arXiv subjects

Andreas Kugi

Publications and source records attributed to Andreas Kugi.

At least 19 recordsLinked to original sources

arXiv2026

FPGA-based LQG controller and hardware-in-the-loop simulator implementation for nanomechanical systems

We present an open-source framework for real-time Linear Quadratic Gaussian (LQG) control and hardware-in-the-loop (HIL) simulation on the affordable Red Pitaya STEMlab FPGA platform. The controller implements a discrete-time Kalman filter and Linear Quadratic Regulator (LQR) for systems with up to three coupled oscillatory degrees of freedom, targeting applications in levitated optomechanics, MEMS/NEMS, and related experimental platforms. Complementing the controller, the HIL simulator provides a~configurable second-order stochastic plant with nonlinear input and output mappings, enabling realistic closed-loop testing under real-time and fixed-point constraints. A MATLAB-based workflow automates model configuration, controller synthesis, numerical scaling, and FPGA deployment without requiring specialized hardware expertise. As an end-to-end demonstration, we present the stabilization of a levitated nanoparticle in a two-dimensional double-well potential, illustrating the complete workflow from model definition and simulation to real-time feedback control.

eess.SY↗

arXiv2026

Optimal State Preparation for Impulse Estimation in Gaussian Quantum Systems

We present an optimal control-based strategy to enhance the estimation of impulse-like disturbances in continuously monitored linear classical and quantum systems by exploiting non-equilibrium states. Using optimal estimation techniques for linear Gaussian systems to collect information from the temporal vicinity of the disturbance, we cast the minimization of disturbance estimation uncertainty as a nonlinear optimal control problem over time-dependent system parameters. The resulting method dynamically shapes the estimation covariances through parametric modulation, maximizing information gain at a known impulse time. This differs fundamentally from conventional squeezing protocols using periodic modulation that effectively degrade inference of impulse-like disturbances. Applied to nanomechanical resonators and levitated nanoparticles, optimal parametric driving reduces estimation variance by up to a factor of two relative to steady-state operation

quant-ph↗

arXiv2026

Conformal Cross-Modal Active Learning

Foundation models for vision have transformed visual recognition with powerful pretrained representations and strong zero-shot capabilities, yet their potential for data-efficient learning remains largely untapped. Active Learning (AL) aims to minimize annotation costs by strategically selecting the most informative samples for labeling, but existing methods largely overlook the rich multimodal knowledge embedded in modern vision-language models (VLMs). We introduce Conformal Cross-Modal Acquisition (CCMA), a novel AL framework that bridges vision and language modalities through a teacher-student architecture. CCMA employs a pretrained VLM as a teacher to provide semantically grounded uncertainty estimates, conformally calibrated to guide sample selection for a vision-only student model. By integrating multimodal conformal scoring with diversity-aware selection strategies, CCMA achieves superior data efficiency across multiple benchmarks. Our approach consistently outperforms state-of-the-art AL baselines, demonstrating clear advantages over methods relying solely on uncertainty or diversity metrics.

cs.CV↗

arXiv2026

SafeFlowMPC: Predictive and Safe Trajectory Planning for Robot Manipulators with Learning-based Policies

The emerging integration of robots into everyday life brings several major challenges. Compared to classical industrial applications, more flexibility is needed in combination with real-time reactivity. Learning-based methods can train powerful policies based on demonstrated trajectories, such that the robot generalizes a task to similar situations. However, these black-box models lack interpretability and rigorous safety guarantees. Optimization-based methods provide these guarantees but lack the required flexibility and generalization capabilities. This work proposes SafeFlowMPC, a combination of flow matching and online optimization to combine the strengths of learning and optimization. This method guarantees safety at all times and is designed to meet the demands of real-time execution by using a suboptimal model-predictive control formulation. SafeFlowMPC achieves strong performance in three real-world experiments on a KUKA 7-DoF manipulator, namely two grasping experiment and a dynamic human-robot object handover experiment. A video of the experiments is available at http://www.acin.tuwien.ac.at/42d6. The code is available at https://github.com/TU-Wien-ACIN-CDS/SafeFlowMPC.

cs.RO↗

arXiv2025

Identification and optimal control strategies for the transversal splitting of ultra--cold Bose gases

