# 🤖 Operating real hardware with Gemini AI

## Executive summary

The video demonstrates a proof-of-concept for operating physical hardware (a robot arm) remotely over a significant distance (1,700 miles). By leveraging Gemini AI, the speakers built a custom 3D web visualizer that translates user input (slider movements) into real-time joint coordinates, enabling distributed control of the physical robot.

## Key takeaways

- Remote Hardware Control Demonstration: A robot arm located in Texas was successfully controlled from New York using a custom 3D web visualizer, bridging a distance of 1,700 miles.
- Real-Time Joint Coordinate Transmission: The system architecture allows the browser to immediately send new joint coordinates whenever a slider is moved in the 3D visualizer.

## Technical details

- AI-Assisted Robotics Interface: Gemini was utilized to process the calibration file required for the physical robot arm, enabling the remote control system to function.
- Web Visualization and Input: The control mechanism relies on a 3D visualizer where user input (slider movements) generates and transmits joint coordinates in real time.

## Practical implications

- Enables distributed control systems for robotics and IoT devices, allowing operation across vast geographical distances.
- Demonstrates the use of generative AI (Gemini) to integrate complex physical hardware constraints (calibration files) into web-based interfaces.
- Provides a blueprint for building remote testing and operational tools for build-engineering environments.

## Topics

Robotics, AI Integration, Distributed Systems, Web Development, Hardware Control, Google AI Studio, Gemini, Google Antigravity

Source: https://www.youtube.com/watch?v=fMgcm9aXZyY