arXiv subjects

Thomas Seel

Publications and source records attributed to Thomas Seel.

At least 19 recordsLinked to original sources

arXiv2026

Time-Series Anomaly Detection for Mobile Robots in Automotive Active Safety Testing using an RNN-VAE

Mobile robots, like the ultra-flat overrunable (UFO) robot platform, used in automotive active safety tests, currently lack self-diagnostic capabilities necessary to detect present hardware defects. This circumstance can lead to more severe failures, causing expensive repairs and operational downtime. This work proposes, for the first time, a reconstructionbased time-series anomaly detection model for these mobile robots, considering defect classes such as unevenly worn full-rubber tires or damaged dampers. Unlike prior publications, the proposed approach leverages the vast quantities of unlabeled data generated during routine operation through a simple pre-training step. Furthermore, it optimizes the hyperparameters of the implemented gated recurrent unit-based variational autoencoder (GRU-VAE) and evaluates both a stateless, windowed training approach and one using truncated backpropagation through time (TBPTT). The model's generalization capabilities are demonstrated by successfully detecting six defect types, with four of them not present in the data used for hyperparameter optimization and threshold selection. This is validated using a test set collected from five system instances at various points over a period of several months, achieving an F1 score of 0.936, indicating strong practical viability.

eess.SY↗

arXiv2026

Inertial Human Motion Capture: From Biomechanics to Recent Sensor Fusion Methods and Back

Inertial measurement units (IMUs) are a promising means to capture human motion, yet obtaining meaningful biomechanical quantities from IMU measurements remains non-trivial. This tutorial-style review focuses on kinematics and introduces four key aspects (inertial human motion capture objective, environmental conditions, subject & attributes, and motion characteristics) to determine how to translate biomechanical problems into adequate formulations for the fusion of inertial sensor measurements. We identify three fundamental challenges for kinematics estimation from IMUs: IMUs do not provide direct information about the joint angle, IMUs do not measure their own orientation, and real-world environments and dynamics compromise sensor reliability. Though there exist widely-used methods to overcome these challenges, they suffer from severe limitations in real-life applications, e.g., the need for sensor-to-segment calibration, and the fact that magnetic field disturbances degrade joint angle accuracy. The full potential for many use-cases hence remains untapped in terms of accuracy and reliability. We share insights into recently proposed methods, e.g. exploiting the human body's kinematic chain constraints, having the potential to overcome these limitations. We also present guiding questions related to the four key aspects and illustrate their use for navigating the methodological landscape for the use-case of lower-extremity joint angle estimation, for which we share open-access code and compare the traditional workflow with three alternatives. Our aim is to bridge the gap between the sensor fusion community developing methods for human motion capture and the biomechanics community in need of accurate, easy-to-use, and reliable methods to study human motion outside of the laboratory.

eess.SP↗

arXiv2026

Neural Network-Based Virtual Wheel-Speed Sensor for Enhanced Low-Velocity State Estimation

Accurate wheel speed information is crucial for vehicle control and state estimation. Conventional sensors suffer from quantization and latency, especially at low velocities, while motor-speed signals in electric vehicles are distorted by drivetrain torsion. This work presents a neural-network-based virtual wheel-speed sensor that fuses wheel-speed and motor-speed signals to reduce errors from both sources. Validated on real-world Volkswagen ID.7 data, the real-time capable model achieves an error reduction of up to 85% compared to the production sensor and 47% compared to an optimized zero-phase filter, providing a smooth signal for driver-assistance functions. The results demonstrate robust generalization across diverse real-world maneuvers within the vehicle platform.

eess.SY↗

arXiv2026

Structure-Preserving Gaussian Processes Via Discrete Euler-Lagrange Equations

In this paper, we propose Lagrangian Gaussian Processes (LGPs) for probabilistic and data-efficient learning of dynamics via discrete forced Euler-Lagrange equations. Importantly, the geometric structure of the Lagrange-d'Alembert principle, which governs the motion of dynamical systems, is preserved by construction in the absence of external forces. This allows learning physically consistent models that overcome erroneous drift in the system's energy, thereby providing stable long-term predictions. At the core of our approach lie linear operators for Gaussian process conditioning, constructed from discrete forced Euler-Lagrange equations and variational discretization schemes. Thereby and unlike prior work, the method enables learning dynamics from discrete position snapshots, i.e., without access to a system's velocities or momenta. This is particularly relevant for a large class of practical scenarios where only position measurements are available, for instance, in motion capture or visual servoing applications. We demonstrate the data-efficiency and generalization capabilities of the LGPs in various synthetic and real-world case studies, including a real-world soft robot with hysteresis. The experimental results underscore that the LGPs learn physically consistent dynamics with uncertainty quantification solely from sparse positional data and enable stable long-term predictions.

