[voxleone]

Very interesting. I’ve been working on a small library called SpinStep — it's a Python-based quaternion traversal framework that steps through trees based on orientation thresholds, rather than positional hierarchy. I designed it with spatial decision-making in mind, and while it's still early-stage, it might complement Ketu’s orientation handling or be interesting for modeling formation logic.

Totally open-ended — no expectations at all. Just thought I’d share it in case it aligns with any future ideas you’re exploring.

https://github.com/VoxleOne/SpinStep

[wanderinglight]

I like the idea. I"m not able to understand how quaternions alone can be used to navigate through space without a position translation vector.

The uses cases in the docs for drone control and manipulating kinetic arms with multiple degrees of freedom look promising though.

What does SpinStep provide? - Is it a traversal framework for quaternions? - Or it is a constraint solver that computes a series of transformations to make each node in a scene comply with a final end state

If it can be used for translating nodes as well via quaternions I would be interested in implementing a formation coordinator for it.

If you have an idea of what quaternion transformation will help achieve in Ketu and what the end state looks like please do create an issue on the Repo. I'll see if I can implement it using the concepts in SpinStep.

[voxleone]

Please excuse the delayed reply. Thanks for the thoughtful question — really appreciate you taking a close look.

SpinStep is primarily a traversal framework, not a constraint solver. It operates on orientation-based logic, where each node has a quaternion, and traversal occurs by stepping into child nodes whose orientations fall within a defined angular threshold from the current orientation.

So instead of computing a path toward a goal state, SpinStep says: “Given my current rotation, which nearby nodes are rotationally reachable?” It’s useful when the orientation itself is the condition for progression, like in:

    Animation state machines

    Scene graph logic

    Robotics command routing

    or potentially drone formations

At the moment, it doesn’t compute transformation sequences to reach a goal state — i.e., it’s not a solver or planner. But I agree: the quaternion logic could be extended to track both rotational and translational progression, especially if you define formations as goal orientations + positions.

I'd be happy to open an issue on Ketu with a concrete idea — possibly describing how formation transitions could be modeled as quaternion steps through orientation "nodes," - each drone is a node - and how that might be used to route coordination logic.

Thanks again for the openness.

[wanderinglight]

Formations in Ketu rely on knowing where nodes need to be placed and which slots in a formation are not filled.

Once this is known we translate the nodes towards the assigned slot in each step.

When you say "which nodes are rotationally reachable", what does that imply for formations? A node will be assigned a slot and has to move towards it while trying to avoid obstacles.

Happy to take a look if you file an issue on Ketu with details.

[voxleone]

>>When you say "which nodes are rotationally reachable", what does that imply for formations?

Here, position is essential, while SpinStep is orientational. Each node (e.g., drone) is assigned a target slot in space. Movement is needed — so SpinStep alone wouldn’t currently manage this. It might help decide orientation steps, but not positional navigation or collision avoidance.

Possibilities Going Forward:

We could extend SpinStep toward SE(3) traversal, or write a complementary module. That would involve:

-Pairing quaternions with positional vectors -- which is an awesome upgrade to the SpSt library.

-Defining a traversal condition based on both angular distance and spatial proximity.

-Possibly adding constraints (like occupied/available slots) and collision checks.

Such an approach could lead to a formation coordinator, where:

-Each drone is a node.

-Target positions + orientations define “goal nodes.”

-You route through “pose-space” rather than just physical space.

I'm a little swamped right now. In the days ahead i will be taking a closer look at how to integrate SpinStep with Ketu. It's gonna be fun.

All the best!