publications

2026

1. ICUAS26

   From Language to Logic: A Theoretical Architecture for VLM-Grounded Safe Navigation

   Kristy Sakano, Kalonji Harrington, and Huan Xu

   In 2026 International Conference on Unmanned Aircraft Systems (ICUAS), 2026

   Abs DOI Slides

   We propose an architecture for integrating high-level, human-provided safety rules and operator-aligned semantic preferences into autonomous robot navigation in unstructured outdoor environments. In our approach, natural-language rules are translated into Signal Temporal Logic (STL) specifications that guide planning and navigation during runtime. Persistent, environment-centric rules and terrain preferences are grounded into a 2D cost map, while temporally dynamic requirements are expressed as STL specifications to be monitored during runtime. We hypothesize the use of Vision-Language Models (VLMs) for zero-shot scene understanding, enabling mapping between human instructions, semantic features, and environmental constraints. Within this framework, we construct an illustrative navigation model that is designed to satisfy a set of STL-encoded specifications and soft operator preferences through formal satisfaction metrics embedded into environmental properties and runtime monitoring.

2. ARR

   DRIVE-SAFE: Data-Driven Robustness and Informed Validation for Evolving Specifications via Formal Evaluation

   Kristy Sakano, Jianyu An, Dinesh Manocha, and 1 more author

   Advanced Robotics Research, 2026

   Abs DOI PDF

   We present a regulator-driven framework for the temporal evaluation of learning-based, black-box autonomous robot policies under evolving safety requirements. Motivated by real-world certification processes in which regulators often assess observable behavior in inaccessible black-box policies, we simulate a scenario where deployed policies must maintain ongoing compliance as human-defined safety rules change over time. Natural-language safety requirements are translated into Signal Temporal Logic (STL) specifications, which are then used to quantitatively evaluate rollout traces from black-box policies. We compute Total Robustness Value (TRV), Largest Robustness Value (LRV), and Average Violation Robustness Voting (AVRV), to summarize average performance, worst-case violation, and average specification violation across a set of rollout trajectories. These metrics are combined with domain-informed importance weights to provide targeted feedback to model designers, indicating where retraining or redesign is most needed. In a virtual driving scenario, iterative retraining increased satisfaction of a global speed limit specification by 72% and a slippage mitigation specification by 32%. When the speed-limit requirement was tightened, the resulting retrained model increased satisfaction by 47% but revealed trade-offs with competing safety objectives. Overall, our approach integrates temporal-logic-based robustness metrics with statistically meaningful evaluation, bridging regulator oversight and model improvement under evolving safety requirements.

2025

1. ICUAS25

   A Framework for Black-Box Controller Design to Automatically Satisfy Specifications Using Signal Temporal Logic

   Kristy Sakano, Joe Mockler, Alexis Chen, and 1 more author

   In 2025 International Conference on Unmanned Aircraft Systems (ICUAS), 2025

   Abs DOI Slides

   We present a framework for including humanreadable specifications in the design of black-box autonomous systems, or systems whose construct are prohibitively complex to analyze or intuit. By integrating parametric Signal Temporal Logic (pSTL), we can systematically evaluate and refine deep reinforcement learning policies to ensure compliance with predefined system specifications. Our approach is tested in a simulated autonomous driving environment, where we train a deep reinforcement learning agent in Mario Kart SNES using Proximal Policy Optimization. The agent is evaluated based on its ability to navigate a structured driving course while satisfying a set of a priori requirements; this evaluation is performed by writing and solving the parameters in a pSTL statement. This work contributes to the broader effort of bridging formal methods and data-driven learning, providing insights for researchers and operators alike in developing AI based controllers for real-world autonomous systems.

2022

1. MNRAS

   η Geminorum: An Eclipsing Semiregular Variable Star Orbited by a Companion Surrounded by an Extended Disc

   Guillermo Torres and Kristy Sakano

   2022

   Abs DOI PDF

   We report 11 yr of spectroscopic monitoring of the M-type asymptotic giant branch star η Gem, a semiregular variable and a known spectroscopic binary with a period of 8.2 yr. We combine our radial velocities with others from the literature to provide an improved spectroscopic orbital solution, giving a period of 2979 d, which we then use to predict past times of eclipse. We examine archival photometry from amateur variable star observers, and other sources, and find many instances of dimmings that occurred at the right time. This confirms previous indications that the system is eclipsing, and it now ranks among those with the longest known periods. No secondary eclipses are seen. The ∼0.4 mag eclipses that last about 5 months are much too deep to be produced by a stellar companion. We propose instead that the companion is surrounded by a large disc that is at least 1.5 au in diameter but is likely larger. We predict the centre of the next eclipse will occur on New Year’s Day, 2029.

2. IEEE AERO

   Onboard Hyperspectral Image Classification via Transfer Learning for Communication-Limited Spacecraft

   Justin S. Goodwill, James P. MacKinnon, Kristy Sakano, and 1 more author

   In 2022 IEEE Aerospace Conference (AERO), 2022

   DOI PDF