Let’s Stop Talking Exclusively about Cobot Safety, Neglecting Cobot Security

How many articles have you read about robots and cobots safety? How many times have you heard people (including us, from Cobotics World) talking about it? Are collaborative robots truly safe? Does a cobot remain safe even if I fix a screwdriver or a drill at the end of them?
Although important this is not the only question that matters. There is another, non-negligible factor, that could eventually be extremely connected with safety: security.

As written in [1], cobots supplier Acutronic allowed Alias Robotics to perform an in-depth analysis of its modular cobot MARA. This study revealed multiple weaknesses and 27 exploitable vulnerabilities in the robot. Even if Acutronic Robotics promptly reacted, ensuring security improvements and transforming MARA in probably the most secure cobot of the market, it is very likely that many other cobots share the same security issues.

Now, when we hear the word “security” we immediately think about protecting data and information that could be stolen from our company. But let’s imagine if someone could take control of the cobots that are working side by side with our employees in our factory. What if this mysterious hacker would start modifying the safety parameters of our collaborative systems? He could, for example, deactivate the force control that makes a cobot stop when a collision is detected, with serious impact on operators safety.

Researchers from Alias Robotics warn that most collaborative robots are totally insecure: “most manufacturers simply don’t think about security”, but they are not the only one that raises the issue. Robotic Industries Association, in [2] pointed out how vulnerabilities in cybersecurity undermine the entire purpose of collaborative robots and put people in harm’s way. We could ask cobot suppliers to design every robot in such a way that they are harmless even in the worst case possible, but this will necessarily represent a huge roadblock on the path of cobots employment.

So, while last week Robotics Industries Association defined new test metrics [3] and methods with the purpose of outlining the optimal testing methods for power and force in power-and-force-limited cobot systems in order to guarantee safety for this kind of system, there is still so much to do for what that concerns cobot security. Where to start? Probably reading the article [4]  by the Robotiq‘s blog about Do’s and Don’ts for Protecting Your Cobots’ Cybersecurity

[1] https://www.prosyscom.tech/robotics/collaborative-robot-insecurities-exposed-mara/

[2] https://www.robotics.org/blog-article.cfm/Collaborative-Robots-and-Cybersecurity-Concerns/66

[3] https://www.mmh.com/article/ria_standards_committee_releases_all_new_methods_and_metrics_for_collaborat

[4] https://blog.robotiq.com/protecting-the-cybersecurity-of-your-cobots-dos-and-donts