cs.LG↗

arXiv2026

Active Stereo-Camera Outperforms Multi-Sensor Setup in ACT Imitation Learning for Humanoid Manipulation

The complexity of teaching humanoid robots new tasks is one of the major reasons hindering their widespread adoption in the industry. While Imitation Learning (IL), particularly Action Chunking with Transformers (ACT), enables rapid task acquisition, there is no consensus yet on the optimal sensory hardware required for manipulation tasks. This paper benchmarks 14 sensor combinations on the Unitree G1 humanoid robot equipped with three-finger hands for two manipulation tasks. We explicitly evaluate the integration of tactile and proprioceptive modalities alongside active vision. Our analysis demonstrates that strategic sensor selection can outperform complex configurations in data-limited regimes while reducing computational overhead. We develop an open-source Unified Ablation Framework that utilizes sensor masking on a comprehensive master dataset. Results indicate that additional modalities often degrade performance for IL with limited data. A minimal active stereo-camera setup outperformed complex multi-sensor configurations, achieving 87.5% success in a spatial generalization task and 94.4% in a structured manipulation task. Conversely, adding pressure sensors to this setup reduced success to 67.3% in the latter task due to a low signal-to-noise ratio. We conclude that in data-limited regimes, active vision offers a superior trade-off between robustness and complexity. While tactile modalities may require larger datasets to be effective, our findings validate that strategic sensor selection is critical for designing an efficient learning process.

cs.RO↗

arXiv2026

Simultaneous State Estimation and Online Model Learning in a Soft Robotic System

Operating complex real-world systems, such as soft robots, can benefit from precise predictive control schemes that require accurate state and model knowledge. This knowledge is typically not available in practical settings and must be inferred from noisy measurements. In particular, it is challenging to simultaneously estimate unknown states and learn a model online from sequentially arriving measurements. In this paper, we show how a recently proposed gray-box system identification tool enables the estimation of a soft robot's current pose while at the same time learning a bending stiffness model. For estimation and learning, we only need a nominal constant-curvature robot model and measurements of the robot's base reactions (e.g., base forces). The estimation scheme -- relying on a marginalized particle filter -- allows us to conveniently interface nominal constant-curvature equations with a Gaussian Process (GP) bending stiffness model to be learned. This, in contrast to estimation via a random walk over stiffness values, enables prediction of bending stiffness and improves overall model quality. We demonstrate, using a real-world soft robot, that the method learns a bending-stiffness model online while accurately estimating the robot's pose. Notably, reduced error in multi-step forward predictions indicates that the learned bending-stiffness GP improves overall model quality.

eess.SY↗

arXiv2026

A Hybrid Autoencoder for Robust Heightmap Generation from Fused Lidar and Depth Data for Humanoid Robot Locomotion

Reliable terrain perception is a critical prerequisite for the deployment of humanoid robots in unstructured, human-centric environments. While traditional systems often rely on manually engineered, single-sensor pipelines, this paper presents a learning-based framework that uses an intermediate, robot-centric heightmap representation. A hybrid Encoder-Decoder Structure (EDS) is introduced, utilizing a Convolutional Neural Network (CNN) for spatial feature extraction fused with a Gated Recurrent Unit (GRU) core for temporal consistency. The architecture integrates multimodal data from an Intel RealSense depth camera, a LIVOX MID-360 LiDAR processed via efficient spherical projection, and an onboard IMU. Quantitative results demonstrate that multimodal fusion improves reconstruction accuracy by 7.2% over depth-only and 9.9% over LiDAR-only configurations. Furthermore, the integration of a 3.2 s temporal context reduces mapping drift.

