Skip to content

BACKEND.md

Latest commit

 

History

History
69 lines (51 loc) · 2.79 KB

BACKEND.md

File metadata and controls

69 lines (51 loc) · 2.79 KB
Raw

Backend

FastAPI entrypoint: backend/main.py (uvicorn backend.main:app).

Layering: API → Service → infrastructure (repository or HTTP client). Coding rules: ../backend/README.md.

System context (frontend, ROS, session lifecycle): architecture.md.

Repository layout

backend/          # FastAPI server
frontend/         # Web UI
ros2_ws/src/      # ROS 2 packages
config/           # Machine-local JSON (calibration.json gitignored)
docs/             # Project documentation

Modules

Path Role
backend/modules/patients/ Patient CRUD
backend/modules/sessions/ Clinical sessions per patient
backend/modules/assessments/ Assessment forms
backend/modules/runtime/ SessionManagerClient + RuntimeService → session manager :9090

ROS packages referenced by runtime: gressus_session, gressus_pgear, gressus_bringup — details in ROS.md.

Runtime API

Proxied to session manager (SESSION_MANAGER_URL, default http://127.0.0.1:9090). Response shapes: backend/modules/runtime/schemas.py.

Endpoint Purpose
GET /api/runtime/status Session manager snapshot
POST /api/runtime/stack/start Launch ROS stack (job: feedback, game, calibrate_apriltag)
POST /api/runtime/stack/stop Stop active launch
POST /api/runtime/pgear/load-profile Load P.GEAR exo profile JSON
POST /api/runtime/pgear/arm Arm exoskeleton
POST /api/runtime/pgear/disarm Disarm
POST /api/runtime/pgear/run Start gait
POST /api/runtime/pgear/stop-gait Stop gait
POST /api/runtime/pgear/estop Emergency stop
POST /api/runtime/pgear/estop-reset Clear faults after E-STOP
POST /api/runtime/pgear/full-cal ODrive FULL CAL (DISARM only)
POST /api/runtime/pgear/calibrate-baseline Empty-exo baseline fit (~30 s, ARM+RUN)
POST /api/runtime/start Legacy alias → stack start (game modes)
POST /api/runtime/stop Legacy alias → stack stop

sessionId and patientId on stack start are forwarded to session manager and exported as ROS env vars — see architecture.md.

Docker

Backend runs in the backend compose service on port :8000. Startup and compose overview: ../README.md.

Dependencies: backend/pyproject.toml (FastAPI, httpx, SQLAlchemy, asyncpg, Alembic).

Data flow (backend-centric)

Frontend ──REST──► Backend
                      │
                      ├── PostgreSQL (patients, sessions, assessments)
                      │
                      └── RuntimeService ──HTTP──► session_manager (:9090)

Insole WebSocket and ROS topics bypass the backend — see architecture.md.