Search Results

Convolutional neural networks are the de facto method of processing camera, radar, and lidar data for use in perception in ADAS and L4 vehicles, yet their operation is a black box to many engineers. Unlike traditional rules-based approaches to coding intelligent systems, networks are trained and the internal structure created during the training process is too complex to be understood by humans, yet in operation networks are able to classify objects of interest at error rates better than rates achieved by humans viewing the same input data.

The conventional process of last-mile delivery logistics often leads to safety problems for road users and a high level of environmental pollution. Delivery drivers must deal with frequent stops, search for a convenient parking spot and sometimes navigate through the narrow streets causing traffic congestion and possibly safety issues for the ego vehicle as well as for other traffic participants. This process is not only time consuming but also environmentally impactful, especially in low-emission zones where prolonged vehicle idling can lead to air pollution and to high operational costs. To overcome these challenges, a reliable system is required that not only ensures the flexible, safe and smooth delivery of goods but also cuts the costs and meets the delivery target.

In the evolving landscape of automated driving systems, the critical role of vehicle localization within the autonomous driving stack is increasingly evident. Traditional reliance on Global Navigation Satellite Systems (GNSS) proves to be inadequate, especially in urban areas where signal obstruction and multipath effects degrade accuracy. Addressing this challenge, this paper details the enhancement of a localization system for autonomous public transport vehicles, focusing on mitigating GNSS errors through the integration of a LiDAR sensor. The approach involves creating a 3D map using the factor graph-based LIO-SAM algorithm based on GNSS, vehicle odometry, IMU and LiDAR data. The algorithm is adapted to the use-case by adding a velocity factor and altitude data from a Digital Terrain model. Based on the map a state estimator is proposed, which combines high-frequency LiDAR odometry based on FAST-LIO with low-frequency absolute multiscale ICP-based LiDAR position estimation.

Autonomous Driving is being utilized in various settings, including indoor areas such as industrial halls. Additionally, LIDAR sensors are currently popular due to their superior spatial resolution and accuracy compared to RADAR, as well as their robustness to varying lighting conditions compared to cameras. They enable precise and real-time perception of the surrounding environment. Several datasets for on-road scenarios such as KITTI or Waymo are publicly available. However, there is a notable lack of open-source datasets specifically designed for industrial hall scenarios, particularly for 3D LIDAR data. Furthermore, for industrial areas where vehicle platforms with omnidirectional drive are often used, 360° FOV LIDAR sensors are necessary to monitor all critical objects. Although high-resolution sensors would be optimal, mechanical LIDAR sensors with 360° FOV exhibit a significant price increase with increasing resolution.

The automotive PowerNet is facing a major transformation. The three main drivers are: • Increasing power • Availability requirements • PowerNet complexity and cost reduction These driving factors result in a wide variety of possible future PowerNet topologies. The increasing power demand is among others caused by the progressive electrification of formerly mechanical components and the trend of increasing number of comfort loads. This leads to a steady increase in installed electrical power. X-by-wire systems and autonomous driving functions result in higher availability requirements. As a result, the power supply of all safety-critical loads must always be kept sufficiently stable. To reduce costs and increase reliability, the car manufacturers aim to reduce the complexity of the PowerNet System, including the wiring harness and the controller network. The wiring harness e.g., is currently one of the costliest parts of modern cars. These challenges are met with different concepts.

Investigating human driver behavior enhances the acceptance of the autonomous driving and increases road safety in heterogeneous environments with human-operated and autonomous vehicles. The previously established driver fingerprint model, focuses on the classification of driving style based on CAN bus signals. However, driving styles are inherently complex and influenced by multiple factors, including changing driving environments and driver states. To comprehensively create a driver profile, an in-car measurement system based on the Driver-Driven vehicle-Driving environment (3D) framework is developed. The measurement system records emotional and physiological signals from the driver, including ECG signal and heart rate. A Raspberry Pi camera is utilized on the dashboard to capture the driver's facial expressions and a trained convolutional neural network (CNN) recognizes emotion. To conduct unobtrusive ECG measurements, an ECG sensor is integrated into the steering wheel.

Autonomous driving is a hot topic in the automotive domain, and there is an increasing need to prove its reliability. They use machine learning techniques, which are themselves stochastic techniques based on some kind of statistical inference. The occurrence of incorrect decisions is part of this approach and often not directly related to correctable errors. The quality of the systems is indicated by statistical key figures such as accuracy and precision. Numerous driving tests and simulations in simulators are extensively used to provide evidence. However, the basis of all descriptive statistics is a random selection from a probability space. The difficulty in testing or constructing the training and test data set is that this probability space is usually not well defined. To systematically address this shortcoming, ontologies have been and are being developed to capture the various concepts and properties of the operational design domain.

The transition from ICE to electric power trains in new vehicles along with the application of advanced active and passive noise reduction solutions has intensified the perception of noise sources not directly linked to the propulsion system. This includes road noise as amplified by the tire cavity resonance. This resonance mainly depends on tire geometry, air temperature inside the tire and vehicle speed and is increasingly audible for larger wheels and heavier vehicles, as they are typical for current electrical SUV designs. Active technologies can be applied to significantly reduce narrow band tire cavity noise with low costs and minimal weight increase. Like ANC systems for ICE powertrains, they make use of the audio system in the vehicle. In this paper, a novel low-cost system for road induced tire cavity noise control (RTNC) is presented that reduces the tire cavity resonance noise inside a car cabin.

Let’s start with the uncomfortable truth, climate change is happening, and the automotive industrial network is one of the main industries contributing to greenhouse gas emissions. SKF is an energy intensive business – directly using energy, mainly in the form of electricity and gas, in its operations around the world. In addition, SKF utilizes materials, predominantly steel, and services which can be energy and carbon intensive – such as transports and raw material in production and processing. The combined impact of these direct and indirect energy uses (scope 1, 2 and 3 upstream) generates an excess of over two million metric tons of CO2e per year. This figure would however be significantly higher were it not for the actions SKF has taken to reduce both energy and carbon intensity. In 2000, we were one of the first companies to actually start to report and set climate targets.