arXiv subjects

Melkior Ornik

Publications and source records attributed to Melkior Ornik.

At least 55 records · Page 3Linked to original sources

arXiv2025

Optimizing a Model-Agnostic Measure of Graph Counterdeceptiveness via Reattachment

Recognition of an adversary's objective is a core problem in physical security and cyber defense. Prior work on target recognition focuses on developing optimal inference strategies given the adversary's operating environment. However, the success of such strategies significantly depends on features of the environment. We consider the problem of optimal counterdeceptive environment design: construction of an environment which promotes early recognition of an adversary's objective, given operational constraints. Viewed as a bounded-length graph-design problem, we introduce a metric for counterdeception and a novel heuristic that maximizes it based on iterative reattachment of trees. We benchmark the performance of this algorithm on synthetic networks as well as a graph inspired by a real-world high-security environment, verifying that the proposed algorithm is computationally feasible and yields meaningful network designs.

math.OC↗

arXiv2025

Online Learning and Control Synthesis for Reachable Paths of Unknown Nonlinear Systems

In this paper, we present a novel method to drive a nonlinear system to a desired state, with limited a priori knowledge of its dynamic model: local dynamics at a single point and the bounds on the rate of change of these dynamics. This method synthesizes control actions by utilizing locally learned dynamics along a trajectory, based on data available up to that moment, and known proxy dynamics, which can generate an underapproximation of the unknown system's true reachable set. An important benefit to the contributions of this paper is the lack of knowledge needed to execute the presented control method. We establish sufficient conditions to ensure that a controlled trajectory reaches a small neighborhood of any provably reachable state within a short time horizon, with precision dependent on the tunable parameters of these conditions.

math.OC↗

arXiv2025

Conformal Contraction for Robust Nonlinear Control with Distribution-Free Uncertainty Quantification

We present a novel robust control framework for continuous-time, perturbed nonlinear dynamical systems with uncertainty that depends nonlinearly on both the state and control inputs. Unlike conventional approaches that impose structural assumptions on the uncertainty, our framework enhances contraction-based robust control with data-driven uncertainty prediction, remaining agnostic to the models of the uncertainty and predictor. We statistically quantify how reliably the contraction conditions are satisfied under dynamics with uncertainty via conformal prediction, thereby obtaining a distribution-free and finite-time probabilistic guarantee for exponential boundedness of the trajectory tracking error. We further propose the probabilistically robust control invariant (PRCI) tube for distributionally robust motion planning, within which the perturbed system trajectories are guaranteed to stay with a finite probability, without explicit knowledge of the uncertainty model. Numerical simulations validate the effectiveness of the proposed robust control framework and the performance of the PRCI tube.

math.OC↗

arXiv2025

Mode-Prefix-Based Control of Switched Linear Systems with Applications to Fault Tolerance

In this paper, we consider the problem of designing prefix-based optimal controllers for switched linear systems over finite horizons. This problem arises in fault-tolerant control, when system faults result in abrupt changes in dynamics. We consider a class of mode-prefix-based linear controllers that depend only on the history of the switching signal. The proposed optimal control problems seek to minimize both expected performance and worst-case performance over switching signals. We show that this problem can be reduced to a convex optimization problem. To this end, we synthesize one controller for each switching signal under a prefix constraint that ensures consistency between controllers. Then, system level synthesis is used to obtain a convex program in terms of the system-level parameters. In particular, it is shown that the prefix constraints are linear in terms of the system-level parameters. Finally, we apply this framework for optimal control of a fighter jet model suffering from system faults, illustrating how fault tolerance is ensured.

math.OC↗

arXiv2025

Enhancing Robot Navigation Policies with Task-Specific Uncertainty Managements

Robots navigating complex environments must manage uncertainty from sensor noise, environmental changes, and incomplete information, with different tasks requiring varying levels of precision in different areas. For example, precise localization may be crucial near obstacles but less critical in open spaces. We present GUIDE (Generalized Uncertainty Integration for Decision-Making and Execution), a framework that integrates these task-specific requirements into navigation policies via Task-Specific Uncertainty Maps (TSUMs). By assigning acceptable uncertainty levels to different locations, TSUMs enable robots to adapt uncertainty management based on context. When combined with reinforcement learning, GUIDE learns policies that balance task completion and uncertainty management without extensive reward engineering. Real-world tests show significant performance gains over methods lacking task-specific uncertainty awareness.

