Commit Graph
44 Commits
Author SHA1 Message Date
Adriano Dal Pastro 7f60238be9 fix(pose_3d): URDF feet a terra (+0.793m) + sfere polso a zero-pose (Z=0.45) 2026-05-21 17:41:31 +02:00
Adriano Dal Pastro bfd4cc13fb fix(pose_3d): URDFLoader path resolution (rimuovi doppio prefisso /static/vendor/g1/) 2026-05-21 17:37:25 +02:00
Adriano Dal Pastro d5bb9cebd3 feat(pose_3d): workspace check client-side (ghost verde/rosso) + SHOULDER da URDF 2026-05-21 17:21:09 +02:00
Adriano Dal Pastro 8c09d79505 feat(pose_3d): modello reale G1 (URDF + STL meshes) al posto della silhouette finta
Sostituisce torso/head/hips/gambe/spalle primitive in tab Pose 3D con il
modello URDF G1 (g1_body29_hand14), caricato via urdf-loader 0.12.6.

Vendor:
- web/vendor/g1/g1.urdf + 64 mesh STL (~52 MB, copiati da
  G1_API_Server/assets/g1/)
- web/vendor/three/addons/loaders/URDFLoader.js (npm urdf-loader@0.12.6),
  import path adattato a 'three/addons/loaders/STLLoader.js' (importmap
  del sito) e ColladaLoader stubbato (G1 URDF è solo STL)
- web/vendor/three/addons/loaders/URDFClasses.js (companion del loader)

pose_3d.js:
- Aggiunge loadG1Urdf(scene) async che configura loader.loadMeshCb per
  servire le STL dal mount /static/vendor/g1/meshes/
- Best-effort: in caso di errore di parsing/fetch il body non viene
  aggiunto ma la tab continua a funzionare (polsi, UCS, gizmo)
2026-05-21 17:14:16 +02:00
Adriano Dal Pastro 18dce0d951 feat(pose_3d): bottone Sincronizza con robot (refetch state + reset target) 2026-05-21 17:08:09 +02:00
Adriano Dal Pastro b852976a4a feat(pose_3d): readout XYZ/RPY editabili (input number, drive il ghost) 2026-05-21 17:01:03 +02:00
Adriano Dal Pastro 8b722ca729 feat(pose_3d): silhouette robot completa (testa+gambe+spalle) + FRONT come marker terra 2026-05-21 16:46:44 +02:00
Adriano Dal Pastro d4cd82551e fix(pose_3d): ghost fallback alla posa dell'arm quando /state e' offline 2026-05-21 16:44:57 +02:00
Adriano Dal Pastro fd506dff9b feat(pose_3d): etichette DX/SX sulle sfere e FRONT al base 2026-05-21 16:41:08 +02:00
Adriano Dal Pastro 25c94d3861 fix(pose_3d): gizmo visibility (use getHelper for r160) + larger wrist hitbox + diagnostics 2026-05-21 16:36:09 +02:00
Adriano Dal Pastro 71baf829e6 feat(pose_3d): bottone Sblocca braccia 2026-05-21 16:06:43 +02:00
Adriano Dal Pastro bb6796bfe3 feat(pose_3d): save/load/delete named poses 2026-05-21 16:03:38 +02:00
Adriano Dal Pastro a81fc9acf9 feat(pose_3d): Jog mode auto-apply + safety + Precise toggle 2026-05-21 16:01:56 +02:00
Adriano Dal Pastro 5b2239e307 feat(pose_3d): Editor mode Apply + Cancel 2026-05-21 16:00:15 +02:00
Adriano Dal Pastro 64500944d6 feat(pose_3d): selezione polso + gizmo translate/rotate sul ghost 2026-05-21 15:58:41 +02:00
Adriano Dal Pastro 7a561cfa67 feat(pose_3d): polling /state + readout sidebar 2026-05-21 15:56:34 +02:00
Adriano Dal Pastro f161894d53 feat(pose_3d): scene Three.js base (torso, polsi, UCS, grid) 2026-05-21 15:54:29 +02:00
Adriano Dal Pastro c4124a2983 feat(pose_3d): markup HTML + CSS della tab 2026-05-21 15:50:49 +02:00
Adriano Dal Pastro 433edc0974 chore(web): vendor TransformControls (per tab Pose 3D) 2026-05-21 15:48:01 +02:00
Adriano Dal PastroandClaude Opus 4.7 57954b5c8d feat(right_arm): perm assi cam->polso, precise=true default, verify_tag UI
Lavoro accumulato nella sessione 2026-05-21 prima del brainstorming
della tab Pose 3D:

- compute_delta_wrist generalizzato con cam_to_wrist_perm: la mappa
  cam->polso non e' solo sign-flip ma una permutazione (default
  identita' per backward compat). DeltaOversize ora espone X/Y/Z
  nella stringa d'errore. Test aggiornati.

