arXiv subjects

Aradhana Nayak

Publications and source records attributed to Aradhana Nayak.

10 recordsLinked to original sources

arXiv2026

GUI agent: Guided Exploration of User-Sensitive Screens

LLM agents are increasingly being used to automate tasks for users within an open GUI environment. They inevitably encounter screens containing user-sensitive information, for which takeover of task execution by the user is highly desirable or even necessary. State-of-the-art LLM-driven agents are usually fine-tuned to complete tasks regardless of the safety implications of their actions. This makes their real-world deployment difficult and adversely affects the reliability. Therefore, it is crucial to identify and categorize user-sensitive states and define user-sensitive queries. This dataset would be to engineers to recognize and request handover to the user in critical scenarios. This short paper develops an explorer agent that systematically explores the query space starting from one demonstrated task to identify queries that, if executed, would lead to user-sensitive states in a GUI environment.

cs.AI↗

arXiv2025

The Safety Challenge of World Models for Embodied AI Agents: A Review

The rapid progress in embodied artificial intelligence has highlighted the necessity for more advanced and integrated models that can perceive, interpret, and predict environmental dynamics. In this context, World Models (WMs) have been introduced to provide embodied agents with the abilities to anticipate future environmental states and fill in knowledge gaps, thereby enhancing agents' ability to plan and execute actions. However, when dealing with embodied agents it is fundamental to ensure that predictions are safe for both the agent and the environment. In this article, we conduct a comprehensive literature review of World Models in the domains of autonomous driving and robotics, with a specific focus on the safety implications of scene and control generation tasks. Our review is complemented by an empirical analysis, wherein we collect and examine predictions from state-of-the-art models, identify and categorize common faults (herein referred to as pathologies), and provide a quantitative evaluation of the results.

cs.AI↗

arXiv2022

A Solution to Adaptive Mobile Manipulator Throwing

Mobile manipulator throwing is a promising method to increase the flexibility and efficiency of dynamic manipulation in factories. Its major challenge is to efficiently plan a feasible throw under a wide set of task specifications. We show that the mobile manipulator throwing problem can be simplified to a planar problem, hence greatly reducing the computational costs. Using machine learning approaches, we build a model of the object's inverted flying dynamics and the robot's kinematic feasibility, which enables throwing motion generation within 1 ms for given query of target position. Thanks to the computational efficiency of our method, we show that the system is adaptive under disturbance, via replanning on the fly for alternative solutions, instead of sticking to the original throwing plan.

cs.RO↗

arXiv2022

Learning High Dimensional Demonstrations Using Laplacian Eigenmaps

This article proposes a novel methodology to learn a stable robot control law driven by dynamical systems. The methodology requires a single demonstration and can deduce a stable dynamics in arbitrary high dimensions. The method relies on the idea that there exists a latent space in which the nonlinear dynamics appears quasi linear. The original nonlinear dynamics is mapped into a stable linear DS, by leveraging on the properties of graph embeddings. We show that the eigendecomposition of the Graph Laplacian results in linear embeddings in two dimensions and quasi-linear in higher dimensions. The nonlinear terms vanish, exponentially as the number of datapoints increase, and for large density of points, the embedding appears linear. We show that this new embedding enables to model highly nonlinear dynamics in high dimension and overcomes alternative techniques in both precision of reconstruction and number of parameters required for the embedding. We demonstrate its applicability to control real robot tasked to perform complex free motion in space.

cs.RO↗

arXiv2021

Stabilizing a spherical pendulum on a quadrotor

In this article we design a backstepping control law based on geometric principles to swing up a spherical pendulum mounted on a moving quadrotor. The available degrees of freedom in the control vector also permit us to position the plane of the quadrotor parallel to the ground. The problem addressed here is, indeed, novel and has many practical applications which arise during the transport of a payload mounted on top of a quadrotor. The modeling and control law are coordinate-free and thus avoid singularity issues. The geometric treatment of the problem greatly simplifies both the modeling and control law for the system. The control action is verified and supported by numerical experiments for aggressive manoeuvres starting very close to the downward stable equilibrium position of the pendulum.

eess.SY↗

arXiv2020

Geometric Optimal Trajectory Tracking of Nonholonomic Mechanical Systems

We study the tracking of a trajectory for a nonholonomic system by recasting the problem as a constrained optimal control problem. The cost function is chosen to minimize the error in positions and velocities between the trajectory of a nonholonomic system and the desired reference trajectory, both evolving on the distribution which defines the nonholonomic constraints. The problem is studied from a geometric framework. Optimality conditions are determined by the Pontryagin Maximum Principle and also from a variational point of view, which allows the construction of geometric integrators. Examples and numerical simulations are shown to validate the results.

math.OC↗

arXiv2019

Optimal Trajectory Tracking of Nonholonomic Mechanical Systems: a geometric approach

We study the tracking of a trajectory for a nonholonomic system by recasting the problem as an optimal control problem. The cost function is chosen to minimize the error in positions and velocities between the trajectory of a nonholonomic system and the desired reference trajectory evolving on the distribution which defines the nonholonomic constraints. We prepose a geometric framework since it describes the class of nonlinear systems under study in a coordinate-free framework. Necessary conditions for the existence of extrema are determined by the Pontryagin Minimum Principle. A nonholonomic fully actuated particle is used as a benchmark example to show how the proposed method is applied.

eess.SY↗

arXiv2017

Almost-global tracking of the unactuated joint in a pendubot

Tracking the unactuated configuration variable in an underactuated system, in a global sense, has not received much attention. Here we present a scheme to do so for a pendubot - a two link robot actuated only at the first link. We propose a control law for almost-global asymptotic tracking (AGAT) of a smooth reference trajectory for the unactuated second joint of the pendubot. The control law achieves almost-global tracking for any smooth reference trajectory specified for the unactuated joint. Further, we generalize the proposed scheme to an n-link system with as many (or more) degrees of actuation than unactuation, and show that the result holds.

eess.SY↗

arXiv2017

Almost-globally stable tracking for on compact Riemannian manifolds

In this article, we propose a control law for almost-global asymptotic tracking (AGAT) of a smooth reference trajectory for a fully actuated simple mechanical system (SMS) evolving on a Riemannian manifold which can be embedded in a Euclidean space. The existing results on tracking for an SMS are either local, or almost-global, only in the case the manifold is a Lie group. In the latter case, the notion of a configuration error is naturally defined by the group operation and facilitates a global analysis. However, such a notion is not intrinsic to a Riemannian manifold. In this paper, we define a configuration error followed by error dynamics on a Riemannian manifold, and then prove AGAT. The results are demonstrated for a spherical pendulum which is an SMS on $S^2$ and for a particle moving on a Lissajous curve in $\mathbb{R}^3$.

eess.SY↗

arXiv2017

Almost-global tracking for a rigid body with internal rotors

Almost-global orientation trajectory tracking for a rigid body with external actuation has been well studied in the literature, and in the geometric setting as well. The tracking control law relies on the fact that a rigid body is a simple mechanical system (SMS) on the $3-$dimensional group of special orthogonal matrices. However, the problem of designing feedback control laws for tracking using internal actuation mechanisms, like rotors or control moment gyros, has received lesser attention from a geometric point of view. An internally actuated rigid body is not a simple mechanical system, and the phase-space here evolves on the level set of a momentum map. In this note, we propose a novel proportional integral derivative (PID) control law for a rigid body with $3$ internal rotors, that achieves tracking of feasible trajectories from almost all initial conditions.

eess.SY↗