Searcharxiv⌕ Search

arXiv subjects

Katharina Winter

Publications and source records attributed to Katharina Winter.

10 recordsLinked to original sources

GridDriver: On-Board BEV-Based Language Planning and Reasoning

Bird's-Eye-View (BEV)-conditioned large language models (LLMs) enable language-interactive driving, but onboard deployment requires efficient perception and continuous planning alongside autoregressive language generation. We present GridDriver, an efficient BEV-conditioned LLM framework combining trajectory prediction, structured driving indicators, natural-language navigation, driving explanations, and scene question answering. Lightweight spatio-temporal BEV perception and spatially structured tokens provide scene context to a shared LLM, while frame-wise scheduling interleaves trajectory prediction with asynchronous language generation. GridDriver achieves competitive performance on public benchmarks, reaching 87.9% PDMS on NAVSIM and 57.6% accuracy on nuScenes-QA with a shared planning and QA checkpoint. Spatial representation analyses demonstrate improved geometric information retention and trajectory prediction compared with Q-Former compression. The system achieves 32.7 ms standalone planning latency and 9.4 plans/s in an onboard ROS 2 pipeline. We further evaluate planning on our own data recordings and demonstrate open-loop onboard operation on a research vehicle. Across public benchmarks and real-world evaluations, we identify strengths and limitations of the BEV-based approach.

cs.CV↗

Generative AI for Autonomous Driving: A Review

Generative AI (GenAI) is rapidly advancing the field of Autonomous Driving (AD), extending beyond traditional applications in text, image, and video generation. We explore how generative models can enhance automotive tasks, such as static map creation, dynamic scenario generation, trajectory forecasting, and vehicle motion planning. By examining multiple generative approaches ranging from Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs) to Generative Transformers (GTs) and Diffusion Models (DMs), we highlight and compare their capabilities and limitations for AD-specific applications. Additionally, we discuss hybrid methods integrating conventional techniques with generative approaches, and emphasize their improved adaptability and robustness. We also identify relevant datasets and outline open research questions to guide future developments in GenAI. Finally, we discuss five core challenges: safety, interpretability, realtime capabilities, training and benchmarking. We also present recommendations for image generation, dynamic scenario generation, and planning.

cs.CV↗

Speed in the Blind Spot: An Interpretability Analysis of Dynamic Perception in VLMs for Autonomous Driving

Vision-Language Models are increasingly used in autonomous-driving systems, yet their ability to recover dynamic physical state from visual input remains insufficiently characterized. We study velocity understanding as a controlled diagnostic across three tasks: surrounding-agent speed, current ego speed, and short-horizon future ego-speed proposal. On nuScenes, we evaluate open-weight general-purpose and PhysicalAI VLMs, together with the driving-oriented Alpamayo-1.5 Vision-Language-Action model, using multiple input and output formulations. We combine verbal evaluation with temporal perturbations, counterfactual ego-speed hints and linear probes of hidden representations. The tasks exhibit distinct failure modes. Surrounding-agent speed is weakly encoded in an agent-specific form, whereas current ego speed is often internally accessible but poorly verbalized: continuous probes achieve 4.7-5.8 km/h MAE compared with 10.2-16.8 km/h MAE for verbal outputs. Multiple frames provide inconsistent verbal gains to single frame inputs, and frame order is rarely exploited. Under non-optimized QLoRA, task-specific adaptation improves both task-relevant latent speed representations and verbal readout, but continuous surrounding-agent speed estimation remains weak, while most future-speed gains survive frame shuffling, indicating limited temporal grounding. Driving specialized Alpamayo-1.5 shows stronger latent representations for surrounding-agent and future ego speed, while current ego-speed decodability is comparable and substantial probe-verbal gaps remain. Thus, driving specialization can strengthen motion representations but does not guarantee stronger encoding across both scene and ego states or reliable readout. The results show that plausible planning outputs do not necessarily imply reliable recovery or temporal grounding of the underlying dynamic state.

cs.CV↗

"Dear LLaVA, Please Drive": A Depth-Aware Vision-Language Agent for Closed-Loop Robotic Control

Vision-language models (VLMs) provide a compelling foundation for reasoning-driven mobile navigation, offering rich contextual understanding and strong generalization from large-scale pretraining. Most existing navigation frameworks rely on imitation learning and therefore require substantial labeled trajectory data, limiting their scalability and robustness. In this work, we propose a parameter-efficient approach to fine-tune a pretrained VLM for autonomous navigation using an Imperative Learning paradigm. By optimizing against differentiable geometric cost fields rather than labeled trajectories, our model learns to generate collision-free paths exclusively from stereoscopic depth observations. We introduce a unified end-to-end navigation pipeline for natural-language-driven robotic control. This system leverages a shared VLM backbone with task-specific Low-Rank Adaptation (LoRA) modules, effectively bridging the gap from semantic target selection to low-level trajectory planning. Our approach achieves competitive Success weighted by Path Length (SPL) in unseen environments while updating less than 1% of the model's total parameters. Qualitative real-world experiments validate sim-to-real generalization and stable path planning without fine-tuning on real-world data. These results highlight a practical approach for deploying VLM-based agents on mobile robots, enabling high-level semantic navigation without the prohibitive requirement for large-scale, labeled trajectory data.

