Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a method for adapting probabilistic system models to handle out-of-Operational Design Domain (ODD) scenarios, enhancing the resilience of autonomous systems. The approach dynamically extends the coverage of system situation capabilities, enabling verification and adaptation of system behavior in unanticipated conditions. Preliminary results demonstrate increased system reliability through behavior adaptation and formal guarantees, even when faced with unforeseen out-of-ODD situations.
Guaranteeing safety outside a robot's training data is now possible with a new method that dynamically adapts its probabilistic model of the world.
Ensuring reliable performance in situations outside the Operational Design Domain (ODD) remains a primary challenge in devising resilient autonomous systems. We explore this challenge by introducing an approach for adapting probabilistic system models to handle out-of-ODD scenarios while, in parallel, providing quantitative guarantees. Our approach dynamically extends the coverage of existing system situation capabilities, supporting the verification and adaptation of the system's behaviour under unanticipated situations. Preliminary results demonstrate that our approach effectively increases system reliability by adapting its behaviour and providing formal guarantees even under unforeseen out-of-ODD situations.