Railway systems, particularly in Germany, require high levels of automation to address legacy infrastructure challenges and increase train traffic safely. A key component of automation is robust long-range perception, essential for early hazard detection, such as obstacles at level crossings or pedestrians on tracks. Unlike automotive systems with braking distances of 70 meters, trains require perception ranges exceeding 1 km. This paper presents an deep-learning-based approach for long-range 3D object detection tailored for autonomous trains. The method relies solely on monocular images, inspired by the Faraway-Frustum approach, and incorporates LiDAR data during training to improve depth estimation. The proposed pipeline consists of four key modules: (1) a modified YOLOv9 for 2.5D object detection, (2) a depth estimation network, and (3–4) dedicated short- and long-range 3D detection heads. Evaluations on the OSDaR23 dataset demonstrate the effectiveness of the approach in detecting objects up to 250 meters. Results highlight its potential for railway automation and outline areas for future improvement.

Robust object detection for Unmanned Surface Vehicles (USVs) in complex water environments is essential for reliable navigation and operation. Specifically, water surface object detection faces challenges from blurred edges and diverse object scales. Although vision-radar fusion offers a feasible solution, existing approaches suffer from cross-modal feature conflicts, which negatively affect model robustness. To address this problem, we propose a robust vision-radar fusion model WS-DETR. In particular, we first introduce a Multi-Scale Edge Information Integration (MSEII) module to enhance edge perception and a Hierarchical Feature Aggregator (HiFA) to boost multi-scale object detection in the encoder. Then, we adopt self-moving point representations for continuous convolution and residual connection to efficiently extract irregular features under the scenarios of irregular point cloud data. To further mitigate cross-modal conflicts, an Adaptive Feature Interactive Fusion (AFIF) module is introduced to integrate visual and radar features through geometric alignment and semantic fusion. Extensive experiments on the WaterScenes dataset demonstrate that WS-DETR achieves state-of-the-art (SOTA) performance, maintaining its superiority even under adverse weather and lighting conditions.

Dynamic Occupancy Grid Mapping is a technique used to generate a local map of the environment, containing both static and dynamic information. Typically, these maps are primarily generated using lidar measurements. However, with improvements in radar sensing, resulting in better accuracy and higher resolution, radar is emerging as a viable alternative to lidar as the primary sensor for mapping. In this paper, we propose a radar-centric dynamic occupancy grid mapping algorithm with adaptations to the state computation, inverse sensor model, and field-of-view computation tailored to the specifics of radar measurements. We extensively evaluate our approach with real data to demonstrate its effectiveness and establish the first benchmark for radar-based dynamic occupancy grid mapping using the publicly available Radarscenes dataset.

This paper introduces a novel hybrid architecture that enhances radar-based Dynamic Occupancy Grid Mapping (DOGM) for autonomous vehicles, integrating deep learning for state-classification. Traditional radar-based DOGM often faces challenges in accurately distinguishing between static and dynamic objects. Our approach addresses this limitation by introducing a neural network-based DOGM state correction mechanism, designed as a semantic segmentation task, to refine the accuracy of the occupancy grid. Additionally a heuristic fusion approach is proposed which allows to enhance performance without compromising on safety. We extensively evaluate this hybrid architecture on the NuScenes Dataset, focusing on its ability to improve dynamic object detection as well grid quality. The results show clear improvements in the detection capabilities of dynamic objects, highlighting the effectiveness of the deep learning-enhanced state correction in radar-based DOGM.

In recent years, the rise in both natural and man-made disasters, along with armed conflicts and terrorist threats, has elevated the demand for Search and Rescue (SAR) missions worldwide. This paper underscores the critical necessity to enhance SAR capacity, safety, and capabilities, with a primary goal of reducing response times through the integration of robots into SAR operations. The examination of research on robotized SAR highlights deficiencies in both software and hardware, particularly focusing on perception systems for robotized SAR platforms.

With the flourishing development of intelligent warehousing systems, the technology of Automated Guided Vehicle (AGV) has experienced rapid growth. Within intelligent warehousing environments, AGV is required to safely and rapidly plan an optimal path in complex and dynamic environments. Most research has studied deep reinforcement learning to address this challenge. However, in the environments with sparse extrinsic rewards, these algorithms often converge slowly, learn inefficiently or fail to reach the target. Random Network Distillation (RND), as an exploration enhancement, can effectively improve the performance of proximal policy optimization, especially enhancing the additional intrinsic rewards of the AGV agent which is in sparse reward environments. Moreover, most of the current research continues to use 2D grid mazes as experimental environments. These environments have insufficient complexity and limited action sets. To solve this limitation, we present simulation environments of AGV path planning with continuous actions and positions for AGVs, so that it can be close to realistic physical scenarios. Based on our experiments and comprehensive analysis of the proposed method, the results demonstrate that our proposed method enables AGV to more rapidly complete path planning tasks with continuous actions in our environments. A video of part of our experiments can be found at https://youtu.be/lwrY9YesGmw.

