The ability to operate safely in increasingly complex traffic scenarios is a fundamental requirement for Automated Driving Systems (ADS). Ensuring the safe release of ADS functions necessitates a precise understanding of the occurring traffic scenarios. To support this objective, this work introduces a pipeline for traffic scenario clustering and the analysis of scenario category completeness. The Clustering Vector Quantized - Variational Autoencoder (CVQ-VAE) is employed for the clustering of highway traffic scenarios and utilized to create various catalogs with differing numbers of traffic scenario categories. Subsequently, the impact of the number of categories on the completeness considerations of the traffic scenario categories is analyzed. The results show an outperforming clustering performance compared to previous work. The trade-off between cluster quality and the amount of required data to maintain completeness is discussed based on the publicly available highD dataset.

Reliable and precise object tracking is an essential requirement for automated driving. The majority of LiDAR-based tracking algorithms resort to raw range measurements only. In contrast, we propose a novel method to extract compact and salient features from LiDAR intensities. Using the example of an evasive steering maneuver of a leading vehicle, we show that leveraging these intensity features allows for a more accurate estimation of object states. The resulting early detection of target object rotation allows an automated driving system additional time for deriving an appropriate driving policy.

Advanced Driver Assistance Systems (ADASs), such as autonomous driving, require the continuous computation and update of detailed environment maps. Today's standard processors in automotive Electronic Control Units (ECUs) struggle to provide enough computing power for those tasks. Here, new architectures, like Graphics Processing Units (GPUs) might be a promising accelerator candidate for ECUs. Current algorithms have to be adapted to these new architectures when possible, or new algorithms have to be designed to take advantage of these architectures. In this paper, we propose a novel parallel update algorithm, called cell-based update algorithm for occupancy grid maps, which exploits the highly parallel architecture of GPUs and overcomes the shortcomings of previous implementations based on the Bresenham algorithm on such architectures. A second contribution is a new hybrid map, which takes the advantages of the classic occupancy grid map and reduces the computational effort of those. All algorithms are parallelized and implemented on a discrete GPU as well as on an embedded GPU (Nvidia Tegra K1 Jetson board). Compared with the state-of-the-art Bresenham algorithm as used in the case of occupancy grid maps, our parallelized cell-based update algorithm and our proposed hybrid map approach achieve speedups of up to 2.5 and 4.5, respectively.

Reliable object perception is a vital requirement for automated driving. Despite the availability of precise contour measurements, most state-of-the-art tracking systems still represent object geometry as bounding boxes. However, there are objects operating in public traffic for which the box assumption is highly inappropriate. We therefore propose to represent object contours using 2D polylines. Taking into account the mutual dependence of object poses and shape, our tracking framework targets at a simultaneous estimation of both states. Moreover, we propose to augment scan segments with free-space information at their boundaries and show how this knowledge can be incorporated into the tracking framework and beyond. Evaluation with real scan data shows that our method produces accurate dynamic estimates and consistent shape reconstructions.

Future Advanced Driver Assistance Systems (ADAS) need to create an accurate model of the environment. Accordingly, an enormous amount of data has to be fused and processed. From this data, information such as the positions of the vehicles, has to be extracted out of the model, e.g., to create a convoy track. Common architectures used today, like single-core processors in automotive Electronic Control Units (ECUs), struggle to provide enough computing power for those tasks. Here, emerging embedded multi-core architectures are appealing such as embedded Graphics Processing Units (GPUs). In this paper, we present a novel parallelization of a convoy track detection algorithm. Moreover, in order to profit best from for embedded GPUs, special techniques such as Zero Copy are exploited to parallelize our application. As an experimental platform, an Nvidia Tegra K1 is used, which is also common in the automotive industry. For different scenarios, we illustrate the limitations of the system and algorithm. Yet, impressive speedups with respect to a single-core CPU solution of up to nine may be achieved using the proposed parallelization techniques in case of high traffic situations.

A road curb detection algorithm for a 3D sensor, e.g. a dense stereo camera, is presented in this paper. The road curb detection is based on a digital elevation map. Different techniques and coordinate systems for mapping the height values are compared theoretically and by simulating a different quality of ego motion data. Furthermore we introduce a new approach of finding road curbs in an elevation map, which is based on a calculation of the most probable path. Using an elevation map the curb height can be calculated in an additional step. For evaluation we use highly accurate reference sensors and compare the detected curbs to a ground truth. Additionally we introduce a novel criteria to describe the quality of an elevation map and discuss the results. The road detection algorithm works in real-time and has a position accuracy of about 10 cm and an height error of about 1.5 cm.

In this paper an approach for 360 degree multi sensor fusion for static and dynamic obstacles is presented. The perception of static and dynamic obstacles is achieved by combining the advantages of model based object tracking and an occupancy map. For the model based object tracking a novel multi reference point tracking system, called best knowledge model, is introduced. The best knowledge model allows to track and describe objects with respect to a best suitable reference point. It is explained how the object tracking and the occupancy map closely interact and benefit from each other. Experimental results of the 360 degree multi sensor fusion system from an automotive test vehicle are shown.

in this paper we present a novel environment perception system based on an occupancy grid mapping and a multi-object tracking. The goal of such a system is to create a harmonic, consistent and complete representation of the vehicle environment as a base for future advanced driver assistance systems. In addition to a mathematical formulation of the problem we present a robust algorithm to detect dynamic obstacles from the occupancy map and show how both, the mapping process and the tracking can benefit from each other. Therefore, the concept of moving objects with associated dynamic cells is introduced. The presented techniques are applicable to both 2D and 3D mapping and can be also extended to correct the ego motion from the occupancy map and the object tracks. Unlike many publications over the last years our work provides real time performance and an accurate detection of obstacles with real laser and radar sensors and can fulfill the requirements of future driver assistance systems.

This paper deals with localization of a mobile robot using received signal strength indicator (RSSI) in low power IEEE 802.15.4 conform wireless communication in an outdoor environment. Hardware modifications are derived to reduce the radio irregularity and to increase the uniqueness of the measured RSSI to distance. A novel algorithm is elaborated allowing for sub meter accuracy. It accounts for the noise and implicitly models the uncertainty. Additionally it is robust to node failures. To further improve the accuracy a particle filter is employed to perform probabilistic sensor fusion of odometry, ultrasonic and RSSI sensors and a map. The suitability of this approach is shown with real measurements achieving a mean positioning error of 0.32 m.