- G1Client.wrist_absolute / wrist_relative: precise=True di default
  (era False). Stessa cosa per wrist_routes.AbsoluteReq Pydantic
  e per agent.tools move_wrist_*. Movimenti reali ora chiedono
  controllo accurato salvo override esplicito.

- Tab Braccio DX:
  - /verify_tag endpoint: snapshot wrist FK + tag id=1 dalla front
    camera. Per verifica relativa moveto vs marker (no calibrazione).
  - GUI: pannello "Verifica tag id=1" + campi camera_offset_wrist_mm.
  - markUserTouchedOnInput esteso a raOff/raSgn/raCamOff.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-21 15:44:48 +02:00
Adriano Dal PastroandClaude Opus 4.7 1fa7d12ad7 feat(right_arm): force, camera↔polso offset, cluster filter sul centroide
Tre miglioramenti al move-toward dopo il primo uso live:

1. checkbox "force" accanto a "Muovi": bypassa max_step_m quando
   l'utente sa che il delta richiesto e' grande ma corretto (es. primo
   avvicinamento all'oggetto).

2. camera_offset_wrist_mm: il modello assumeva camera e TCP coincidenti
   (sole sign flips). La camera Nuwa e' invece montata con traslazione
   costante rispetto al polso. La formula diventa:
       delta_wrist = sign * (centroid_cam - offset_cam) + t_wrist_cam
   Esposto in config (default 0,0,0), GUI (3 input "Camera rispetto a
   polso (mm)") e in /state. Il termine compensa il fatto che muovere
   il polso porta la camera, non il TCP, sul punto di presa.

3. filter_nearest_cluster: il centroide era media di tutti i punti
   SAM21 della mask. In presenza di sfondo lontano visibile attraverso
   buchi della depth del primo piano la media veniva trascinata a
   metri di distanza. Ora si tiene il cluster piu' vicino:
       z_near = p5(Z)
       keep   = punti con Z <= z_near + cluster_depth_window_m
   Default window 0.20 m, configurabile. La risposta di /detect/wrist
   espone n_sam21_points, n_centroid_points e z_near/z_cutoff per
   diagnosi.

Aggiornata anche la persistenza: /offset accetta camera_offset_wrist_mm.
GUI: i campi Offset/Sign/CamOff rispettano "user touched" (refresh non
li cancella mentre l'utente digita).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-21 12:18:24 +02:00
Adriano Dal PastroandClaude Opus 4.7 2f6444c0a5 fix(right_arm): bbox dict, image_size da /api/status, tab non seleziona robot
Tre problemi della tab Braccio DX dopo il primo rodaggio:

1. /detect/front e /detect/wrist ritornavano bbox raw dal YOLO server,
   che lo manda come lista [x1,y1,x2,y2]; la GUI legge bbox.x1 e mostra
   NaN. Ora il router normalizza il payload via perception/detector
   ._normalize, allineandosi al resto del codice. Aggiunta regressione
   con payload reale del server.

2. La tab gateava su robot.active==g1 && wrist_side==right con banner
   "Attiva G1 right" e POST /activate. Confonde: la selezione del
   robot e' globale e non si flippa per-feature. Rimossi banner,
   bottone, endpoint /activate e _guard_g1_right(); gli endpoint
   operano sul polso destro del robot attivo qualunque esso sia.

