Matthias Althoff
YOU?
Author Swipe
View article: SanDRA: Safe Large-Language-Model-Based Decision Making for Automated Vehicles Using Reachability Analysis
SanDRA: Safe Large-Language-Model-Based Decision Making for Automated Vehicles Using Reachability Analysis Open
Large language models have been widely applied to knowledge-driven decision-making for automated vehicles due to their strong generalization and reasoning capabilities. However, the safety of the resulting decisions cannot be ensured due t…
View article: Falsification-Driven Reinforcement Learning for Maritime Motion Planning
Falsification-Driven Reinforcement Learning for Maritime Motion Planning Open
Compliance with maritime traffic rules is essential for the safe operation of autonomous vessels, yet training reinforcement learning (RL) agents to adhere to them is challenging. The behavior of RL agents is shaped by the training scenari…
View article: A Design Co-Pilot for Task-Tailored Manipulators
A Design Co-Pilot for Task-Tailored Manipulators Open
Although robotic manipulators are used in an ever-growing range of applications, robot manufacturers typically follow a ``one-fits-all'' philosophy, employing identical manipulators in various settings. This often leads to suboptimal perfo…
View article: Safe Reinforcement Learning using Action Projection: Safeguard the Policy or the Environment?
Safe Reinforcement Learning using Action Projection: Safeguard the Policy or the Environment? Open
Projection-based safety filters, which modify unsafe actions by mapping them to the closest safe alternative, are widely used to enforce safety constraints in reinforcement learning (RL). Two integration strategies are commonly considered:…
View article: An image-based, unified feature extraction framework for experimental and synthetic data for data-driven modelling in stamping and bending
An image-based, unified feature extraction framework for experimental and synthetic data for data-driven modelling in stamping and bending Open
Stamping describes cutting and forming processes in a combined, usually multi-stage process. The order of the process sequence, material, its composition, spring-back behavior and any pre-deformations of the material, influence the result …
View article: Automatic Geometric Decomposition for Analytical Inverse Kinematics
Automatic Geometric Decomposition for Analytical Inverse Kinematics Open
Calculating the inverse kinematics (IK) is a fundamental challenge in robotics. Compared to numerical or learning-based approaches, analytical IK provides higher efficiency and accuracy. However, existing analytical approaches are difficul…
View article: Approximability of the Containment Problem for Zonotopes and Ellipsotopes
Approximability of the Containment Problem for Zonotopes and Ellipsotopes Open
View article: Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanations
Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanations Open
Despite significant advancements in post-hoc explainability techniques for neural networks, many current methods rely on heuristics and do not provide formally provable guarantees over the explanations provided. Recent work has shown that …
View article: Out of the Shadows: Exploring a Latent Space for Neural Network Verification
Out of the Shadows: Exploring a Latent Space for Neural Network Verification Open
Neural networks are ubiquitous. However, they are often sensitive to small input changes. Hence, to prevent unexpected behavior in safety-critical applications, their formal verification -- a notoriously hard problem -- is necessary. Many …
View article: Language Models That Walk the Talk: A Framework for Formal Fairness Certificates
Language Models That Walk the Talk: A Framework for Formal Fairness Certificates Open
As large language models become integral to high-stakes applications, ensuring their robustness and fairness is critical. Despite their success, large language models remain vulnerable to adversarial attacks, where small perturbations, suc…
View article: Comparative Analysis of T5 Model Performance for Indonesian Abstractive Text Summarization
Comparative Analysis of T5 Model Performance for Indonesian Abstractive Text Summarization Open
The rapid growth of digital content has created significant challenges in information processing, particularly in languages like Indonesian, where automatic summarization remains complex. This study evaluates the performance of different T…
View article: Preface for Feature Topic on Testing of Highly Automated Vehicles
Preface for Feature Topic on Testing of Highly Automated Vehicles Open
View article: Scenario Factory 2.0: Scenario-Based Testing of Automated Vehicles with CommonRoad
Scenario Factory 2.0: Scenario-Based Testing of Automated Vehicles with CommonRoad Open
View article: Formally Verifying Analog Neural Networks with Device Mismatch Variations
Formally Verifying Analog Neural Networks with Device Mismatch Variations Open
View article: Trajectory Planning with Signal Temporal Logic Costs using Deterministic Path Integral Optimization
Trajectory Planning with Signal Temporal Logic Costs using Deterministic Path Integral Optimization Open
Formulating the intended behavior of a dynamic system can be challenging. Signal temporal logic (STL) is frequently used for this purpose due to its suitability in formalizing comprehensible, modular, and versatile spatiotemporal specifica…