cs.RO↗

arXiv2025

The Lazy Student's Dream: ChatGPT Passing an Engineering Course on Its Own

This paper presents a comprehensive investigation into the capability of Large Language Models (LLMs) to successfully complete a semester-long undergraduate control systems course. Through evaluation of 115 course deliverables, we assess LLM performance using ChatGPT under a "minimal effort" protocol that simulates realistic student usage patterns. The investigation employs a rigorous testing methodology across multiple assessment formats, from auto-graded multiple choice questions to complex Python programming tasks and long-form analytical writing. Our analysis provides quantitative insights into AI's strengths and limitations in handling mathematical formulations, coding challenges, and theoretical concepts in control systems engineering. The LLM achieved a B-grade performance (82.24\%), approaching but not exceeding the class average (84.99\%), with strongest results in structured assignments and greatest limitations in open-ended projects. The findings inform discussions about course design adaptation in response to AI advancement, moving beyond simple prohibition towards thoughtful integration of these tools in engineering education. Additional materials including syllabus, examination papers, design projects, and example responses can be found at the project website: https://gradegpt.github.io.

cs.CY↗

arXiv2025

Energetic Resilience of Linear Driftless Systems

When a malfunction causes a control system to lose authority over a subset of its actuators, achieving a task may require spending additional energy in order to compensate for the effect of uncontrolled inputs. To understand this increase in energy, we introduce an energetic resilience metric that quantifies the maximal additional energy required to achieve finite-time regulation in linear driftless systems that suffer this malfunction. We first derive optimal control signals and minimum energies to achieve this task in both the nominal and malfunctioning systems. We then obtain a bound on the worst-case energy used by the malfunctioning system, and its exact expression in the special case of loss of authority over one actuator. Further considering this special case, we derive a bound on the metric for energetic resilience. A simulation example on a model of an underwater robot demonstrates that this bound is useful in quantifying the increased energy used by a system suffering such a malfunction.

math.OC↗

arXiv2025

Analysis of the Unscented Transform Controller for Systems with Bounded Nonlinearities

In this paper, we present an analysis of the Unscented Transform Controller (UTC), a technique to control nonlinear systems motivated as a dual to the Unscented Kalman Filter (UKF). We consider linear, discrete-time systems augmented by a bounded nonlinear function of the state. For such systems, we review 1-step and N-step versions of the UTC. Using a Lyapunov-based analysis, we prove that the states and inputs converge to a bounded ball around the origin, whose radius depends on the bound on the nonlinearity. Using examples of a fighter jet model and a quadcopter, we demonstrate that the UTC achieves satisfactory regulation and tracking performance on these nonlinear models.

eess.SY↗

arXiv2025

Target Prediction Under Deceptive Switching Strategies via Outlier-Robust Filtering of Partially Observed Incomplete Trajectories

Motivated by a study on deception and counter-deception, this paper addresses the problem of identifying an agent's target as it seeks to reach one of two targets in a given environment. In practice, an agent may initially follow a strategy to aim at one target but decide to switch to another midway. Such a strategy can be deceptive when the counterpart only has access to imperfect observations, which include heavily corrupted sensor noise and possible outliers, making it difficult to visually identify the agent's true intent. To counter deception and identify the true target, we utilize prior knowledge of the agent's dynamics and the imprecisely observed partial trajectory of the agent's states to dynamically update the estimation of the posterior probability of whether a deceptive switch has taken place. However, existing methods in the literature have not achieved effective deception identification within a reasonable computation time. We propose a set of outlier-robust change detection methods to track relevant change-related statistics efficiently, enabling the detection of deceptive strategies in hidden nonlinear dynamics with reasonable computational effort. The performance of the proposed framework is examined for Weapon-Target Assignment (WTA) detection under deceptive strategies using random simulations in the kinematics model with external forcing.

stat.AP↗

arXiv2025

Sum-of-Squares Data-driven Robustly Stabilizing and Contracting Controller Synthesis for Polynomial Nonlinear Systems

This work presents a computationally efficient approach to data-driven robust contracting controller synthesis for polynomial control-affine systems based on a sum-of-squares program. In particular, we consider the case in which a system alternates between periods of high-quality sensor data and low-quality sensor data. In the high-quality sensor data regime, we focus on robust system identification based on the data informativity framework. In low-quality sensor data regimes we employ a robustly contracting controller that is synthesized online by solving a sum-of-squares program based on data acquired in the high-quality regime, so as to limit state deviation until high-quality data is available. This approach is motivated by real-life control applications in which systems experience periodic data blackouts or occlusion, such as autonomous vehicles undergoing loss of GPS signal or solar glare in machine vision systems. We apply our approach to a planar unmanned aerial vehicle model subject to an unknown wind field, demonstrating its uses for verifiably tight control on trajectory deviation.

eess.SY↗

arXiv2025

Motion Planning and Control with Unknown Nonlinear Dynamics through Predicted Reachability