3. image_size: [0,0] perche' Nuwa /api/intrinsics non include
   width/height; vivono in /api/status > data.resolution. Aggiunto
   NuwaClient.image_size() e usato in detect_front. Aggiornato
   docstring di intrinsics() per coerenza con il payload reale.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-21 11:43:05 +02:00
Adriano Dal PastroandClaude Opus 4.7 848cd962c4 fix(gui): conferma esplicita su Muovi quando dry-run off
Coerente con le conferme di Home e Custom: gesto fisico richiede consenso.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 12:47:34 +02:00
Adriano Dal Pastro 8297abb61d feat(gui): right_arm.js — auto-refresh, detect, move, home, custom 2026-05-20 12:43:55 +02:00
Adriano Dal PastroandClaude Opus 4.7 03ecbbae4e fix(gui): allinea CSS tab Braccio DX al tema dark esistente
- Sostituisce colori hardcoded con CSS variables (--bg/--panel/--border/--dim/--warn/--ok/--danger)
- Scope .dim a #tab-right-arm per non alterare altre tab che usano la classe
- Aggiunge white-space pre-wrap a .ra-result per testi multi-riga

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-20 12:40:55 +02:00
Adriano Dal Pastro 2f8c7006ba feat(gui): markup + CSS tab Braccio DX 2026-05-20 12:38:01 +02:00
Adriano Dal PastroandClaude Opus 4.7 b0a2d7141d feat(find_model gui): toggle to overlay work plane as a violet polygon
The find_model response now returns plane_polygon_px (4 corners of a
30 cm square anchored at the work plane centroid, projected to RGB
pixels via the same K used for the SAM21 cloud). The GUI gains a
'Piano (viola)' checkbox that draws the polygon as a translucent
violet quad with an outline, helping the operator see whether the
plane sits where expected relative to the object.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-13 11:53:16 +02:00
Adriano Dal PastroandClaude Opus 4.7 2fe7a7a7fa feat(find_model gui): toggle to overlay full SAM21 cluster (yellow)
Adds a 4th checkbox 'Punti SAM21 (gialli)' that draws the full SAM21
segmentation (input to both the plane partition and Open3DModel) on
the RGB result canvas. The backend now returns sam21_points_px and
n_sam21 alongside the existing accepted/excluded splits.

Useful as a sanity check: visualise what was actually segmented vs
what the plane filter kept/rejected vs where the STL got placed.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-13 11:40:24 +02:00
Adriano Dal PastroandClaude Opus 4.7 767d91acdc feat(work_plane gui): show fit inliers as blue scatter in 3D, drop wrong plane mesh
The 0.4x0.4 PlaneGeometry square was a fixed-size proxy that did not
reflect the actual ROI extent — visually misleading. Replace it with
an explicit blue point cloud of the inliers fed to fit_plane_svd, so
the user can see *exactly* which depth pixels the plane was fit on.

- Backend: fit response now returns fit_points_camera (CV frame) instead
  of the (parallax-shifted) RGB pixel projection.
- Frontend: drop the purple plane mesh; new three.setFitInliers paints
  the inliers in blue. The RGB overlay no longer carries blue dots.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-13 11:20:03 +02:00
Adriano Dal PastroandClaude Opus 4.7 ae510e4387 feat(work_plane gui): overlay inliers used for fit as blue dots on RGB
The fit endpoint now returns fit_points_px (the post-outlier-filter
inliers projected to RGB pixels via Nuwa's RGB intrinsics). The GUI
draws them on the same overlay as the ROI rectangle so the user can
visually verify which depth pixels were fed to fit_plane_svd.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-13 11:10:25 +02:00
Adriano Dal PastroandClaude Opus 4.7 76e0739620 fix(work_plane): allow negative offset_mm + readable validation errors
- Drop the ge=0 constraint on FitReq.offset_mm (negative values are a
  documented usage: aggressive noise cull above the plane).
- HTML input min lowered to -200 to match.
- Render FastAPI 422 validation arrays as a human-readable string in the
  status chip (was showing "[object Object]").

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-13 11:03:36 +02:00
Adriano Dal PastroandClaude Opus 4.7 92506d62c6 fix(work_plane gui): preserve ROI overlay across RGB live-refresh
Setting canvas.width/height clears its bitmap; the live RGB tick was
firing the img.load handler every second, wiping the ROI rectangle and
breaking the two-click flow (third click was interpreted as restart).

