Improving real-time safety checks for autonomous system controls
Real-Time Synthesis of Robust Controlled Invariant Sets for Monotone Systems
Artificial IntelligenceRobotics
Summary
Keeping autonomous systems safe while they operate needs quick safety checks that update as things change. The authors show a new way to do these checks faster for certain systems where safety rules have specific mathematical shapes. Their method breaks down the problem into many small searches that can run at the same time, making the process much quicker. Experiments showed this method can check safety on enormous grids within seconds or minutes, even while the system is running.
What this means in practice
- •For autonomous vehicle engineers: Generate safety guarantees online for self-driving cars adapting to changing environments quickly and reliably with scalable algorithms.
- •For robotics control teams: Enable rapid synthesis of safe operating regions for robot motion planning under dynamic conditions using efficient invariance checking.
Authors
Yasin Sonmez, Mahmoud Khaled, Majid Zamani, Murat Arcak
Abstract
Safety-critical control of autonomous systems requires formal safety certificates, such as controlled invariant sets, that must be computed online as conditions change. Although standard synthesis algorithms scale poorly with state dimension, monotone dynamical systems with lower-closed safety specifications allow for accelerated computation of controlled invariant sets. In particular, lazy fixed-point algorithms exploit monotonicity and track only the antichain basis of the set. However, membership tests and redundancy checks against an evolving basis remain major bottlenecks. We introduce a threshold-function reformulation in which a lower-closed set on a d-dimensional grid is represented by its column heights along a designated axis. This reformulates the greatest-fixed-point iteration as independent one-dimensional binary searches, one per grid column, yielding an embarrassingly parallel iteration with asymptotically lower computational complexity than the lazy fixed-point algorithm. Experiments synthesize invariant sets on 3D grids with 10^9 cells in under 50 ms and 10^14 cells in under two minutes. We further demonstrate online re-synthesis in a safety-informed model predictive controller example.