Disengagement Analysis and Field Tests of a Prototypical Open-Source Level 4 Autonomous Driving System

Proprietary Autonomous Driving Systems are typically evaluated through disengagements, unplanned manual interventions to alter vehicle behavior, as annually reported by the California Department of Motor Vehicles. However, the real-world capabilities of prototypical open-source Level 4 vehicles over substantial distances remain largely unexplored. This study evaluates a research vehicle running an Autoware-based software stack across 236 km of mixed traffic. By classifying 30 disengagements across 26 rides with a novel five-level criticality framework, we observed a spatial disengagement rate of 0.127 1/km. Interventions predominantly occurred at lower speeds near static objects and traffic lights. Perception and Planning failures accounted for 40% and 26.7% of disengagements, respectively, largely due to object-tracking losses and operational deadlocks caused by parked vehicles. Frequent, unnecessary interventions highlighted a lack of trust on the part of the safety driver. These results show that while open-source software enables extensive operations, disengagement analysis is vital for uncovering robustness issues missed by standard metrics.

StyleVLA: Driving Style-Aware Vision Language Action Model for Autonomous Driving

Vision Language Models (VLMs) bridge visual perception and linguistic reasoning. In Autonomous Driving (AD), this synergy has enabled Vision Language Action (VLA) models, which translate high-level multimodal understanding into driving behaviors, typically represented as future trajectories. However, existing VLA models mainly generate generic collision-free trajectories. Beyond collision avoidance, adapting to diverse driving styles (e.g., sporty, comfortable) is essential for personalized driving. Moreover, many methods treat trajectory generation as naive token prediction, which can produce kinematically infeasible actions. To address these limitations, we present StyleVLA, a physics-informed VLA framework for generating diverse and physically plausible driving behaviors. We introduce a hybrid loss that combines a kinematic consistency constraint with a continuous regression head to improve trajectory feasibility. To train StyleVLA, built on Qwen3-VL-4B, we construct a large-scale instruction dataset with over 1.2k scenarios, 76k Bird's Eye View (BEV) samples, and 42k First Person View (FPV) samples, with ground-truth trajectories for five driving styles and natural-language instructions. Experiments show that our 4B-parameter StyleVLA significantly outperforms proprietary models (e.g., Gemini-3-Pro) and state-of-the-art VLA models. Using a composite driving score measuring success rate, physical feasibility, and style adherence, StyleVLA achieves 0.55 on BEV and 0.51 on FPV, versus 0.32 and 0.35 for Gemini-3-Pro. These results show that a specialized, physics-informed, lightweight model can surpass closed-source models on domain-specific tasks.

Head-to-Head autonomous racing at the limits of handling in the A2RL challenge

Autonomous racing presents a complex challenge involving multi-agent interactions between vehicles operating at the limit of performance and dynamics. As such, it provides a valuable research and testing environment for advancing autonomous driving technology and improving road safety. This article presents the algorithms and deployment strategies developed by the TUM Autonomous Motorsport team for the inaugural Abu Dhabi Autonomous Racing League (A2RL). We showcase how our software emulates human driving behavior, pushing the limits of vehicle handling and multi-vehicle interactions to win the A2RL. Finally, we highlight the key enablers of our success and share our most significant learnings.

Modular Autonomy with Conversational Interaction: An LLM-driven Framework for Decision Making in Autonomous Driving

Recent advancements in Large Language Models (LLMs) offer new opportunities to create natural language interfaces for Autonomous Driving Systems (ADSs), moving beyond rigid inputs. This paper addresses the challenge of mapping the complexity of human language to the structured action space of modular ADS software. We propose a framework that integrates an LLM-based interaction layer with Autoware, a widely used open-source software. This system enables passengers to issue high-level commands, from querying status information to modifying driving behavior. Our methodology is grounded in three key components: a taxonomization of interaction categories, an application-centric Domain Specific Language (DSL) for command translation, and a safety-preserving validation layer. A two-stage LLM architecture ensures high transparency by providing feedback based on the definitive execution status. Evaluation confirms the system's timing efficiency and translation robustness. Simulation successfully validated command execution across all five interaction categories. This work provides a foundation for extensible, DSL-assisted interaction in modular and safety-conscious autonomy stacks.

A Kalman Filter-Based Disturbance Observer for Steer-by-Wire Systems

Steer-by-Wire systems replace mechanical linkages, which provide benefits like weight reduction, design flexibility, and compatibility with autonomous driving. However, they are susceptible to high-frequency disturbances from unintentional driver torque, known as driver impedance, which can degrade steering performance. Existing approaches either rely on direct torque sensors, which are costly and impractical, or lack the temporal resolution to capture rapid, high-frequency driver-induced disturbances. We address this limitation by designing a Kalman filter-based disturbance observer that estimates high-frequency driver torque using only motor state measurements. We model the drivers passive torque as an extended state using a PT1-lag approximation and integrate it into both linear and nonlinear Steer-by-Wire system models. In this paper, we present the design, implementation and simulation of this disturbance observer with an evaluation of different Kalman filter variants. Our findings indicate that the proposed disturbance observer accurately reconstructs driver-induced disturbances with only minimal delay 14ms. We show that a nonlinear extended Kalman Filter outperforms its linear counterpart in handling frictional nonlinearities, improving estimation during transitions from static to dynamic friction. Given the study's methodology, it was unavoidable to rely on simulation-based validation rather than real-world experimentation. Further studies are needed to investigate the robustness of the observers under real-world driving conditions.