- Only mutate canvas size when image resolution actually changes.
- Re-paint the overlay after each load so the ROI persists.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-13 10:58:41 +02:00
Adriano Dal PastroandClaude Opus 4.7 33d66019ee feat(web): Cerca modello — model picker + find_model overlay handlers
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-13 10:12:46 +02:00
Adriano Dal PastroandClaude Opus 4.7 af8d4d6fc4 feat(web): rebuild Cerca modello tab — model picker + result canvas
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-13 10:10:35 +02:00
Adriano Dal PastroandClaude Opus 4.7 56f8888217 fix(gui): work_plane 3D view = camera POV + red = excluded points
- Camera at origin (CV frame) looking at +Z with screen-up = CV −Y, so
  the rendered scene matches the camera's vantage point. OrbitControls
  retargets to the plane centroid on fit so rotation feels natural.
- Simplify scatter coloring to two tones: red for filter-excluded points
  (signed_distance < −offset_mm/1000), cyan-white for everything kept.
  Matches the backend filter contract exactly.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-11 17:41:03 +02:00
Adriano Dal PastroandClaude Opus 4.7 652fa4b912 fix(gui): two-click ROI on Calibrazione Piano (was shift+drag)
User asked for "2 punti" — replace the drag-to-rectangle interaction with
two discrete clicks: first click places corner A (cross marker), second
click defines the rectangle, third click starts a new selection.
Rectangles smaller than 10×10 px are rejected. Status chip guides the
flow.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-11 17:38:43 +02:00
Adriano Dal PastroandClaude Opus 4.7 b4db712cd0 fix(gui): work_plane RGB feed via same-origin proxy
Direct fetch to http://localhost:8877/api/rgb fails from the browser due
to CORS (page is served from :8890). Switch to /api/calibration/rgb.jpg
which presa-mano already proxies same-origin.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-11 17:36:56 +02:00
Adriano Dal PastroandClaude Opus 4.7 57bd252375 fix(gui+api): cleanup before merge
- Rename "TrovaUtensile" tab → "Cerca modello".
- Remove the "Precisione (ultime 5 detection)" stats panel and the
  "Dettagli" JSON dump from the grasp tab; keep hidden stubs so the
  existing app.js continuous-loop code doesn't null-deref.
- Add work_plane_routes to api/__init__.py imports + __all__ (consistency
  with the other routers).
- Trim the misleading "drag = pan · wheel = zoom · doppio click = reset"
  hint from the Calibrazione Piano RGB label — pan/zoom isn't wired.
- Disable wpClearBtn initially so it doesn't flash clickable before
  fetchSaved() resolves.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-11 17:23:29 +02:00
Adriano Dal PastroandClaude Opus 4.7 ae84739c64 feat(gui): Calibrazione Piano tab (ROI + fit preview + 3D viewer)
New tab replaces the removed Servo Visuale. Shift+drag on the RGB image
defines a rectangle, 'Calcola piano' calls POST /api/work_plane/fit and
renders the plane + a colored scatter in 3D. Adjusting the offset slider
re-colors the scatter client-side. 'Salva piano' persists; 'Rimuovi piano
salvato' clears.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-11 17:08:32 +02:00
Adriano Dal PastroandClaude Opus 4.7 eb53f30e23 refactor: remove visual-servo feature
The marker-relative servo (endpoints /api/grasp/visual_servo/{plan,step,
plane/*}, helpers in grasp_routes.py, the bulk of grasp/visual_servo.py,
web/servo.js, and the Servo Visuale GUI tab) is gone. Plane fitting now
lives in /api/work_plane/* and perception/work_plane.py; Task 5 wires the
new Calibrazione Piano tab.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-11 17:04:00 +02:00
Adriano Dal PastroandClaude Opus 4.7 ba0259aa90 feat(gui): surface SAM21 source + n_points in TrovaUtensile status bar
Status line on "Pianifica grasp" now shows source=sam21 / source=bbox_roi
(with fallback reason if any) plus n_points, so SAM21 vs fallback path is
visible at a glance without opening DevTools.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-11 15:52:04 +02:00
Adriano Dal PastroandClaude Opus 4.7 0c211630cc refactor: rimuove feature Auto-Calibrazione (tab + endpoint + job runner)
Elimina il flusso di acquisizione griglia automatica: tab GUI, modulo
autocal3d.js, AutoRunner/auto_job, e gli endpoint /auto/* + /auto_grid.
La calibrazione resta operabile via sample manuali + /solve.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-11 11:03:23 +02:00
Adriano Dal PastroandClaude Opus 4.7 f6197488f9 refactor: cleanup GUI + AI movement controls + depth-correction polish
GUI:
- Rimossa tab "Calibrazione" (sostituita dalla calibrazione via AI con saved
  task `calibrazione_dx`/`calibrazione_sx`, oppure dalla tab "Auto-
  Calibrazione" per il flusso a griglia automatica). Backend
  calibration_routes invariato — tool AI continuano a usarlo.