Splitting a Bose--Einstein condensate (BEC) is a key operation in fundamental physics experiments and emerging quantum technologies, where precise preparation of well--defined initial states requires fast yet coherent control of the condensate's nonlinear dynamics. This work formulates the BEC splitting process as an optimal feedforward control problem based on a physically interpretable, reduced--order model identified from limited experimental data. We introduce a systematic calibration strategy that combines optimal experiment selection and constrained nonlinear parameter estimation, enabling accurate system identification with minimal experimental overhead. Using this calibrated model, we compute energy--optimal trajectories via indirect optimal control to realize shortcuts to adiabaticity (STAs), achieving rapid transitions to the ground state of a double--well potential while suppressing excitations. Experiments confirm that the proposed control framework yields high--fidelity state transfers across multiple configurations, demonstrating its robustness and scalability for quantum control applications.

eess.SY↗

arXiv2025

Lang2Lift: A Language-Guided Autonomous Forklift System for Outdoor Industrial Pallet Handling

Automating pallet handling in outdoor logistics and construction environments remains challenging due to unstructured scenes, variable pallet configurations, and changing environmental conditions. In this paper, we present Lang2Lift, an end-to-end language-guided autonomous forklift system designed to support practical pallet pick-up operations in real-world outdoor settings. The system enables operators to specify target pallets using natural language instructions, allowing flexible selection among multiple pallets with different loads and spatial arrangements. Lang2Lift integrates foundation-model-based perception modules with motion planning and control in a closed-loop autonomy pipeline. Language-grounded visual perception is used to identify and segment target pallets, followed by 6D pose estimation and geometric refinement to generate manipulation-feasible insertion poses. The resulting pose estimates are directly coupled with the forklift planning and control modules to execute fully autonomous pallet pick-up maneuvers. We deploy and evaluate the proposed system on the ADAPT autonomous outdoor forklift platform across diverse real-world scenarios, including cluttered scenes, variable lighting, and different payload configurations. Tolerance-based pose evaluation further indicates accuracy sufficient for successful fork insertion. Timing and failure analyses highlight key deployment trade-offs and practical limitations, providing insights into integrating language-guided perception within industrial automation systems. Video demonstrations are available at https://eric-nguyen1402.github.io/lang2lift.github.io/

cs.RO↗

arXiv2025

Feedback stabilization of a nanoparticle at the intensity minimum of an optical double-well potential

In this work, we develop and analyze adaptive feedback control strategies to stabilize and confine a nanoparticle at the unstable intensity minimum of an optical double-well potential. The resulting stochastic optimal control problem for a noise-driven mechanical particle in a nonlinear optical potential must account for unavoidable experimental imperfections such as measurement nonlinearities and slow drifts of the optical setup. To address these issues, we simplify the model in the vicinity of the unstable equilibrium and employ indirect adaptive control techniques to dynamically follow changes in the potential landscape. Our approach leads to a simple and efficient Linear Quadratic Gaussian (LQG) controller that can be implemented on fast and cost-effective FPGAs, ensuring accessibility and reproducibility. We demonstrate that this strategy successfully tracks the intensity minimum and significantly reduces the nanoparticle's residual state variance, effectively lowering its center-of-mass temperature. While conventional optical traps rely on confining optical forces in the light field at the intensity maxima, trapping at intensity minima mitigates absorption heating, which is crucial for advanced quantum experiments. Since LQG control naturally extends into the quantum regime, our results provide a promising pathway for future experiments on quantum state preparation beyond the current absorption heating limitation, like matter-wave interference and tests of the quantum-gravity interface.

eess.SY↗

arXiv2025

Incremental Language Understanding for Online Motion Planning of Robot Manipulators

Human-robot interaction requires robots to process language incrementally, adapting their actions in real-time based on evolving speech input. Existing approaches to language-guided robot motion planning typically assume fully specified instructions, resulting in inefficient stop-and-replan behavior when corrections or clarifications occur. In this paper, we introduce a novel reasoning-based incremental parser which integrates an online motion planning algorithm within the cognitive architecture. Our approach enables continuous adaptation to dynamic linguistic input, allowing robots to update motion plans without restarting execution. The incremental parser maintains multiple candidate parses, leveraging reasoning mechanisms to resolve ambiguities and revise interpretations when needed. By combining symbolic reasoning with online motion planning, our system achieves greater flexibility in handling speech corrections and dynamically changing constraints. We evaluate our framework in real-world human-robot interaction scenarios, demonstrating online adaptions of goal poses, constraints, or task objectives. Our results highlight the advantages of integrating incremental language understanding with real-time motion planning for natural and fluid human-robot collaboration. The experiments are demonstrated in the accompanying video at www.acin.tuwien.ac.at/42d5.

cs.RO↗

arXiv2025

Fast Coherent Splitting of Bose-Einstein Condensates

Preparation of non-trivial quantum states without introducing unwanted excitations or decoherence remains a central challenge in utilizing ultracold atomic systems for quantum simulation. We employ optimal control methods to realize fast, coherent splitting of a one-dimensional Bose-Einstein condensate, achieving minimal classical excitations while preserving quantum correlations. Furthermore, we explore two-step protocols in which controlled classical motion is first induced and subsequently suppressed via tailored control sequences. Our experiments highlight the potential of optimal control for quantum state engineering and dynamical control in many-body quantum systems.