cs.RO↗

arXiv2025

Learning Dynamics from Input-Output Data with Hamiltonian Gaussian Processes

Embedding non-restrictive prior knowledge, such as energy conservation laws, into learning methods is a key motive to construct physically consistent dynamics models from limited data, relevant for, e.g., model-based control. Recent work incorporates Hamiltonian dynamics into Gaussian Processes (GPs) to obtain uncertainty-quantifying, energy-consistent models, but these methods rely on -- rarely available -- velocity or momentum data. In this paper, we study dynamics learning using Hamiltonian GPs and focus on learning solely from input-output data, without relying on velocity or momentum measurements. Adopting a non-conservative formulation, energy exchange with the environment, e.g., through external forces or dissipation, can be captured. We provide a fully Bayesian scheme for estimating probability densities of unknown hidden states, GP hyperparameters, as well as structural hyperparameters, such as damping coefficients. The proposed method is evaluated in a nonlinear simulation case study and compared to a state-of-the-art approach that relies on momentum measurements.

cs.LG↗

arXiv2025

Dual Iterative Learning Control for Multiple-Input Multiple-Output Dynamics with Validation in Robotic Systems

Solving motion tasks autonomously and accurately is a core ability for intelligent real-world systems. To achieve genuine autonomy across multiple systems and tasks, key challenges include coping with unknown dynamics and overcoming the need for manual parameter tuning, which is especially crucial in complex Multiple-Input Multiple-Output (MIMO) systems. This paper presents MIMO Dual Iterative Learning Control (DILC), a novel data-driven iterative learning scheme for simultaneous tracking control and model learning, without requiring any prior system knowledge or manual parameter tuning. The method is designed for repetitive MIMO systems and integrates seamlessly with established iterative learning control methods. We provide monotonic convergence conditions for both reference tracking error and model error in linear time-invariant systems. The DILC scheme -- rapidly and autonomously -- solves various motion tasks in high-fidelity simulations of an industrial robot and in multiple nonlinear real-world MIMO systems, without requiring model knowledge or manually tuning the algorithm. In our experiments, many reference tracking tasks are solved within 10-20 trials, and even complex motions are learned in less than 100 iterations. We believe that, because of its rapid and autonomous learning capabilities, DILC has the potential to serve as an efficient building block within complex learning frameworks for intelligent real-world systems.

eess.SY↗

arXiv2025

Bayesian Inference and Learning in Nonlinear Dynamical Systems: A Framework for Incorporating Explicit and Implicit Prior Knowledge

Accuracy and generalization capabilities are key objectives when learning dynamical system models. To obtain such models from limited data, current works exploit prior knowledge and assumptions about the system. However, the fusion of diverse prior knowledge, e. g. partially known system equations and smoothness assumptions about unknown model parts, with information contained in the data remains a challenging problem, especially in input-output settings with latent system state. In particular, learning functions that are nested inside known system equations can be a laborious and error-prone expert task. This paper considers inference of latent states and learning of unknown model parts for fusion of data information with different sources of prior knowledge. The main contribution is a general-purpose system identification tool that, for the first time, provides a consistent solution for both, online and offline Bayesian inference and learning while allowing to incorporate explicit and implicit prior system knowledge. We propose a novel interface for combining known dynamics functions with a learning-based approximation of unknown system parts. Based on the proposed model structure, closed-form densities for efficient parameter marginalization are derived. No user-tailored coordinate transformations or model inversions are needed, making the presented framework a general-purpose tool for inference and learning. The broad applicability of the devised framework is illustrated in three distinct case studies, including an experimental data set.

stat.ML↗

arXiv2025

Adaptive Model-Predictive Control of a Soft Continuum Robot Using a Physics-Informed Neural Network Based on Cosserat Rod Theory

