this repo focuses on conceptualising the agent-simulation based geometry generation for reactive 3D printing below other modules of the robotic fabrication platform are drafted to, to give the context

diffusion mechanism several layers of the environment effects the agents' behaviour the agents react to the environment and form it in the same time

the layers of the environment: voxel-like objects (numpy arrays) agents in the environment: general objects mechanism: the diffusion

examples:

• Earth solid ground
• Air we could consider forming structures by building negative spaces
• PathPheromons way of agent communication
• FoodPheromons way of agent attraction
• AsBuilt input from the 3Dscan
• CollisionArea precalculated unreachable area based on the ground and asbuilt layer

• density the voxel array
• diffusion_strength
• show_color
• attraction_strength the attraction strength between a Layer and an Agent class

• Builders
• Explorers
• others?

methods:

• move *
• check_move_options
• get_impulse
• get_neighbor_voxels

properties:

• position
• movement_limitations
• pathpheromon_strength

global properties:

• pheromon_attraction_strength

it could perhaphs make sense to create mechanism Classes instead of inbuilt functions

• VoxelEngine every layer refers to this object to build its own array accordingly
• DiffuseALayer
• OverlaySomeLayers
• DumpState dumps a state of a layer or the entire world
• LoadAsBuilt
• Wind
• Gravity

standalone visualiser of one layer or the world State

1. initial design goal-parameters-constrains definition
2. enabling realtime designer interaction (3D or text-based)

tool to generate a robotic path plan to 3Dprint the sequence from the as built state to a given state