When designing path tracking controllers for car-like vehicles, two main aspects are the tracking performance and the characteristics of the actuation signal. Our work bases on an existing variable structure controller that was designed with the geometrically optimal solution of a Dubins car, but with chattering on the output. In this contribution, we extend this approach to achieve an actuation signal that is$n$-smooth and differentially bounded. While the global stability under matched disturbances is maintained, the finite time reaching behavior is exchanged for asymptotic convergence. With the$n$-smooth output,$n$new parameters are introduced, weighing the control characteristics between the tracking performance and the magnitude of the steering angles derivatives. The controller is also evaluated in simulation, demonstrating the tuning capabilities, as well as the reaching and tracking behavior. The main contribution of this work is a control law designed to produce a smooth steering angle with implicit satisfaction of bounds on its derivatives.

Virtual testing and validation of autonomous systems require real-time capable sensor models to couple the system under test with the simulated environment. This paper proposes a lidar modeling approach using back projection of 360$^{\circ}$depth images for point cloud computation. Beam incident angles are derived from estimated surface normal vectors and allow for intensity calculation. Furthermore, spatial beam divergence and multiple return modes can be simulated. Using the depth buffer as primary input source facilitates the integration with different environment simulation tools. Virtual Test Drive (VTD) and CARLA serve as example cases for interfacing with the developed model. An analysis of sampling errors resulting from the underlying model principles is presented. Finally, the normal vector estimation precision and the computation time are evaluated.

Cooperative Intelligent Transportation Systems (C-ITSs) facilitate safe, efficient, and sustainable driving through vehicle-to-everything (V2X) communication. Standardized services and messages regulate the exchange of basic information required for an unrestricted view of the traffic situation. However, many applications need information that goes beyond. Furthermore, compliance with these standards is cumbersome when testing on a small-scale or in simulation. The result: standards are often bypassed. Therefore, this paper presents a standard-compliant, lightweight, easy-to-integrate library that is application-independent and open source. We integrate the library in a small-scale test bed and use Decentralized Environmental Notification Messages (DENMs) to trigger the speed reduction of a vehicle in a bad weather situation. Through Cooperative Awareness Messages (CAMs), the vehicle shares its status information, including its current velocity. The experiment demonstrates a working location-dependent exchange of information between a Road Side Unit (RSU) and the vehicle, utilizing LEVEL as a communication stack.

Path tracking controllers for non-holonomic vehicles are commonly designed with primary focus on robustness to different kinds of disturbances, vehicle dynamics and other effects. Subsequently, the path tracking behavior is improved by parameter optimization or controller extensions. The possible improvement, however, is already limited by the control design itself. To overcome this drawback, this paper presents an alternative approach: In a first step, we formulate the geometrically optimal solution for a Dubins car reaching a straight target path as a globally stabilizing sliding-mode feedback law. In a second step, we introduce a single tuning parameter that achieves robustness with respect to exemplary matched disturbances. The result is a sub-optimal but robust controller. We investigate the impact of tuning this parameter towards optimality or robustness with respect to the transient and stationary tracking behavior in simulation scenarios. Following the presented approach, it is possible to adapt the control law with respect to various disturbances facing and utilizing the trade-off between optimality and robustness. This is a significant advantage in the design of path tracking controllers due to the variety of vehicle actuation and measurement systems and operational design domains.

The assessment of automated driving relies increasingly on scenario-based virtual tests to achieve sufficient test coverage. Scenarios are generally based on ground truth information. Therefore, it is necessary to reproduce the view of the environment of the automated vehicle as it is seen by the autonomous driving function in the simulation. Typically, this view is erroneous compared to the ground truth due to sensor errors. This paper presents a novel approach to cluster, identify and finally to reproduce sensor errors by maneuver-dependent statistical models for the detection of other traffic objects. Errors are classified by their static and dynamic influences and incorporated into individual error models. These are evaluated in a final step based on real driving data.

Assessment and testing are among the biggest challenges for the release of automated driving. Up to this date, the exact procedure to achieve homologation is not settled. Current research focuses on scenario-based approaches that represent driving scenarios as test cases within a scenario space. This avoids redundancies in testing, enables the inclusion of virtual testing into the process, and makes a statement about test coverage possible. However, it is unclear how to define such a scenario space and the coverage criterion. This work presents a novel approach to the definition of the scenario space. Spatiotemporal filtering on naturalistic highway driving data provides a large amount of driving scenarios as a foundation. A custom distance measure between scenarios enables hierarchical agglomerative clustering, categorizing the scenarios into subspaces. The members of a resulting cluster found through this approach reveal a common structure that is visually observable. We discuss a data-driven solution to define the necessary test coverage for the assessment of automated driving. Finally, the contribution of the findings to achieve homologation is elaborated.

