antelocnova

You're welcome ;-)

I remember coming across this research paper, and I thought it was really impressive. I clearly remember the bricks guitar, built by robots =)

Currently, Nova doesn't handle any physical constraints regarding stability, just collision.

I think physics handling is a very desirable feature, maybe an agent using a physics engine directly or maybe via MCP that agent could get feedback about a model's stability, fix problems, back to more feedback, etc.

This would be similar to the current way finds and iteratively corrects defects by rendering images from the model, different perspectives, and directly examining them.

A very similar idea crossed my mind several times, but with a twist: altering LEGO CAD parts, but keeping the connectors the same.

Main idea was, if LEGO parts don't cover the needs for an agent to build a model, let it use the best it can find, and manufacture others if needed, but keeping same compatibility with the rest of the catalog.

And for that, instead of generating new parts from scratch, maybe mutating pre-existing parts into new shapes it needs, but leaving connectors as they were, or even adding new connectors but still compatible.

Sure! Thanks dang ;)

And regarding building those, well... actually, I didn't even try, the amount of parts required alone for the larger models is actually in the order of the thousands, my guess is it would be difficult to get all the required parts for just one of those models.

I'm currently considering 3D-printing a small one, however, in order to see how it looks IRL.

Physics modeling is something Nova is currently lacking, and I think very interesting, given that agents currently don't verify physics, stability... only placements.

Well, I'd really like to see that, robots building models generated by agents.

Something that comes to my mind very often is... adding more parts to the LDraw parts catalog, the ones used to build models, but these extra ones would be real-sized blocks, like the ones used to build actual buildings ;-)

Well, that greatly depends on the thinking required to produce a single LDraw model, not so much regarding how much of a "thinker" a model is, but how much thinking it must do to get a model right.

Technic models, for example, are usually expensive, given that initial placements are usually further from the right position than non-Technic models, the "bricks" ones.

A (wild!) estimate: Astra, building a Technic mechanism, around 5$...

This is one of the reasons for creating this project: making a tooling that can be used by low-end agents to iteratively build, while high-end agents, like e.g. Astra or Opus, would do the planning, vision models the inspection... etc.

Well, who knows :)

That's unfinished work, however... didn't get good results with Technic models yet, except for very specific cases, like e.g. the Atlas Crane.

That's still WIP, started already with Technic assemblies, not mechanisms yet, in order to first produce the required scaffolding to attach things like e.g. cogs, transmissions, etc. to them.

Getting mechanisms to work is one of my goals, though =)

A mix of both, I think.

When giving them enough examples, like e.g. houses, they tend to do well in that category, but not perfect.

My guess is that what happens is, given that most LEGO parts have a very limited, discrete set of legal positions and rotations, expressed as integer numbers, the agent finds patterns between different models, like e.g. a brick close to the center of a model usually has very similar (integer) coordinates and rotation components to another central brick from another model.

However, it still messes up very often, with things like e.g. +/- signs that flip parts due to rotations becoming -90deg when they should be +90deg; then, validation and correction cycles, running tools that verify the presence/absence of collisions, really helps in finishing the job.

Hi there, thanks!

I'm taking a look at this paper right now, didn't came across it before.

I think it's really good, and also an approach I tried.

In my case, it was working pretty well in terms of geometry and correctness, but not in terms of creativity: too many constraints and an agentic LLM would end up building very similar buildings all the time. Too few ones, and it would drift into unbuildable models.

In the end, I think the current iterative process gives it a good balance: let it try things first and be creative, parts will be placed incorrectly at the beginning, but it will fix their placements, colours and so on bit by bit, until the end result is (hopefully!) correct.

And regarding Opus, according to my experience, I'd say definitely models quality is way better when produced by Opus 5 and even better by Opus 5.5