Dynamic control of soft continuum robots (SCRs) holds great potential for expanding their applications, but remains a challenging problem due to the high computational demands of accurate dynamic models. While data-driven approaches like Koopman-operator-based methods have been proposed, they typically lack adaptability and cannot reconstruct the full robot shape, limiting their applicability. This work introduces a real-time-capable nonlinear model-predictive control (MPC) framework for SCRs based on a domain-decoupled physics-informed neural network (DD-PINN) with adaptable bending stiffness. The DD-PINN serves as a surrogate for the dynamic Cosserat rod model with a speed-up factor of up to 44,000. It is also used within an unscented Kalman filter for estimating the model states and bending compliance from end-effector position measurements. We implement a nonlinear evolutionary MPC running at 70 Hz on the GPU. In simulation, it demonstrates accurate tracking of dynamic trajectories and setpoint control with end-effector position errors below 3 mm (2.3\% of the actuator's length). In real-world experiments, the controller achieves similar accuracy and accelerations up to 3.55 m/s2.

cs.RO↗

arXiv2025

Fast Contact Detection via Fusion of Joint and Inertial Sensors for Parallel Robots in Human-Robot Collaboration

Fast contact detection is crucial for safe human-robot collaboration. Observers based on proprioceptive information can be used for contact detection but have first-order error dynamics, which results in delays. Sensor fusion based on inertial measurement units (IMUs) consisting of accelerometers and gyroscopes is advantageous for reducing delays. The acceleration estimation enables the direct calculation of external forces. For serial robots, the installation of multiple accelerometers and gyroscopes is required for dynamics modeling since the joint coordinates are the minimal coordinates. Alternatively, parallel robots (PRs) offer the potential to use only one IMU on the end-effector platform, which already presents the minimal coordinates of the PR. This work introduces a sensor-fusion method for contact detection using encoders and only one low-cost, consumer-grade IMU for a PR. The end-effector accelerations are estimated by an extended Kalman filter and incorporated into the dynamics to calculate external forces. In real-world experiments with a planar PR, we demonstrate that this approach reduces the detection duration by up to 50% compared to a momentum observer and enables the collision and clamping detection within 3-39ms.

cs.RO↗

arXiv2025

Generalizable and Fast Surrogates: Model Predictive Control of Articulated Soft Robots using Physics-Informed Neural Networks

Soft robots can revolutionize several applications with high demands on dexterity and safety. When operating these systems, real-time estimation and control require fast and accurate models. However, prediction with first-principles (FP) models is slow, and learned black-box models have poor generalizability. Physics-informed machine learning offers excellent advantages here, but it is currently limited to simple, often simulated systems without considering changes after training. We propose physics-informed neural networks (PINNs) for articulated soft robots (ASRs) with a focus on data efficiency. The amount of expensive real-world training data is reduced to a minimum -- one dataset in one system domain. Two hours of data in different domains are used for a comparison against two gold-standard approaches: In contrast to a recurrent neural network, the PINN provides a high generalizability. The prediction speed of an accurate FP model is exceeded with the PINN by up to a factor of 467 at slightly reduced accuracy. This enables nonlinear model predictive control (MPC) of a pneumatic ASR. Accurate position tracking with the MPC running at 47 Hz is achieved in six dynamic experiments.

cs.RO↗

arXiv2025

SafePR: Unified Approach for Safe Parallel Robots by Contact Detection and Reaction with Redundancy Resolution

Fast and safe motion is crucial for the successful deployment of physically interactive robots. Parallel robots (PRs) offer the potential for higher speeds while maintaining the same energy limits due to their low moving masses. However, they require methods for contact detection and reaction while avoiding singularities and self-collisions. We address this issue and present SafePR - a unified approach for the detection and localization, including the distinction between collision and clamping to perform a reaction that is safe for humans and feasible for PRs. Our approach uses information from the encoders and motor currents to estimate forces via a generalized-momentum observer. Neural networks and particle filters classify and localize the contacts. We introduce reactions with redundancy resolution to avoid self-collisions and type-II singularities. Our approach detected and terminated 72 real-world collision and clamping contacts with end-effector speeds of up to 1.5 m/s, each within 25-275 ms. The forces were below the thresholds from ISO/TS 15066. By using built-in sensors, SafePR enables safe interaction with already assembled PRs without the need for new hardware components.

cs.RO↗

arXiv2024

Learning-based Nonlinear Model Predictive Control of Articulated Soft Robots using Recurrent Neural Networks

