Event
Study proposes goal insertion attack; SIREN can cause humanoid robot collisions
1 reports · 1 sources
Tiffany Bao and other authors propose a "goal insertion attack" targeting humanoid robot safety filters and develop an automated tool, SIREN. The paper states that the mathematical guarantees of such safety filters hold only when safety parameters are configured correctly, and that the parameters depend on the specific task and environment and may fail after deployment. The authors state that SIREN can automatically find vulnerable goals that can be inserted for a given task and environment, identifying attacks in all vulnerable scenarios in the SPARK dataset, three times the number found by baseline methods; they also generated attacks on a real humanoid robot, causing it to collide while executing tasks.
Report timeline
- Study Reveals Flaws in Humanoid Robot Safety Filters; SIREN Can Automatically Generate Collision Attacks
The paper points out that the mathematical guarantees of humanoid robot safety filters hold only when parameters are configured correctly, while the parameters depend on the specific task and environment and may fail after deployment. The authors propose goal insertion attacks and develop an automated tool, SIREN, which identified attacks in all vulnerable scenarios in the SPARK dataset, three times the number found by baseline methods, and triggered collisions on a real humanoid robot.
arXiv cs.RO ·