Get Started
Research questionHow can runtime safety monitors reduce growing zonotopes without losing precision or missing real-time deadlines?As uncertain sensor measurements accumulate, zonotopes grow and require over-approximating reductions. These reductions trade computational cost against the precision needed to assess whether robot behavior satisfies safety specifications.
Alignment & Safety
Inference Optimization
Robotics
Latest papersRecent research connected to this question, newest first.Predictive Zonotope Reduction: Precise Runtime Monitoring under UncertaintyThe source studies RLola monitoring for a simulated 5-degree-of-freedom robotic arm in MuJoCo, with sensor uncertainty modeled according to ISO 5725. Experiments on a Raspberry Pi 5 compare dynamic and static reduction strategies and report lower false-positive rates for dynamic reduction; the evidence is limited to this setup.research paper · Sep 3, 2026
Related questions
How can tool-using agents reduce serial action–observation latency without sacrificing task completion?How can remote robots maintain reliable control over bandwidth-limited, changing wireless links without transmitting every high-dimensional observation?How can safe reinforcement learning enforce infinitely many continuous-space constraints while remaining computationally tractable?How can lunar robots maintain accurate, bounded-latency instance segmentation under low light, tight compute, and radiation faults?
Home
Topics
Search
Library