cond-mat.quant-gas↗

arXiv2025

ADAPT: An Autonomous Forklift for Construction Site Operation

Efficient material logistics play a critical role in controlling costs and schedules in the construction industry. However, manual material handling remains prone to inefficiencies, delays, and safety risks. Autonomous forklifts offer a promising solution to streamline on-site logistics, reducing reliance on human operators and mitigating labor shortages. This paper presents the development and evaluation of ADAPT (Autonomous Dynamic All-terrain Pallet Transporter), a fully autonomous off-road forklift designed for construction environments. Unlike structured warehouse settings, construction sites pose significant challenges, including dynamic obstacles, unstructured terrain, and varying weather conditions. To address these challenges, our system integrates AI-driven perception techniques with traditional approaches for decision making, planning, and control, enabling reliable operation in complex environments. We validate the system through extensive real-world testing, comparing its continuous performance against an experienced human operator across various weather conditions. Our findings demonstrate that autonomous outdoor forklifts can operate near human-level performance, offering a viable path toward safer and more efficient construction logistics.

cs.RO↗

arXiv2025

Federated Learning for Data-Driven Feedforward Control: A Case Study on Vehicle Lateral Dynamics

In many control systems, tracking accuracy can be enhanced by combining (data-driven) feedforward (FF) control with feedback (FB) control. However, designing effective data-driven FF controllers typically requires large amounts of high-quality data and a dedicated design-of-experiment process. In practice, relevant data are often distributed across multiple systems, which not only introduces technical challenges but also raises regulatory and privacy concerns regarding data transfer. To address these challenges, we propose a framework that integrates Federated Learning (FL) into the data-driven FF control design. Each client trains a data-driven, neural FF controller using local data and provides only model updates to the global aggregation process, avoiding the exchange of raw data. We demonstrate our method through simulation for a vehicle trajectory-tracking task. Therein, a neural FF controller is learned collaboratively using FL. Our results show that the FL-based neural FF controller matches the performance of the centralized neural FF controller while reducing communication overhead and increasing data privacy.

cs.LG↗

arXiv2025

BoundPlanner: A convex-set-based approach to bounded manipulator trajectory planning

Online trajectory planning enables robot manipulators to react quickly to changing environments or tasks. Many robot trajectory planners exist for known environments but are often too slow for online computations. Current methods in online trajectory planning do not find suitable trajectories in challenging scenarios that respect the limits of the robot and account for collisions. This work proposes a trajectory planning framework consisting of the novel Cartesian path planner based on convex sets, called BoundPlanner, and the online trajectory planner BoundMPC. BoundPlanner explores and maps the collision-free space using convex sets to compute a reference path with bounds. BoundMPC is extended in this work to handle convex sets for path deviations, which allows the robot to optimally follow the path within the bounds while accounting for the robot's kinematics. Collisions of the robot's kinematic chain are considered by a novel convex-set-based collision avoidance formulation independent on the number of obstacles. Simulations and experiments with a 7-DoF manipulator show the performance of the proposed planner compared to state-of-the-art methods. The source code is available at github.com/TU-Wien-ACIN-CDS/BoundPlanner and videos of the experiments can be found at www.acin.tuwien.ac.at/42d4.

cs.RO↗

arXiv2025

Towards Autonomous Wood-Log Grasping with a Forestry Crane: Simulator and Benchmarking

Forestry machines operated in forest production environments face challenges when performing manipulation tasks, especially regarding the complicated dynamics of underactuated crane systems and the heavy weight of logs to be grasped. This study investigates the feasibility of using reinforcement learning for forestry crane manipulators in grasping and lifting heavy wood logs autonomously. We first build a simulator using Mujoco physics engine to create realistic scenarios, including modeling a forestry crane with 8 degrees of freedom from CAD data and wood logs of different sizes. We further implement a velocity controller for autonomous log grasping with deep reinforcement learning using a curriculum strategy. Utilizing our new simulator, the proposed control strategy exhibits a success rate of 96% when grasping logs of different diameters and under random initial configurations of the forestry crane. In addition, reward functions and reinforcement learning baselines are implemented to provide an open-source benchmark for the community in large-scale manipulation tasks. A video with several demonstrations can be seen at https://www.acin.tuwien.ac.at/en/d18a/

cs.RO↗

arXiv2024

Optimal Sensing of Momentum Kicks with a Feedback-Controlled Nanomechanical Resonator