As racing sports pushed the technological limits of vehicles in the past, automated racing has the potential to directly evaluate new technologies from research in a competitive environment. The required speed of implementation and the fair evaluation criteria enable very fast progress in selecting potential strategies, architectures and algorithms. By this, automated racing strongly contributes to science as well as to future every day implementations of driving automation. In this publication, the approaches used by the team Autonomous-Racing-Graz at the ROBORACE final races in season alpha are revealed. The contribution describes how to generate a smooth reference line in case of non-smooth borders and additional precision gates on a race track. The generation of an initial smooth reference line is key to assure convergence in later optimization. The described data processing is based on track geometry defined by borders only. The specific challenge addressed in the approach was the incorporation of tight, short gates and the low computational effort of the algorithms, enabling fast tuning and adaptation during racing events onsite.

Cooperative driving is a key enabler for safe and efficient driving. The European Telecommunications Standards Institute (ETSI) published standards for message dissemination and packet formats for continuous status notification, i.e., Cooperative Awareness Message (CAM), and asynchronous event notification, i.e., Decentralized Environmental Notification Message (DENM). Cooperative Intelligent Transportation Systems (C-ITS), such as platooning, rely on these messages that are exchanged between vehicles and infrastructure. However, CAM and DENM do not target cooperativeness: future paths can not be shared. Therefore, in this paper we present a modified CAM structure that allows a vehicle to indicate its own future trajectory over a variable spatial and temporal horizon. It may be used in combination with DENM to cooperatively handle events and limiting their impact. The presented approach is tested in simulation using Omnet++ and Simulation of Urban Mobility (SUMO) considering an IEEE 802.11p network.

Reachability analysis is used to define spatial areas based on physical constraints, where the pedestrian might go. A new jerk constrained model for the pedestrian is introduced and combined with a Gaussian Process prediction model. A mixed integer linear programming approach is used for motion planning to compute optimal trajectories for the vehicle. Based on the safety constraints it is ensured that the vehicle does not collide with the pedestrian. The approach is physically inspired and it can be proven for all kind of pedestrians. Only the maximum values for the jerk, acceleration, and velocity of the pedestrian are necessary to predict future trajectory with absolute certainty.

Safe autonomous driving in urban areas requires robust algorithms to avoid collisions with other traffic participants with limited perception ability. Current deployed approaches relying on Autonomous Emergency Braking (AEB) systems are often overly conservative. In this work, we formulate the problem as a partially observable Markov decision process (POMDP), to derive a policy robust to uncertainty in the pedestrian location. We investigate how to integrate such a policy with an AEB system that operates only when a collision is unavoidable. In addition, we propose a rigorous evaluation methodology on a set of well-defined scenarios. We show that combining the two approaches provides a robust autonomous braking system that reduces unnecessary braking caused by using the AEB system on its own.

Among the actual trends that will affect society in the coming years, autonomous driving stands out as having the potential to disruptively change the automotive industry as we know it today. As a consequence, this will also highly impact the semiconductor industry and open new market opportunities, since semiconductors play an indispensable role as enablers for automated vehicles. Fully automated driving has been identified as one major enabler to master the Grand Societal Challenges of safe, clean, and efficient mobility. For this, fail-operational behavior is essential in the sense, plan, and act stages of the automation chain in order to handle safety-critical situations by its own, which currently is not reached with state-of-the-art approaches also due to missing reliable environment perception and sensor fusion. PRYSTINE, which was the highest ranked ECSEL project proposal in 2017, will realize Fail-operational Urban Surround perceptION (FUSION) which is based on robust Radar and LiDAR sensor fusion and control functions in order to enable safe automated driving in urban and rural environments. Furthermore, PRYSTINE will strengthen and extend traditional core competencies of the European industry, research organizations, and universities in smart mobility and in particular in the electronic component and systems and cyber-physical systems domain.

Co-simulation is a powerful approach to verify a system design and to support concept decisions early in the automotive development process. Due to the heterogeneous nature of the co-simulation framework there is a lot of potential for variability requiring the systematic handling of it.

The introduction of FlexRay in the automotive industry as well as the standardization efforts for the hardware and software architecture rise different challenges: How to efficiently integrate the new services in the system, how to validate the communication architecture with respect to the effects influencing the network? The research project TEODACS aims at answering these questions. We present in this document a new validation approach for the communication architecture based on the simulation of the entire FlexRay network within a co-simulation framework. This environment largely reduces the processing resources required for simulation and thus makes the analysis of such complex system possible.