Benchmarking Vision Foundation Models for Input Monitoring in Autonomous Driving

Deep neural networks (DNNs) remain challenged by distribution shifts in complex open-world domains like automated driving (AD): Absolute robustness against yet unknown novel objects (semantic shift) or styles like lighting conditions (covariate shift) cannot be guaranteed. Hence, reliable operation-time monitors for identification of out-of-training-data-distribution (OOD) scenarios are imperative. Current approaches for OOD classification are untested for complex domains like AD, are limited in the kinds of shifts they detect, or even require supervision with OOD samples. To prepare for unanticipated shifts, we instead establish a framework around a principled, unsupervised, and model-agnostic method that unifies detection of all kinds of shifts: Find a full model of the training data's feature distribution, to then use its density at new points as in-distribution (ID) score. To implement this, we propose to combine the newly available Vision Foundation Models (VFM) as feature extractors with one of four alternative density modeling techniques. In an extensive benchmark of 4 VFMs against 20 baselines, we show the superior performance of VFM feature encodings compared to shift-specific OOD monitors. Additionally, we find that sophisticated architectures outperform larger latent space dimensionality; and our method identifies samples with higher risk of errors on downstream tasks, despite being model-agnostic. This suggests that VFMs are promising to realize model-agnostic, unsupervised, reliable safety monitors in complex vision tasks.

Learning Monocular Depth from Events via Egomotion Compensation

Event cameras are neuromorphically inspired sensors that sparsely and asynchronously report brightness changes. Their unique characteristics of high temporal resolution, high dynamic range, and low power consumption make them well-suited for addressing challenges in monocular depth estimation (e.g., high-speed or low-lighting conditions). However, current existing methods primarily treat event streams as black-box learning systems without incorporating prior physical principles, thus becoming over-parameterized and failing to fully exploit the rich temporal information inherent in event camera data. To address this limitation, we incorporate physical motion principles to propose an interpretable monocular depth estimation framework, where the likelihood of various depth hypotheses is explicitly determined by the effect of motion compensation. To achieve this, we propose a Focus Cost Discrimination (FCD) module that measures the clarity of edges as an essential indicator of focus level and integrates spatial surroundings to facilitate cost estimation. Furthermore, we analyze the noise patterns within our framework and improve it with the newly introduced Inter-Hypotheses Cost Aggregation (IHCA) module, where the cost volume is refined through cost trend prediction and multi-scale cost consistency constraints. Extensive experiments on real-world and synthetic datasets demonstrate that our proposed framework outperforms cutting-edge methods by up to 10\% in terms of the absolute relative error metric, revealing superior performance in predicting accuracy.

Virtualization & Microservice Architecture for Software-Defined Vehicles: An Evaluation and Exploration

The emergence of Software-Defined Vehicles (SDVs) signifies a shift from a distributed network of electronic control units (ECUs) to a centralized computing architecture within the vehicle's electrical and electronic systems. This transition addresses the growing complexity and demand for enhanced functionality in traditional E/E architectures, with containerization and virtualization streamlining software development and updates within the SDV framework. While widely used in cloud computing, their performance and suitability for intelligent vehicles have yet to be thoroughly evaluated. In this work, we conduct a comprehensive performance evaluation of containerization and virtualization on embedded and high-performance AMD64 and ARM64 systems, focusing on CPU, memory, network, and disk metrics. In addition, we assess their impact on real-world automotive applications using the Autoware framework and further integrate a microservice-based architecture to evaluate its start-up time and resource consumption. Our extensive experiments reveal a slight 0-5% performance decline in CPU, memory, and network usage for both containerization and virtualization compared to bare-metal setups, with more significant reductions in disk operations-5-15% for containerized environments and up to 35% for virtualized setups. Despite these declines, experiments with actual vehicle applications demonstrate minimal impact on the Autoware framework, and in some cases, a microservice architecture integration improves start-up time by up to 18%.

Gassidy: Gaussian Splatting SLAM in Dynamic Environments

3D Gaussian Splatting (3DGS) allows flexible adjustments to scene representation, enabling continuous optimization of scene quality during dense visual simultaneous localization and mapping (SLAM) in static environments. However, 3DGS faces challenges in handling environmental disturbances from dynamic objects with irregular movement, leading to degradation in both camera tracking accuracy and map reconstruction quality. To address this challenge, we develop an RGB-D dense SLAM which is called Gaussian Splatting SLAM in Dynamic Environments (Gassidy). This approach calculates Gaussians to generate rendering loss flows for each environmental component based on a designed photometric-geometric loss function. To distinguish and filter environmental disturbances, we iteratively analyze rendering loss flows to detect features characterized by changes in loss values between dynamic objects and static components. This process ensures a clean environment for accurate scene reconstruction. Compared to state-of-the-art SLAM methods, experimental results on open datasets show that Gassidy improves camera tracking precision by up to 97.9% and enhances map quality by up to 6%.

Adopting RAG for LLM-Aided Future Vehicle Design

In this paper, we explore the integration of Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) to enhance automated design and software development in the automotive industry. We present two case studies: a standardization compliance chatbot and a design copilot, both utilizing RAG to provide accurate, context-aware responses. We evaluate four LLMs-GPT-4o, LLAMA3, Mistral, and Mixtral -- comparing their answering accuracy and execution time. Our results demonstrate that while GPT-4 offers superior performance, LLAMA3 and Mistral also show promising capabilities for local deployment, addressing data privacy concerns in automotive applications. This study highlights the potential of RAG-augmented LLMs in improving design workflows and compliance in automotive engineering.

