Rhino as a 3D pointcloud, see the attached file with a top view of a room, the outline is clearly visible.

There are still some improvements necessary to get from a 'simple' prototype to a proper working device:

Better accuracy, in a later stage a solution would be a more costly but better LIDAR sensor. Now there are tolerances up to 10 cm.

Code improvements, software is stable but gets unstable/ data get a little unreliable if much sensor data are collected. Edit: code is a bit slow and could be speed up for a faster measuring experience.

Hardware: Once I blowed up the SD memory card in smoke.. probably a software issue as mentioned above or a simple short circuit. Didn't use it since then.

To answer your last question, data-postprocessing might be the main subject which should be focused on. Pointclouds are nice to view but (smooth) surfaces should be the final resultat as you can do nearly nothing with just points.

For boat/ship hulls: scanning the inside should be no issue if the space falls within the range of the LIDAR ( roughly 12 m), to scan the outside several scans from different points of view have to be done as hulls are curved and the scanner can't look around the whole hull. Edit: multiple scans needs to be combined in postprocessing, requires reference points etc.

Just out of curiosity, how well does the room scan match the real room? I regularly use my iPhone Pro equipped with Lidar to scan, and on shapes like a room, I typically get an accuracy of about 1%. It has a max range of 5 meters, but you can scan much bigger objects by walking around. Pretty good for something not really intended to be used as a professional measuring tool, using a free software downloaded from Appstore. It's of course not comparable to a professional tool, but it sort of comes in the box when you buy the phone.

The examples shows an 180 sqm office which took say ten minute walk to scan, a couple of scans of sculptures that took say 5 minutes each. The last one was dumped into my 3D printer and printed without any problem. The app can also be used for augmented reality (picture to the right), generate presentation movies, and be imported into my CAD app for further processing.

 

[Remmlinger]

Better accuracy, in a later stage a solution would be a more costly but better LIDAR sensor. Now there are tolerances up to 10 cm.

Following your description, it seems that there is no DIY-solution. The tolerance should be 2 mm max. There are engineering companies, especially in the architectural business, which offer a scan and the conversion into a CAD-file (IGES) that has the required accuracy. Their service might be less costly than the price of the necessary equipment. I am looking for a file of offsets as input into my VPP, similar to the offsets, which are created for the ORC-certificate.