External disturbances exciting a mechanical resonator can be exploited to gain information on the environment. Many of these interactions manifest as momentum kicks, such as the recoil of residual gas, radioactive decay, or even hypothetical interactions with dark matter. These disturbances are often rare enough that they can be resolved as singular events rather than cumulated as force noise. While high-Q resonators with low masses are particularly sensitive to such momentum kicks, they will strongly excite the resonator, leading to nonlinear effects that deteriorate the sensing performance. Hence, this paper utilizes optimal estimation methods to extract individual momentum kicks from measured stochastic trajectories of a mechanical resonator kept in the linear regime through feedback control. The developed scheme is illustrated and tested experimentally using a pre-stressed SiN trampoline resonator. Apart from enhancing a wide range of sensing scenarios mentioned above, our results indicate the feasibility of novel single-molecule mass spectrometry approaches.

eess.SY↗

arXiv2024

Stabilizing nanoparticles in the intensity minimum: feedback levitation on an inverted potential

We demonstrate the stable trapping of a levitated nanoparticle on top of an inverted potential using a combination of optical readout and electrostatic control. The feedback levitation on an inverted potential (FLIP) method stabilizes the particle at an intensity minimum. By using a Kalman-filter-based linear-quadratic-Gaussian (LQG) control method, we confine a particle to within $\sigma_x = (9.0 \pm 0.5) nm$ of the potential maximum at an effective temperature of $(16 \pm 1) K$ in a room-temperature environment. Despite drifts in the absolute position of the potential maximum, we can keep the nanoparticle at the apex by estimating the drift from the particle dynamics using the Kalman filter. Our approach may enable new levitation-based sensing schemes with enhanced bandwidth. It also paves the way for optical levitation at zero intensity of an optical potential, which alleviates decoherence effects due to material-dependent absorption and is hence relevant for macroscopic quantum experiments.

physics.optics↗

arXiv2024

Combining Federated Learning and Control: A Survey

This survey provides an overview of combining Federated Learning (FL) and control to enhance adaptability, scalability, generalization, and privacy in (nonlinear) control applications. Traditional control methods rely on controller design models, but real-world scenarios often require online model retuning or learning. FL offers a distributed approach to model training, enabling collaborative learning across distributed devices while preserving data privacy. By keeping data localized, FL mitigates concerns regarding privacy and security while reducing network bandwidth requirements for communication. This survey summarizes the state-of-the-art concepts and ideas of combining FL and control. The methodical benefits are further discussed, culminating in a detailed overview of expected applications, from dynamical system modeling over controller design, focusing on adaptive control, to knowledge transfer in multi-agent decision-making systems.

cs.LG↗

arXiv2024

Language-Driven Closed-Loop Grasping with Model-Predictive Trajectory Replanning

Combining a vision module inside a closed-loop control system for a \emph{seamless movement} of a robot in a manipulation task is challenging due to the inconsistent update rates between utilized modules. This task is even more difficult in a dynamic environment, e.g., objects are moving. This paper presents a \emph{modular} zero-shot framework for language-driven manipulation of (dynamic) objects through a closed-loop control system with real-time trajectory replanning and an online 6D object pose localization. We segment an object within $\SI{0.5}{\second}$ by leveraging a vision language model via language commands. Then, guided by natural language commands, a closed-loop system, including a unified pose estimation and tracking and online trajectory planning, is utilized to continuously track this object and compute the optimal trajectory in real-time. Our proposed zero-shot framework provides a smooth trajectory that avoids jerky movements and ensures the robot can grasp a non-stationary object. Experiment results exhibit the real-time capability of the proposed zero-shot modular framework for the trajectory optimization module to accurately and efficiently grasp moving objects, i.e., up to \SI{30}{\hertz} update rates for the online 6D pose localization module and \SI{10}{\hertz} update rates for the receding-horizon trajectory optimization. These advantages highlight the modular framework's potential applications in robotics and human-robot interaction; see the video in https://www.acin.tuwien.ac.at/en/6e64/.

cs.RO↗

arXiv2024

Model Predictive Trajectory Planning for Human-Robot Handovers

This work develops a novel trajectory planner for human-robot handovers. The handover requirements can naturally be handled by a path-following-based model predictive controller, where the path progress serves as a progress measure of the handover. Moreover, the deviations from the path are used to follow human motion by adapting the path deviation bounds with a handover location prediction. A Gaussian process regression model, which is trained on known handover trajectories, is employed for this prediction. Experiments with a collaborative 7-DoF robotic manipulator show the effectiveness and versatility of the proposed approach.

cs.RO↗