Online Collision Risk Estimation via Monocular Depth-Aware Object Detectors and Fuzzy Inference

This paper presents a monitoring framework that infers the level of autonomous vehicle (AV) collision risk based on its object detector's performance using only monocular camera images. Essentially, the framework takes two sets of predictions produced by different algorithms and associates their inconsistencies with the collision risk via fuzzy inference. The first set of predictions is obtained through retrieving safety-critical 2.5D objects from a depth map, and the second set comes from the AV's 3D object detector. We experimentally validate that, based on Intersection-over-Union (IoU) and a depth discrepancy measure, the inconsistencies between the two sets of predictions strongly correlate to the safety-related error of the 3D object detector against ground truths. This correlation allows us to construct a fuzzy inference system and map the inconsistency measures to an existing collision risk indicator. In particular, we apply various knowledge- and data-driven techniques and find using particle swarm optimization that learns general fuzzy rules gives the best mapping result. Lastly, we validate our monitor's capability to produce relevant risk estimates with the large-scale nuScenes dataset and show it can safeguard an AV in closed-loop simulations.

Motion Planning for Robotics: A Review for Sampling-based Planners

Recent advancements in robotics have transformed industries such as manufacturing, logistics, surgery, and planetary exploration. A key challenge is developing efficient motion planning algorithms that allow robots to navigate complex environments while avoiding collisions and optimizing metrics like path length, sweep area, execution time, and energy consumption. Among the available algorithms, sampling-based methods have gained the most traction in both research and industry due to their ability to handle complex environments, explore free space, and offer probabilistic completeness along with other formal guarantees. Despite their widespread application, significant challenges still remain. To advance future planning algorithms, it is essential to review the current state-of-the-art solutions and their limitations. In this context, this work aims to shed light on these challenges and assess the development and applicability of sampling-based methods. Furthermore, we aim to provide an in-depth analysis of the design and evaluation of ten of the most popular planners across various scenarios. Our findings highlight the strides made in sampling-based methods while underscoring persistent challenges. This work offers an overview of the important ongoing research in robotic motion planning.

Unveiling Ontological Commitment in Multi-Modal Foundation Models

Ontological commitment, i.e., used concepts, relations, and assumptions, are a corner stone of qualitative reasoning (QR) models. The state-of-the-art for processing raw inputs, though, are deep neural networks (DNNs), nowadays often based off from multimodal foundation models. These automatically learn rich representations of concepts and respective reasoning. Unfortunately, the learned qualitative knowledge is opaque, preventing easy inspection, validation, or adaptation against available QR models. So far, it is possible to associate pre-defined concepts with latent representations of DNNs, but extractable relations are mostly limited to semantic similarity. As a next step towards QR for validation and verification of DNNs: Concretely, we propose a method that extracts the learned superclass hierarchy from a multimodal DNN for a given set of leaf concepts. Under the hood we (1) obtain leaf concept embeddings using the DNN's textual input modality; (2) apply hierarchical clustering to them, using that DNNs encode semantic similarities via vector distances; and (3) label the such-obtained parent concepts using search in available ontologies from QR. An initial evaluation study shows that meaningful ontological class hierarchies can be extracted from state-of-the-art foundation models. Furthermore, we demonstrate how to validate and verify a DNN's learned representations against given ontologies. Lastly, we discuss potential future applications in the context of QR.

Learning Task Planning from Multi-Modal Demonstration for Multi-Stage Contact-Rich Manipulation

Large Language Models (LLMs) have gained popularity in task planning for long-horizon manipulation tasks. To enhance the validity of LLM-generated plans, visual demonstrations and online videos have been widely employed to guide the planning process. However, for manipulation tasks involving subtle movements but rich contact interactions, visual perception alone may be insufficient for the LLM to fully interpret the demonstration. Additionally, visual data provides limited information on force-related parameters and conditions, which are crucial for effective execution on real robots. In this paper, we introduce an in-context learning framework that incorporates tactile and force-torque information from human demonstrations to enhance LLMs' ability to generate plans for new task scenarios. We propose a bootstrapped reasoning pipeline that sequentially integrates each modality into a comprehensive task plan. This task plan is then used as a reference for planning in new task configurations. Real-world experiments on two different sequential manipulation tasks demonstrate the effectiveness of our framework in improving LLMs' understanding of multi-modal demonstrations and enhancing the overall planning performance.

TacDiffusion: Force-domain Diffusion Policy for Precise Tactile Manipulation

Assembly is a crucial skill for robots in both modern manufacturing and service robotics. However, mastering transferable insertion skills that can handle a variety of high-precision assembly tasks remains a significant challenge. This paper presents a novel framework that utilizes diffusion models to generate 6D wrench for high-precision tactile robotic insertion tasks. It learns from demonstrations performed on a single task and achieves a zero-shot transfer success rate of 95.7% across various novel high-precision tasks. Our method effectively inherits the self-adaptability demonstrated by our previous work. In this framework, we address the frequency misalignment between the diffusion policy and the real-time control loop with a dynamic system-based filter, significantly improving the task success rate by 9.15%. Furthermore, we provide a practical guideline regarding the trade-off between diffusion models' inference ability and speed.