UXDuck
UXDuck is a ground control station for teams building autonomous vehicles, offering a browser-based ArduPilot simulator and a desktop app for connecting to real drones, rovers, boats, and submarines. It provides a full history of parameter changes, mission planning and verification, and MCP integration that lets AI agents like Claude Code read telemetry and stage changes for operator approval. The platform supports MAVLink forwarding to Mission Planner or QGroundControl and runs on macOS, Windows, Linux, and Raspberry Pi.
- Autonomous Vehicles
- Drones & Uncrewed Systems
- Developer Tools
- Robotics
- Software Only
- Artificial Intelligence
Funding
Raised to date
$500KRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Founder details are not available yet.
Product
Problem
Developing software for autonomous vehicles such as drones, rovers, boats, and submarines is risky and time-consuming. Testing code directly on physical hardware is expensive, slow, and can lead to crashes or equipment damage, while existing simulation tools often lack support for multiple vehicle types or seamless integration with modern development workflows.
Solution
UXDuck provides a sandboxed development platform where engineers can code, simulate, and test autonomous vehicle software before deploying to real hardware. The platform supports planes, copters, rovers, boats, and submarines, allowing teams to validate behavior across different vehicle types in one environment. UXDuck includes a built-in simulation engine that runs on the developer's computer, with an in-app AI chat and MCP (Model Context Protocol) integration that lets external AI agents read live telemetry, parameters, and vehicle messages to assist with debugging and analysis. The platform is designed to help developers iterate quickly and launch with confidence, reducing the risk of failures in the field.
Target Audience
Primary users are software engineers and robotics developers building autonomous vehicle applications for drones, rovers, boats, and submarines, particularly those working in industries such as agriculture, surveying, defense, and research.
Features
- Sandboxed simulation environment supporting planes, copters, rovers, boats, and submarines
- MCP integration enabling AI agents to read vehicle telemetry, parameters, and messages for assisted debugging
- Tool-based permission system with granular approvals for commands and changes, including allow-once and always-allow options
- Telemetry window analysis that retains up to five minutes of signal data for diagnosing voltage dips, GPS issues, and vibration problems
- Forwarding diagnostics with recorded routing and loop evidence to troubleshoot connection issues
- Cross-platform desktop app available for Linux, macOS (Apple silicon and Intel), and Windows
- Browser-based simulation with no install or account required, using real ArduPilot and MAVLink
- Parameter change history with reasons, staging, and protection against overwriting newer values
- Mission planning, upload, read-back verification, and terrain checks
- MAVLink forwarding to Mission Planner or QGroundControl
- CLI and RPC for shell scripts and CI integration
- Video support for H.264, H.265, MPEG-TS over UDP or RTSP, plus webcam and USB capture