Autonomous motion planning under unknown nonlinear dynamics presents significant challenges. An agent needs to continuously explore the system dynamics to acquire its properties, such as reachability, in order to guide system navigation adaptively. In this paper, we propose a hybrid planning-control framework designed to compute a feasible trajectory toward a target. Our approach involves partitioning the state space and approximating the system by a piecewise affine (PWA) system with constrained control inputs. By abstracting the PWA system into a directed weighted graph, we incrementally update the existence of its edges via affine system identification and reach control theory, introducing a predictive reachability condition by exploiting prior information of the unknown dynamics. Heuristic weights are assigned to edges based on whether their existence is certain or remains indeterminate. Consequently, we propose a framework that adaptively collects and analyzes data during mission execution, continually updates the predictive graph, and synthesizes a controller online based on the graph search outcomes. We demonstrate the efficacy of our approach through simulation scenarios involving a mobile robot operating in unknown terrains, with its unknown dynamics abstracted as a single integrator model.

cs.RO↗

arXiv2025

TRACE: A Self-Improving Framework for Robot Behavior Forecasting with Vision-Language Models

Predicting the near-term behavior of a reactive agent is crucial in many robotic scenarios, yet remains challenging when observations of that agent are sparse or intermittent. Vision-Language Models (VLMs) offer a promising avenue by integrating textual domain knowledge with visual cues, but their one-shot predictions often miss important edge cases and unusual maneuvers. Our key insight is that iterative, counterfactual exploration--where a dedicated module probes each proposed behavior hypothesis, explicitly represented as a plausible trajectory, for overlooked possibilities--can significantly enhance VLM-based behavioral forecasting. We present TRACE (Tree-of-thought Reasoning And Counterfactual Exploration), an inference framework that couples tree-of-thought generation with domain-aware feedback to refine behavior hypotheses over multiple rounds. Concretely, a VLM first proposes candidate trajectories for the agent; a counterfactual critic then suggests edge-case variations consistent with partial observations, prompting the VLM to expand or adjust its hypotheses in the next iteration. This creates a self-improving cycle where the VLM progressively internalizes edge cases from previous rounds, systematically uncovering not only typical behaviors but also rare or borderline maneuvers, ultimately yielding more robust trajectory predictions from minimal sensor data. We validate TRACE on both ground-vehicle simulations and real-world marine autonomous surface vehicles. Experimental results show that our method consistently outperforms standard VLM-driven and purely model-based baselines, capturing a broader range of feasible agent behaviors despite sparse sensing. Evaluation videos and code are available at trace-robotics.github.io.

cs.RO↗

arXiv2025

Hallucination Detection in Foundation Models for Decision-Making: A Flexible Definition and Review of the State of the Art

Autonomous systems are soon to be ubiquitous, spanning manufacturing, agriculture, healthcare, entertainment, and other industries. Most of these systems are developed with modular sub-components for decision-making, planning, and control that may be hand-engineered or learning-based. While these approaches perform well under the situations they were specifically designed for, they can perform especially poorly in out-of-distribution scenarios that will undoubtedly arise at test-time. The rise of foundation models trained on multiple tasks with impressively large datasets has led researchers to believe that these models may provide "common sense" reasoning that existing planners are missing, bridging the gap between algorithm development and deployment. While researchers have shown promising results in deploying foundation models to decision-making tasks, these models are known to hallucinate and generate decisions that may sound reasonable, but are in fact poor. We argue there is a need to step back and simultaneously design systems that can quantify the certainty of a model's decision, and detect when it may be hallucinating. In this work, we discuss the current use cases of foundation models for decision-making tasks, provide a general definition for hallucinations with examples, discuss existing approaches to hallucination detection and mitigation with a focus on decision problems, present guidelines, and explore areas for further research in this exciting field.

cs.AI↗

arXiv2024

Guaranteed Reachability on Riemannian Manifolds for Unknown Nonlinear Systems

Determining the reachable set for a given nonlinear system is critically important for autonomous trajectory planning for reach-avoid applications and safety critical scenarios. Providing the reachable set is generally impossible when the dynamics are unknown, so we calculate underapproximations of such sets using local dynamics at a single point and bounds on the rate of change of the dynamics determined from known physical laws. Motivated by scenarios where an adverse event causes an abrupt change in the dynamics, we attempt to determine a provably reachable set of states without knowledge of the dynamics. This paper considers systems which are known to operate on a manifold. Underapproximations are calculated by utilizing the aforementioned knowledge to derive a guaranteed set of velocities on the tangent bundle of a complete Riemannian manifold that can be reached within a finite time horizon. We then interpret said set as a control system; the trajectories of this control system provide us with a guaranteed set of reachable states the unknown system can reach within a given time. The results are general enough to apply on systems that operate on any complete Riemannian manifold. To illustrate the practical implementation of our results, we apply our algorithm to a model of a pendulum operating on a sphere and a three-dimensional rotational system which lives on the abstract set of special orthogonal matrices.