Soft robots pose difficulties in terms of control, requiring novel strategies to effectively manipulate their compliant structures. Model-based approaches face challenges due to the high dimensionality and nonlinearities such as hysteresis effects. In contrast, learning-based approaches provide nonlinear models of different soft robots based only on measured data. In this paper, recurrent neural networks (RNNs) predict the behavior of an articulated soft robot (ASR) with five degrees of freedom (DoF). RNNs based on gated recurrent units (GRUs) are compared to the more commonly used long short-term memory (LSTM) networks and show better accuracy. The recurrence enables the capture of hysteresis effects that are inherent in soft robots due to viscoelasticity or friction but cannot be captured by simple feedforward networks. The data-driven model is used within a nonlinear model predictive control (NMPC), whereby the correct handling of the RNN's hidden states is focused. A training approach is presented that allows measured values to be utilized in each control cycle. This enables accurate predictions of short horizons based on sensor data, which is crucial for closed-loop NMPC. The proposed learning-based NMPC enables trajectory tracking with an average error of 1.2deg in experiments with the pneumatic five-DoF ASR.

cs.RO↗

arXiv2024

Predictive Energy Management for Recuperation Axles in Refrigerated Trailers

Refrigerated truck trailers are currently mainly operated with environmentally harmful diesel units; an alternative is to operate the refrigeration unit with electrical energy. However, this requires a battery, the size of which can be reduced by using a recuperation axle, which recovers energy during braking. Current systems work purely reactively and often in so-called towing mode, in which a generator torque is provided without a braking request from the driver in order to secure the energy supply. However, this drag leads to additional consumption in the truck. This work quantifies the potential of predictive energy management that uses route and environmental data to minimize CO2 emissions. This was done using simulation data obtained with the help of VECTO. It was shown that there is still considerable potential for savings, so this paper provides an important basis for the later development of predictive energy management and, thus, for the electrification of refrigerated truck transports.

eess.SY↗

arXiv2024

Efficient Online Inference and Learning in Partially Known Nonlinear State-Space Models by Learning Expressive Degrees of Freedom Offline

Intelligent real-world systems critically depend on expressive information about their system state and changing operation conditions, e.g., due to variation in temperature, location, wear, or aging. To provide this information, online inference and learning attempts to perform state estimation and (partial) system identification simultaneously. Current works combine tailored estimation schemes with flexible learning-based models but suffer from convergence problems and computational complexity due to many degrees of freedom in the inference problem (i.e., parameters to determine). To resolve these issues, we propose a procedure for data-driven offline conditioning of a highly flexible Gaussian Process (GP) formulation such that online learning is restricted to a subspace, spanned by expressive basis functions. Due to the simplicity of the transformed problem, a standard particle filter can be employed for Bayesian inference. In contrast to most existing works, the proposed method enables online learning of target functions that are nested nonlinearly inside a first-principles model. Moreover, we provide a theoretical quantification of the error, introduced by restricting learning to a subspace. A Monte-Carlo simulation study with a nonlinear battery model shows that the proposed approach enables rapid convergence with significantly fewer particles compared to a baseline and a state-of-the-art method.

eess.SY↗

arXiv2024

Autonomous Iterative Motion Learning (AI-MOLE) of a SCARA Robot for Automated Myocardial Injection

Stem cell therapy is a promising approach to treat heart insufficiency and benefits from automated myocardial injection which requires highly precise motion of a robotic manipulator that is equipped with a syringe. This work investigates whether sufficiently precise motion can be achieved by combining a SCARA robot and learning control methods. For this purpose, the method Autonomous Iterative Motion Learning (AI-MOLE) is extended to be applicable to multi-input/multi-output systems. The proposed learning method solves reference tracking tasks in systems with unknown, nonlinear, multi-input/multi-output dynamics by iteratively updating an input trajectory in a plug-and-play fashion and without requiring manual parameter tuning. The proposed learning method is validated in a preliminary simulation study of a simplified SCARA robot that has to perform three desired motions. The results demonstrate that the proposed learning method achieves highly precise reference tracking without requiring any a priori model information or manual parameter tuning in as little as 15 trials per motion. The results further indicate that the combination of a SCARA robot and learning method achieves sufficiently precise motion to potentially enable automatic myocardial injection if similar results can be obtained in a real-world setting.

cs.RO↗