Build a Robot-Cell Thermal Warpage Inspection Planner — implementation

## Build
Analyze 3D optical scanner point clouds for thermal warpage against MoveIt collision meshes. Plan robot-assisted flame straightening trajectories for out-of-tolerance structural frames, or divert to manual repair stations.

## User
Welding Cell Quality Specialist

## Workflow
1. Ingest 3D scanner point clouds of welded assemblies and baseline CAD collision meshes from local storage.
2. Calculate maximum thermal warpage distortion vectors using point cloud registration against baseline geometries.
3. Evaluate kinematic reachability in MoveIt for secondary flame straightening torch trajectories along distorted areas.
4. Emit a structured heating runbook detailing torch approach vectors, dwell times, or manual diversion routings.

## Interface
Metrology dashboard displaying 3D thermal distortion heatmaps, point cloud registrations, and robotic trajectory overlays.

## Data
Normalized 3D point cloud coordinate files and simulated robot arm DH-parameters stored as simplified synthetic JSON schema files.
Starter examples below are synthetic. They are not downloaded records.

## Source
[MoveIt robotics](https://raw.githubusercontent.com/moveit/moveit2/main/README.md)
MoveIt motion planning framework for kinematic reachability and collision mesh evaluations.

## Tools
agno, open3d, numpy, pydantic.
Lightweight agentic orchestration to package calculated warpage vectors and coordinate MoveIt solver queries into structured heating tasks.
Fast hosted or local model capable of translating 3D distortion metrics into structured JSON trajectory plans.

## Deliverables
Submit runnable source, pinned dependencies, setup commands, a working interface, example outputs, and a recorded end-to-end demo. Include tests and an action log.

## Acceptance
- Calculates accurate deviation magnitudes from given nominal and scanned 3D coordinates.
- Rejects automated flame straightening for warpage exceeding 10.0mm, routing to manual repair.
- Calculates dwell_time_s as exactly 10 times the max_deviation_mm for robot routing.
- Emits complete payload containing specific dwell times, passes, and manual intervention flags.

## Starter Examples
```json
[
  {
    "input": {
      "part_id": "FRAME_SUB_849",
      "distortion_type": "BOWING",
      "max_deviation_mm": 4.5,
      "coordinates": {
        "x": 320,
        "y": 140,
        "z": 210
      },
      "kinematic_reachable": true
    },
    "expected_output": {
      "routing": "ROBOT_FLAME_STRAIGHTENING",
      "tool_action": "LOCALIZED_HEATING",
      "parameters": {
        "dwell_time_s": 45,
        "passes": 1
      },
      "manual_intervention_required": false
    }
  },
  {
    "input": {
      "part_id": "FRAME_SUB_850",
      "distortion_type": "TWIST",
      "max_deviation_mm": 12.5,
      "coordinates": {
        "x": 320,
        "y": 140,
        "z": 210
      },
      "kinematic_reachable": true
    },
    "expected_output": {
      "routing": "MANUAL_REPAIR_STATION",
      "tool_action": "REJECT_FROM_ROBOT_CELL",
      "parameters": {},
      "manual_intervention_required": true
    }
  }
]
```

## Safety
Read-only metrology evaluation; physical localized heating tasks require prior human review and cell safety clearance.
Use mock external writes. Require human review before any real deployment.