eess.SY↗

arXiv2024

InfraLib: Enabling Reinforcement Learning and Decision-Making for Large-Scale Infrastructure Management

Efficient management of infrastructure systems is crucial for economic stability, sustainability, and public safety. However, infrastructure sustainment is challenging due to the vast scale of systems, stochastic deterioration of components, partial observability, and resource constraints. Decision-making strategies that rely solely on human judgment often result in suboptimal decisions over large scales and long horizons. While data-driven approaches like reinforcement learning offer promising solutions, their application has been limited by the lack of suitable simulation environments. We present InfraLib, an open-source modular and extensible framework that enables modeling and analyzing infrastructure management problems with resource constraints as sequential decision-making problems. The framework implements hierarchical, stochastic deterioration models, supports realistic partial observability, and handles practical constraints including cyclical budgets and component unavailability. InfraLib provides standardized environments for benchmarking decision-making approaches, along with tools for expert data collection and policy evaluation. Through case studies on both synthetic benchmarks and real-world road networks, we demonstrate InfraLib's ability to model diverse infrastructure management scenarios while maintaining computational efficiency at scale.

cs.AI↗

arXiv2024

TAB-Fields: A Maximum Entropy Framework for Mission-Aware Adversarial Planning

Autonomous agents operating in adversarial scenarios face a fundamental challenge: while they may know their adversaries' high-level objectives, such as reaching specific destinations within time constraints, the exact policies these adversaries will employ remain unknown. Traditional approaches address this challenge by treating the adversary's state as a partially observable element, leading to a formulation as a Partially Observable Markov Decision Process (POMDP). However, the induced belief-space dynamics in a POMDP require knowledge of the system's transition dynamics, which, in this case, depend on the adversary's unknown policy. Our key observation is that while an adversary's exact policy is unknown, their behavior is necessarily constrained by their mission objectives and the physical environment, allowing us to characterize the space of possible behaviors without assuming specific policies. In this paper, we develop Task-Aware Behavior Fields (TAB-Fields), a representation that captures adversary state distributions over time by computing the most unbiased probability distribution consistent with known constraints. We construct TAB-Fields by solving a constrained optimization problem that minimizes additional assumptions about adversary behavior beyond mission and environmental requirements. We integrate TAB-Fields with standard planning algorithms by introducing TAB-conditioned POMCP, an adaptation of Partially Observable Monte Carlo Planning. Through experiments in simulation with underwater robots and hardware implementations with ground robots, we demonstrate that our approach achieves superior performance compared to baselines that either assume specific adversary policies or neglect mission constraints altogether. Evaluation videos and code are available at https://tab-fields.github.io.

cs.RO↗

arXiv2024

Koopman-Based Learning of Infinitesimal Generators without Operator Logarithm

The Koopman operator has gained significant attention in recent years for its ability to verify evolutionary properties of continuous-time nonlinear systems by lifting state variables into an infinite-dimensional linear vector space. The challenge remains in providing estimations for transitional properties pertaining to the system's vector fields based on discrete-time observations. To retrieve such infinitesimal system transition information, leveraging the structure of Koopman operator learning, current literature focuses on developing techniques free of time derivatives through the use of the Koopman operator logarithm. However, the soundness of these methods has so far been demonstrated only for maintaining effectiveness within a restrictive function space, together with knowledge of the operator spectrum properties. To better adapt to the practical applications in learning and control of unknown systems, we propose a logarithm-free technique for learning the infinitesimal generator without disrupting the Koopman operator learning framework. This approach claims compatibility with other system verification tools using the same set of training data. We provide numerical examples to demonstrate its effectiveness in applications of system identification and stability prediction.

math.DS↗

arXiv2024

Autonomous Excavation of Challenging Terrain using Oscillatory Primitives and Adaptive Impedance Control

This paper addresses the challenge of autonomous excavation of challenging terrains, in particular those that are prone to jamming and inter-particle adhesion when tackled by a standard penetrate-drag-scoop motion pattern. Inspired by human excavation strategies, our approach incorporates oscillatory rotation elements -- including swivel, twist, and dive motions -- to break up compacted, tangled grains and reduce jamming. We also present an adaptive impedance control method, the Reactive Attractor Impedance Controller (RAIC), that adapts a motion trajectory to unexpected forces during loading in a manner that tracks a trajectory closely when loads are low, but avoids excessive loads when significant resistance is met. Our method is evaluated on four terrains using a robotic arm, demonstrating improved excavation performance across multiple metrics, including volume scooped, protective stop rate, and trajectory completion percentage.

cs.RO↗