What is RosFit?
RosFit is an open-source bridge and management platform that connects ROS 2 robots to the cloud via MQTT. It gives robotics teams a production-ready stack for telemetry collection, remote command execution, device shadows, OTA firmware updates, and a real-time fleet dashboard — without writing a single line of infrastructure code.
Problem
Building cloud connectivity for robots today means stitching together dozens of bespoke tools. Teams spend months on plumbing instead of shipping autonomy features.
| Pain | Today's reality | RosFit solution |
|---|---|---|
| Connecting ROS 2 to the cloud | Custom MQTT bridges take 3+ months to build and harden | Drop-in YAML-configured bridge, production-ready on day one |
| Monitoring a fleet | Build custom dashboards from scratch per project | Pre-built React dashboard with telemetry charts, fleet map, and command panel |
| Sending commands to robots | Ad-hoc WebSocket hacks or SSH tunnels | Structured command API with audit trail and acknowledgement |
| Debugging remote robots | SSH into each robot individually | Centralized logs, live topic echo, and device shadows from the dashboard |
| Firmware updates | Custom update scripts, manual USB flashing | Managed OTA with versioning, rollback, and progress tracking |
| Security | Open MQTT brokers with no auth | TLS, JWT, X.509 mTLS, and role-based access control out of the box |
Solution
RosFit replaces all of the above with a single, composable platform. A lightweight bridge runs on each robot and maps ROS 2 topics to MQTT. A cloud-side message handler persists telemetry to TimescaleDB, manages device shadows in Redis, and stores firmware artefacts in MinIO. The FastAPI backend exposes a REST + WebSocket API that powers both the React dashboard and your own integrations. The entire stack runs with docker compose up for local development or deploys to Kubernetes for production fleets.
Who is it for
- Robotics engineers who want to focus on autonomy, not cloud infrastructure.
- Fleet operators who need a single pane of glass across dozens or hundreds of robots.
- IoT developers connecting ESP32, Raspberry Pi, or any MQTT device to a unified platform.
- Research labs that need reproducible, self-hosted robot management without vendor lock-in.
- Warehouse and factory teams running mixed fleets of AMRs, arms, and sensor nodes.
Key capabilities
- Real-time telemetry — Stream sensor data from every robot to a time-series database and visualize it live on the dashboard.
- Remote command execution — Send structured commands (navigation goals, joint trajectories, parameter updates) to any device through the API or dashboard.
- Fleet management — Group devices by location, type, or project. View aggregate fleet health at a glance.
- Device provisioning — Register new robots via the API or CLI. Automatic certificate generation for mTLS.
- Device shadows — AWS IoT-style desired/reported state management. Update configuration even when a robot is offline — it syncs when it reconnects.
- OTA firmware updates — Upload firmware, target a device or group, track rollout progress, and roll back if something goes wrong.
- Role-based access control — Admin, operator, and viewer roles. Scoped API keys for CI/CD and service integrations.