GuardROS

A Non-Invasive ROS 2 Integration Framework for the GuardBot RollerBot 23 (RB23)

---

ROS 2 • Teleoperation • Telemetry • Multimedia Streaming • Experiment Logging

---

Overview

GuardROS is a non-invasive ROS 2 interoperability framework developed for the GuardBot RollerBot 23 (RB23) platform.

The framework acts as a middleware integration layer capable of bridging the proprietary RB23 communication ecosystem with modular ROS 2 workflows without requiring modifications to the original firmware or hardware.

GuardROS enables:

• ROS 2 teleoperation
• Real-time telemetry acquisition
• Video streaming
• Bidirectional audio communication
• Experimental logging
• ROS 2 topic integration
• Middleware-based extensibility
• Future autonomous robotics integration

The framework was experimentally validated under real-world operational conditions, demonstrating stable bidirectional communication, multimedia integration, and long-duration operational stability.

---

System Architecture

GuardROS operates as a bidirectional interoperability layer between:

• The proprietary GuardBot communication infrastructure
• ROS 2 applications and robotics workflows

The framework decodes telemetry, video, audio, and operational packets from the RB23 platform and exposes them as ROS 2-compatible interfaces and topics.

It also supports the inverse communication flow, allowing ROS 2 nodes to send teleoperation and autonomous control commands back to the robot.

---

Features

Core Capabilities

• ROS 2 interoperability
• Modular architecture
• Real-time teleoperation
• Structured telemetry access
• Multimedia streaming
• Bidirectional audio
• Experiment logging
• Reproducible experimentation

ROS 2 Integration

• ROS 2 topic mapping
• Middleware-based communication
• External tool compatibility
• Distributed robotics workflows

Future Extensions

GuardROS was designed to support future robotics applications such as:

• SLAM
• Nav2
• Computer vision
• Sensor fusion
• Autonomous navigation
• Distributed robotic systems

---

Workspace Structure

guardros_ws/
├── scripts/
│   ├── run_guardros.sh
│   └── stop_guardros.sh
│
├── experiments/
│   ├── logs/
│   └── analysis/
│       └── analyze_guardros_logs.py
│
├── src/
├── build/
├── install/
└── log/

---

Requirements

Recommended Environment

• Ubuntu 22.04
• ROS 2 Humble
• Python 3
• GuardBot RollerBot 23 (RB23)

---

Installation

1. Clone the Repository

git clone https://github.com/<your-repository>/guardros.git

2. Build the Workspace

cd guardros_ws

colcon build

3. Source the Workspace

source install/setup.bash

---

Configuration

Before running GuardROS, configure the workspace path inside:

~/guardros_ws/scripts/run_guardros.sh

Locate:

WORKSPACE="/home/your_user/guardros_ws"

Replace it with your actual workspace path.

Also verify the ROS 2 installation path:

ROS_SETUP="/opt/ros/humble/setup.bash"

---

Running GuardROS

Make Scripts Executable

chmod +x scripts/run_guardros.sh
chmod +x scripts/stop_guardros.sh

Start the Framework

./scripts/run_guardros.sh

The script will request the ROBOT_ID.

After initialization, multiple terminals are automatically launched, including:

• RB23 Driver
• Keyboard Teleoperation
• Video Viewer
• Telemetry Viewer
• Audio Node
• Experiment Logger

---

Logging System

Experiment logs are automatically stored at:

~/guardros_ws/experiments/logs/

Each session creates a timestamped directory containing logs such as:

control_log.csv
telemetry_log.csv
video_log.csv
audio_log.csv
typed_telemetry_log.csv
diagnostics_driver_log.jsonl

---

Analysis Tools

GuardROS includes an analysis utility:

experiments/analysis/analyze_guardros_logs.py

The script generates:

• Statistical summaries
• Timing analysis
• Jitter analysis
• Battery plots
• Temperature plots
• Ping plots
• PNG figures

Example Usage

python3 experiments/analysis/analyze_guardros_logs.py \
experiments/logs/SESSION_NAME

---

Experimental Validation

The framework was experimentally validated across multiple operational scenarios:

Test	Objective
T1	Idle stability validation
T2	Teleoperation validation
T3	Video stress validation
T4	Audio stress validation
T5	Long-duration operational validation
T6	Native application comparison

Results demonstrated:

• Stable 20 Hz command stream
• Stable telemetry acquisition
• Stable ~20 FPS video streaming
• No UDP reception errors
• Sustained operation without manual reconnection

---

Comparison with Native RB23 Ecosystem

Capability	Native Apps	GuardROS
ROS 2 interoperability	No	Yes
ROS topic integration	No	Yes
Structured telemetry access	Limited	Yes
Experimental logging	No	Yes
Middleware integration	No	Yes
Autonomous robotics integration	No	Supported

---

Recommended Workflow

1. Power on the RB23 robot
2. Wait for network connection
3. Run GuardROS
4. Perform experiments
5. Stop GuardROS
6. Analyze generated logs

---

Current Limitations

The current implementation still depends on the original proprietary GuardBot communication infrastructure and server ecosystem.

Future work aims to reduce this dependency through direct or embedded communication architectures.

---

Future Work

Planned future developments include:

• Embedded communication architectures
• Direct robot communication
• SLAM integration
• Nav2 support
• Sensor fusion pipelines
• Autonomous robotics workflows
• Distributed robotic systems

---

Citation

If you use GuardROS in academic work, please cite:

@article{guardros2026,
  title={GuardROS: A Non-Invasive ROS 2 Integration Framework for the GuardBot RollerBot 23},
  author={Mosconi, Denis and Marsicano, Joao Aires and Becker, Marcelo},
  journal={IEEE},
  year={2026}
}

---

License

Apache-2.0

---

Authors

• Denis Mosconi
• Joao Aires Marsicano
• Marcelo Becker

---

Acknowledgements

This work was carried out with the support of Petrobras, with resources from the ANP R&D clause, in partnership with the University of São Paulo (USP) and FAFQ.

---

Repository Structure Suggestions

docs/
├── images/
│   ├── logo.png
│   ├── architecture.png
│   └── rb23.png