Tactic: Initial Access

Technique: Compromise Controller

Compromising the robots controller is a effective way of gaining control over the robot. This can be done in a few ways:
- Recording and replaying the commands that the controller sends to the robot.
- Hacking the controller itself by overwriting certain parts of the code used in the controller.
- DOS'ing the controller communication to the robot. This will cause the robot to not respond to the operator.
- Having the same controller sending commands at the same time blocking or even breaking the robots communication.

Technique in practice

Mitigations

As for mitifations a few additions in the robots security can fix these issues:
- Sending ID's or keys with each command send so the robot can verify and check if the command is a new one and authorized to send this command.
- Keeping the controller in a safe location.
- Make it possible to interact with the robot in multiple ways to minimize the DOS attack risk and making it return upon encountering loss of communication.
- Pair a controller to the robot is a safe way to make sure the same type of controller cant control the robot you are using.

Detections

Detecting the multiple ways a controller can be compromised depends on the attack:
- With the ID system in place for messaging between the controller and robot, it is easy to detect fake or forged commands.
- Loss of communication is a sure tell way of detecting a DOS attack.
- Having the controller be paired to the robot makes detecting a roque controller easier as it does not have authorized permission to command the robot.

Documented incidents with autonomous robots

No incidents known at the time of writing.

Documented incidents in other domains