AI tools:
- Esposto `precise` su `move_wrist_absolute` e `move_wrist_relative`
  (forwarded al campo omonimo del bridge robot, controlla la modalità di
  controllo: false=veloce/tolleranza ampia, true=lento/posa accurata).
- Aggiunto `arms_go_neutral` come tool dedicato (POST /arms/go_neutral del
  bridge): usato dall'AI per "torna a casa / posizione di riposo". Niente
  side, agisce su entrambe le braccia.
- Fix `device_status`: ora wrappa la risposta con `_ok(...)` così il
  dispatcher la classifica correttamente.

Prompt:
- System prompt aggiornato: l'AI ora passa SEMPRE `precise=true` di default
  sui movimenti, escluso quando l'utente chiede esplicitamente velocità.
  Sezione dedicata su come tornare a riposo (arms_go_neutral).
- Documentato che `measured_side_mm` (ritornato da get_marker_3d) viene
  distorto dalla correzione affine sul depth — non usarlo per validare la
  size_mm dichiarata.

Saved tasks:
- `calibrazione_dx` e `calibrazione_sx` ora partono e finiscono con
  `arms_go_neutral`, cancellano i sample preesistenti per quel lato, e
  acquisiscono ≥10 sample con `precise=true` su una griglia ±80mm + ±15° di
  rotazione attorno alla posa home.

Config:
- Marker corrente: id=10, size=40mm (cambiato il tag fisico sul polso).
- Addendum prompt allentato: niente più "max 3 righe", ora chiede di
  riportare il risultato di OGNI tool chiamato.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-06 17:05:52 +02:00
Adriano Dal PastroandClaude Opus 4.7 ac573505b7 feat: initial commit — orchestratore presa-mano con AI assistant + dual-robot
Pipeline completa per calibrazione hand-eye e grasp planning su robot
Unitree G1/H1 con camera depth Nuwa-HP60C. Esposta come FastAPI + GUI vanilla
JS, con assistente AI (OpenRouter + tool calling curato) per pilotare il
sistema in linguaggio naturale.

Architettura:
- FastAPI orchestrator (porta da .env, default 8890) che coordina 4
  microservizi upstream: Nuwa-HP60C, marker-finder, YOLO, robot bridge.
- Dual-robot: services.g1 e services.h1 in config.yaml; switchabili a
  runtime via PUT /api/home/robot.
- Calibrazione hand-eye eye-to-hand split per-side (right/left).
- Marker pose: marker-finder riceve gli intrinseci RGB della Nuwa al boot,
  PnP server-side con quei dati. Nessun re-PnP locale richiesto.
- Depth-to-RGB: registrazione iterativa IR↔RGB per back-projection 3D
  corretta (Nuwa /api/measure usa coords IR; il nostro /api/perception/
  measure_rgb prende pixel RGB e ritorna 3D nel frame RGB OpenCV).
- Correzione affine sul depth bias (calibrabile via marker come
  ground-truth) — riduce errore Z da ~6% a ~1.5% nei test.
- AI assistant: 16 tool curati (move_wrist_*, hand_action, get_marker[_3d],
  detect_object, calibration_*, set_active_robot…), system prompt
  dinamico che riflette robot attivo + marker config + addendum
  customizzabile da GUI. Temperature di default 0.1 per evitare invenzioni.

GUI (7 tab):
- Home: stato comunicazione device + selettore robot attivo G1/H1
- Assistente AI: chat con OpenRouter, history persistita lato browser,
  task riutilizzabili (prompt salvati), trace tool calls
- Impostazioni AI: modello da combo curata, temperature, max iterations,
  prompt addendum, anteprima system prompt
- Calibrazione, Auto-Calibrazione, TrovaUtensile, Servo Visuale, Vista 3D,
  Movimento EE — invariate dalle versioni precedenti

Setup:
- Variabili in presa-mano/.env (template in .env.example)
- OPENROUTER_API_KEY richiesta solo per il tab Assistente AI
- PRESA_MANO_PORT configurabile

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-06 13:28:23 +02:00