View article: Intelligent Sailing Model for Open Sea Navigation
Intelligent Sailing Model for Open Sea Navigation Open
Autonomous vessels potentially enhance safety and reliability of seaborne trade. To facilitate the development of autonomous vessels, high-fidelity simulations are required to model realistic interactions with other vessels. However, model…
View article: Leveraging Label Preprocessing for Effective End-to-End Indonesian Automatic Speech Recognition
Leveraging Label Preprocessing for Effective End-to-End Indonesian Automatic Speech Recognition Open
This research explores the potential of improving low-resource Automatic Speech Recognition (ASR) performance by leveraging label preprocessing techniques in conjunction with the wav2vec2-large Self-Supervised Learning (SSL) model. ASR tec…
View article: Backward Reachability Analysis of Perturbed Continuous-Time Linear Systems Using Set Propagation
Backward Reachability Analysis of Perturbed Continuous-Time Linear Systems Using Set Propagation Open
View article: Holistic Construction Automation With Modular Robots: From High-Level Task Specification to Execution
Holistic Construction Automation With Modular Robots: From High-Level Task Specification to Execution Open
In situ robotic automation in construction is challenging due to constantly changing environments, a shortage of robotic experts, and a lack of standardized frameworks bridging robotics and construction practices. This work proposes a holi…
View article: Underapproximative Methods for the Order Reduction of Zonotopes
Underapproximative Methods for the Order Reduction of Zonotopes Open
Zonotopes are a widely used set representation in set-based computations due to their compact representation size and their closure under many relevant set operations. However, certain set operations, such as the Minkowski sum, increase th…
View article: CommonPower: A Framework for Safe Data-Driven Smart Grid Control
CommonPower: A Framework for Safe Data-Driven Smart Grid Control Open
View article: Traffic-Rule-Compliant Trajectory Repair via Satisfiability Modulo Theories and Reachability Analysis
Traffic-Rule-Compliant Trajectory Repair via Satisfiability Modulo Theories and Reachability Analysis Open
Complying with traffic rules is challenging for automated vehicles, as numerous rules need to be considered simultaneously. If a planned trajectory violates traffic rules, it is common to replan a new trajectory from scratch. We instead pr…
View article: No More Traffic Tickets: A Tutorial to Ensure Traffic-Rule Compliance of Automated Vehicles
No More Traffic Tickets: A Tutorial to Ensure Traffic-Rule Compliance of Automated Vehicles Open
View article: Leveraging Analytic Gradients in Provably Safe Reinforcement Learning
Leveraging Analytic Gradients in Provably Safe Reinforcement Learning Open
The deployment of autonomous robots in safety-critical applications requires safety guarantees. Provably safe reinforcement learning is an active field of research that aims to provide such guarantees using safeguards. These safeguards sho…
View article: Holistic Construction Automation with Modular Robots: From High-Level Task Specification to Execution
Holistic Construction Automation with Modular Robots: From High-Level Task Specification to Execution Open
In situ robotic automation in construction is challenging due to constantly changing environments, a shortage of robotic experts, and a lack of standardized frameworks bridging robotics and construction practices. This work proposes a holi…
View article: Fair Cost Allocation in Energy Communities Under Forecast Uncertainty
Fair Cost Allocation in Energy Communities Under Forecast Uncertainty Open
Energy communities (ECs) are an increasingly studied path toward improving prosumer coordination. A central challenge of ECs is to allocate cost savings fairly to members. While many allocation mechanisms have been developed, existing lite…
View article: A General Safety Framework for Autonomous Manipulation in Human Environments
A General Safety Framework for Autonomous Manipulation in Human Environments Open
Autonomous robots are projected to significantly augment the manual workforce, especially in repetitive and hazardous tasks. For a successful deployment of such robots in human environments, it is crucial to guarantee human safety. State-o…
View article: Results of the 2023 CommonRoad Motion Planning Competition for Autonomous Vehicles
Results of the 2023 CommonRoad Motion Planning Competition for Autonomous Vehicles Open
In recent years, different approaches for motion planning of autonomous vehicles have been proposed that can handle complex traffic situations. However, these approaches are rarely compared on the same set of benchmarks. To address this is…
View article: The ARCH-COMP Friendly Verification Competition for Continuous and Hybrid Systems
The ARCH-COMP Friendly Verification Competition for Continuous and Hybrid Systems Open
View article: Stepping Out of the Shadows: Reinforcement Learning in Shadow Mode
Stepping Out of the Shadows: Reinforcement Learning in Shadow Mode Open
Reinforcement learning (RL) is not yet competitive for many cyber-physical systems, such as robotics, process automation, and power systems, as training on a system with physical components cannot be accelerated, and simulation models do n…