cs.RO↗

WorldVLM: Combining World Model Forecasting and Vision-Language Reasoning

Autonomous driving systems depend on on models that can reason about high-level scene contexts and accurately predict the dynamics of their surrounding environment. Vision- Language Models (VLMs) have recently emerged as promising tools for decision-making and scene understanding, offering strong capabilities in contextual reasoning. However, their limited spatial comprehension constrains their effectiveness as end-to-end driving models. World Models (WM) internalize environmental dynamics to predict future scene evolution. Recently explored as ego-motion predictors and foundation models for autonomous driving, they represent a promising direction for addressing key challenges in the field, particularly enhancing generalization while maintaining dynamic prediction. To leverage the complementary strengths of context-based decision making and prediction, we propose WorldVLM: A hybrid architecture that unifies VLMs and WMs. In our design, the high-level VLM generates behavior commands to guide the driving WM, enabling interpretable and context-aware actions. We evaluate conditioning strategies and provide insights into the hybrid design challenges.

cs.CV↗

Context-based Motion Retrieval using Open Vocabulary Methods for Autonomous Driving

Autonomous driving systems must operate reliably in safety-critical scenarios, particularly those involving unusual or complex behavior by Vulnerable Road Users (VRUs). Identifying these edge cases in driving datasets is essential for robust evaluation and generalization, but retrieving such rare human behavior scenarios within the long tail of large-scale datasets is challenging. To support targeted evaluation of autonomous driving systems in diverse, human-centered scenarios, we propose a novel context-aware motion retrieval framework. Our method combines Skinned Multi-Person Linear (SMPL)-based motion sequences and corresponding video frames before encoding them into a shared multimodal embedding space aligned with natural language. Our approach enables the scalable retrieval of human behavior and their context through text queries. This work also introduces our dataset WayMoCo, an extension of the Waymo Open Dataset. It contains automatically labeled motion and scene context descriptions derived from generated pseudo-ground-truth SMPL sequences and corresponding image data. Our approach outperforms state-of-the-art models by up to 27.5% accuracy in motion-context retrieval, when evaluated on the WayMoCo dataset.

cs.CV↗

BEV-LLM: Leveraging Multimodal BEV Maps for Scene Captioning in Autonomous Driving

Autonomous driving technology has the potential to transform transportation, but its wide adoption depends on the development of interpretable and transparent decision-making systems. Scene captioning, which generates natural language descriptions of the driving environment, plays a crucial role in enhancing transparency, safety, and human-AI interaction. We introduce BEV-LLM, a lightweight model for 3D captioning of autonomous driving scenes. BEV-LLM leverages BEVFusion to combine 3D LiDAR point clouds and multi-view images, incorporating a novel absolute positional encoding for view-specific scene descriptions. Despite using a small 1B parameter base model, BEV-LLM achieves competitive performance on the nuCaption dataset, surpassing state-of-the-art by up to 5\% in BLEU scores. Additionally, we release two new datasets - nuView (focused on environmental conditions and viewpoints) and GroundView (focused on object grounding) - to better assess scene captioning across diverse driving scenarios and address gaps in current benchmarks, along with initial benchmarking results demonstrating their effectiveness.

cs.CV↗

BEVDriver: Leveraging BEV Maps in LLMs for Robust Closed-Loop Driving

Autonomous driving has the potential to set the stage for more efficient future mobility, requiring the research domain to establish trust through safe, reliable and transparent driving. Large Language Models (LLMs) possess reasoning capabilities and natural language understanding, presenting the potential to serve as generalized decision-makers for ego-motion planning that can interact with humans and navigate environments designed for human drivers. While this research avenue is promising, current autonomous driving approaches are challenged by combining 3D spatial grounding and the reasoning and language capabilities of LLMs. We introduce BEVDriver, an LLM-based model for end-to-end closed-loop driving in CARLA that utilizes latent BEV features as perception input. BEVDriver includes a BEV encoder to efficiently process multi-view images and 3D LiDAR point clouds. Within a common latent space, the BEV features are propagated through a Q-Former to align with natural language instructions and passed to the LLM that predicts and plans precise future trajectories while considering navigation instructions and critical scenarios. On the LangAuto benchmark, our model reaches up to 18.9% higher performance on the Driving Score compared to SoTA methods.

cs.RO↗

MEDIATE: Mutually Endorsed Distributed Incentive Acknowledgment Token Exchange

Recent advances in multi-agent systems (MAS) have shown that incorporating peer incentivization (PI) mechanisms vastly improves cooperation. Especially in social dilemmas, communication between the agents helps to overcome sub-optimal Nash equilibria. However, incentivization tokens need to be carefully selected. Furthermore, real-world applications might yield increased privacy requirements and limited exchange. Therefore, we extend the PI protocol for mutual acknowledgment token exchange (MATE) and provide additional analysis on the impact of the chosen tokens. Building upon those insights, we propose mutually endorsed distributed incentive acknowledgment token exchange (MEDIATE), an extended PI architecture employing automatic token derivation via decentralized consensus. Empirical results show the stable agreement on appropriate tokens yielding superior performance compared to static tokens and state-of-the-art approaches in different social dilemma environments with various reward distributions.

cs.MA↗

How do Quantifiers Affect the Quality of Requirements?

Context: Requirements quality can have a substantial impact on the effectiveness and efficiency of using requirements artifacts in a development process. Quantifiers such as "at least", "all", or "exactly" are common language constructs used to express requirements. Quantifiers can be formulated by affirmative phrases ("At least") or negative phrases ("Not less than"). Problem: It is long assumed that negation in quantification negatively affects the readability of requirements, however, empirical research on these topics remains sparse. Principal Idea: In a web-based experiment with 51 participants, we compare the impact of negations and quantifiers on readability in terms of reading effort, reading error rate and perceived reading difficulty of requirements. Results: For 5 out of 9 quantifiers, our participants performed better on the affirmative phrase compared to the negative phrase. Only for one quantifier, the negative phrase was more effective. Contribution: This research focuses on creating an empirical understanding of the effect of language in Requirements Engineering. It furthermore provides concrete advice on how to phrase requirements.

cs.SE↗