Unsupervised Learning for Detection of Rare Driving Scenarios

The detection of rare and hazardous driving scenarios is a critical challenge for ensuring the safety and reliability of autonomous systems. This research explores an unsupervised learning framework for detecting rare and extreme driving scenarios using naturalistic driving data (NDD). We leverage the recently proposed Deep Isolation Forest (DIF), an anomaly detection algorithm that combines neural network-based feature representations with Isolation Forests (IFs), to identify non-linear and complex anomalies. Data from perception modules, capturing vehicle dynamics and environmental conditions, is preprocessed into structured statistical features extracted from sliding windows. The framework incorporates t-distributed stochastic neighbor embedding (t-SNE) for dimensionality reduction and visualization, enabling better interpretability of detected anomalies. Evaluation is conducted using a proxy ground truth, combining quantitative metrics with qualitative video frame inspection. Our results demonstrate that the proposed approach effectively identifies rare and hazardous driving scenarios, providing a scalable solution for anomaly detection in autonomous driving systems. Given the study's methodology, it was unavoidable to depend on proxy ground truth and manually defined feature combinations, which do not encompass the full range of real-world driving anomalies or their nuanced contextual dependencies.

RoGER-SLAM: A Robust Gaussian Splatting SLAM System for Noisy and Low-light Environment Resilience

The reliability of Simultaneous Localization and Mapping (SLAM) is severely constrained in environments where visual inputs suffer from noise and low illumination. Although recent 3D Gaussian Splatting (3DGS) based SLAM frameworks achieve high-fidelity mapping under clean conditions, they remain vulnerable to compounded degradations that degrade mapping and tracking performance. A key observation underlying our work is that the original 3DGS rendering pipeline inherently behaves as an implicit low-pass filter, attenuating high-frequency noise but also risking over-smoothing. Building on this insight, we propose RoGER-SLAM, a robust 3DGS SLAM system tailored for noise and low-light resilience. The framework integrates three innovations: a Structure-Preserving Robust Fusion (SP-RoFusion) mechanism that couples rendered appearance, depth, and edge cues; an adaptive tracking objective with residual balancing regularization; and a Contrastive Language-Image Pretraining (CLIP)-based enhancement module, selectively activated under compounded degradations to restore semantic and structural fidelity. Comprehensive experiments on Replica, TUM, and real-world sequences show that RoGER-SLAM consistently improves trajectory accuracy and reconstruction quality compared with other 3DGS-SLAM systems, especially under adverse imaging conditions.

Enhancing Physical Consistency in Lightweight World Models

A major challenge in deploying world models is the trade-off between size and performance. Large world models can capture rich physical dynamics but require massive computing resources, making them impractical for edge devices. Small world models are easier to deploy but often struggle to learn accurate physics, leading to poor predictions. We propose the Physics-Informed BEV World Model (PIWM), a compact model designed to efficiently capture physical interactions in bird's-eye-view (BEV) representations. PIWM uses Soft Mask during training to improve dynamic object modeling and future prediction. We also introduce a simple yet effective technique, Warm Start, for inference to enhance prediction quality with a zero-shot model. Experiments show that at the same parameter scale (400M), PIWM surpasses the baseline by 60.6% in weighted overall score. Moreover, even when compared with the largest baseline model (400M), the smallest PIWM (130M Soft Mask) achieves a 7.4% higher weighted overall score with a 28% faster inference speed.

Model-Structured Neural Networks to Control the Steering Dynamics of Autonomous Race Cars

Autonomous racing has gained increasing attention in recent years, as a safe environment to accelerate the development of motion planning and control methods for autonomous driving. Deep learning models, predominantly based on neural networks (NNs), have demonstrated significant potential in modeling the vehicle dynamics and in performing various tasks in autonomous driving. However, their black-box nature is critical in the context of autonomous racing, where safety and robustness demand a thorough understanding of the decision-making algorithms. To address this challenge, this paper proposes MS-NN-steer, a new Model-Structured Neural Network for vehicle steering control, integrating the prior knowledge of the nonlinear vehicle dynamics into the neural architecture. The proposed controller is validated using real-world data from the Abu Dhabi Autonomous Racing League (A2RL) competition, with full-scale autonomous race cars. In comparison with general-purpose NNs, MS-NN-steer is shown to achieve better accuracy and generalization with small training datasets, while being less sensitive to the weights' initialization. Also, MS-NN-steer outperforms the steering controller used by the A2RL winning team. Our implementation is available open-source in a GitHub repository.

Towards Safe Autonomous Driving: A Real-Time Safeguarding Concept for Motion Planning Algorithms

Ensuring the functional safety of motion planning modules in autonomous vehicles remains a critical challenge, especially when dealing with complex or learning-based software. Online verification has emerged as a promising approach to monitor such systems at runtime, yet its integration into embedded real-time environments remains limited. This work presents a safeguarding concept for motion planning that extends prior approaches by introducing a time safeguard. While existing methods focus on geometric and dynamic feasibility, our approach additionally monitors the temporal consistency of planning outputs to ensure timely system response. A prototypical implementation on a real-time operating system evaluates trajectory candidates using constraint-based feasibility checks and cost-based plausibility metrics. Preliminary results show that the safeguarding module operates within real-time bounds and effectively detects unsafe trajectories. However, the full integration of the time safeguard logic and fallback strategies is ongoing. This study contributes a modular and extensible framework for runtime trajectory verification and highlights key aspects for deployment on automotive-grade hardware. Future work includes completing the safeguarding logic and validating its effectiveness through hardware-in-the-loop simulations and vehicle-based testing. The code is available at: https://github.com/TUM-AVS/motion-planning-supervisor