[Paper Review] Robotics CTF (RCTF), a playground for robot hacking

Gorka Olalde Mendia, Lander Usategui San Juan+10 more|arXiv (Cornell University)|Oct 1, 2018

Advanced Malware Detection Techniques22 references4 citations

TL;DR

This paper introduces Robotics CTF (RCTF), an online, browser-based platform that enables white-hat hackers to exploit security vulnerabilities in simulated robotic systems using real-world attack techniques. Built on containerized environments and ROS/ROS2, RCTF provides 9 progressively complex scenarios focused on common flaws like cleartext data transmission, hardcoded credentials, and insecure command execution, with all scenarios open-sourced to foster collaborative robot security research.

ABSTRACT

Robots state of insecurity is onstage. There is an emerging concern about major robot vulnerabilities and their adverse consequences. However, there is still a considerable gap between robotics and cybersecurity domains. For the purpose of filling that gap, the present technical report presents the Robotics CTF (RCTF), an online playground to challenge robot security from any browser. We describe the architecture of the RCTF and provide 9 scenarios where hackers can challenge the security of different robotic setups. Our work empowers security researchers to a) reproduce virtual robotic scenarios locally and b) change the networking setup to mimic real robot targets. We advocate for hacker powered security in robotics and contribute by open sourcing our scenarios.

Motivation & Objective

• To address the growing gap between robotics and cybersecurity by creating an accessible, hands-on platform for offensive security research in robotics.
• To empower security researchers and roboticists to reproduce and test robot security flaws in isolated, customizable virtual environments.
• To promote hacker-powered security in robotics through gamified Capture The Flag (CTF) challenges accessible via web browser.
• To encourage community contributions by open-sourcing 9 initial security scenarios and providing a framework for new scenario development.
• To raise awareness about critical vulnerabilities in robot frameworks like ROS and ROS2, particularly those stemming from poor out-of-the-box security configurations.

Proposed method

• RCTF is implemented as a containerized, stateless platform using Docker and Docker Compose, with each scenario deployed as an isolated, self-contained environment.
• Each scenario emulates real robotic systems using ROS and ROS2, with intentional security flaws such as unencrypted communication, hardcoded credentials, and command injection vulnerabilities.
• The platform provides web-based access to terminal sessions and Gazebo visualization environments, enabling interactive exploitation and observation of attack outcomes.
• Scenarios are designed with a 'Jeopardy' CTF mode, where participants solve challenges to capture 'flags' by exploiting specific vulnerabilities.
• The architecture avoids centralized orchestration, enabling easy migration across cloud platforms and simplifying deployment and management.
• All scenarios are hosted on GitHub at https://github.com/aliasrobotics/rctf-list, with a contribution workflow via pull requests for community-submitted challenges.

Experimental results

Research questions

• RQ1 How can a scalable, browser-accessible platform be designed to enable ethical hackers to explore and exploit real-world vulnerabilities in robotic systems?
• RQ2 What types of security flaws are most prevalent and exploitable in ROS and ROS2-based robotic systems under controlled, isolated conditions?
• RQ3 To what extent can gamified CTF-style challenges effectively train and engage security researchers in robot-specific attack surfaces?
• RQ4 How can open-sourcing robotic security scenarios foster community-driven improvement and broader adoption in robot security research?
• RQ5 Can containerized, self-contained environments effectively replicate real-world robot networking and attack surfaces without requiring physical hardware?

Key findings

• RCTF successfully provides a 24/7, browser-based environment for ethical hackers to explore and exploit security vulnerabilities in ROS and ROS2-based robotic systems.
• Nine distinct security scenarios were developed, each targeting common vulnerabilities such as cleartext data transmission (CWE-319), hardcoded credentials (CWE-788), and improper command neutralization (CWE-78).
• The platform enables local reproduction of scenarios and flexible network configuration, allowing researchers to emulate real robot deployment environments.
• The use of containerized, isolated deployments without centralized orchestration simplifies deployment and enhances portability across cloud and on-premise infrastructures.
• The open-sourcing of all scenarios via GitHub has enabled community contributions and integration, with a clear workflow for submitting new challenges.
• The platform demonstrates that gamified, accessible CTF environments can effectively raise awareness and drive engagement in robot security research.