SemuBot WebAPP

Overview

SemuBot WebAPP is a touchscreen-first robot control interface built with FastAPI and a static frontend. The application exposes a REST API that forwards movement commands and chat messages to a ROS system via roslibpy, and serves a mobile-friendly joystick UI.

The project is designed for use with a ROS bridge server (rosbridge_server) and includes a companion ROS2 speech node that listens for messages published to /speech_input.

This repository continues the webapp from Karl Johan Krasilnikov's 2026 thesis, Krasilnikov-thesis-2026-Webapp. The thesis repository keeps the version submitted with the thesis; development continues here.

Features

• Touch joystick and rotation controls for driving the robot

• Quick-chat message publishing to robot

• Status polling for ROS connection and battery level

• Docker-ready deployment with configurable ROS bridge host/port

Architecture

• app/main.py - FastAPI web server and static file host

• app/api/robot.py - HTTP API routes for robot movement, chat, and status

• app/services/ros_client.py - ROS bridge client using roslibpy

• app/static/ - frontend assets, including HTML, CSS, JS, and chat messages

• speech_node/speech_node.py - ROS2 node subscribing to /speech_input and speaking received messages

Quick Start

Installing Docker to the server

Follow the official installation guide: https://docs.docker.com/engine/install/

Installing rosbridge to SemuBot

Install rosbridge-suites, if not already installed:

sudo apt update
sudo apt install ros-<ros2-distro>-rosbridge-suite

Start the ROS bridge server:

ros2 launch rosbridge_server rosbridge_websocket_launch.xml

Run the application

Clone the repository:

git clone https://github.com/SemuBot/semubot_webapp.git
cd semubot_webapp

The application can run either on the robot’s onboard computer or on a separate server within the same network. In either case, the ROSBRIDGE_HOST must be configured in .env accordingly.

From the project root start the application:

docker compose up --build

Then open the web interface on a mobile device:

http://server_ip:8000

Usage

• Load the web app in a touch-capable browser.

• Use the joystick to send linear movement commands.

• Use the rotate left/right buttons to send angular velocity commands.

• Tap the chat button to send predefined chat messages to the speech node.

• The header shows connection status and battery level.

Notes

• The server forwards /api/robot/move payloads directly to ROS on /cmd_vel.

• The speech node must be built and launched within the robot’s ROS2 environment

• The frontend loads api.js, joystick.js, and status.js only on touch-capable devices.

Used ROS Topics

• /cmd_vel (geometry_msgs/Twist) - receives movement commands from the client

• /speech_input (std_msgs/String) - client publishes chat messages to this topic for the speech node to speak

• /semubot/battery (sensor_msgs/BatteryState) - published by the robot's battery monitor; the status endpoint reports percentage (0.0 to 1.0) and voltage

• /semubot/mode (std_msgs/String) - client publishes the speech mode to switch to (manual, local, cloud)

• /semubot/mode_state (std_msgs/String) - published by the brain node with the current speech mode; shown by the status endpoint

• /emergency_stop (std_msgs/Bool) - published by the client to indicate emergency stop status (true for active)

Project Files

• .env.example - example environment variables for ROS bridge configuration

• docker-compose.yml - container orchestration for the web UI

• Dockerfile - builds the Python service image

• requirements.txt - Python dependencies

• app/static/chat_messages.json - chat options displayed in the UI (Change the messages you want to quick send in this file)