Automatisiertes Fahren 2019

The validation and verification of level 4-5 highly automated driving (HAD) functions requires a vast number of test kilometers. If real test drives are applied exclusively, it can be safely assumed that the required test scope cannot be covered within feasible limits of time and money [Winner]. This is not only due to the functional complexity of HAD applications, but also due to short development cycles and legal safety requirements.

Architecture Trend: Architectures move from Distributed to Centralized to Service-Oriented Architectures Complexity: Adoption of new automotive trends (e.g. zones & servers) helps to cope with complexity Process: Champion best practices in Systems Engineering Development enhanced by dedicated system architecture views.

Bringing automated driving functions and systems on the road is a highly complex task. We expect SW with >20 million lines of code per high performance computer and a significant increase in complexity especially for HAD applications. From the point of view of a software Tier 1 company, there are major game changers when it comes to developing such functions.

A large number of sensors and actuators, combination of high-performance computer and safety critical applications, complex data flows with a huge amount of data and many Tier-2 software suppliers are fundamental challenges for the development of autonomous driving systems. To handle the technical and organizational complexity the most proven development methods and tools are just good enough. Many customer projects have shown focussing early on timing is a key for success. Using state-of-the-art simulation and worst-case analysis tools for evaluation of real-time requirements and design optimization does massively reduce project costs and risks.

PCIe Interface will be the standard interface in data acquisition Ethernet will be the standard for the in-vehicle network architecture Data amount will increase due to amount and resolution of sensors.

It has been demonstrated that driver assistance systems are making a significant contribution toward increasing road safety. For more than a decade, the Accident Research Department of the Allianz Center for Technology (AZT) has been analyzing impact potentials as part of several research projects, including AKTIV, TRACE and euroFOT [1, 2, 3], and was largely responsible for the fact that effective driver assistance systems are now taken into account in several countries when pricing insurance premiums. Potential effectiveness is regularly assessed on the basis of third-party liability (TPL) and motor own damage (MoD) claims reported to Allianz Versicherung [4, 5, 6, 7].

As todays modern state of the art vehicles are underlying incremental changes with regard especially to ‘Automated Driving’ (e.g. sensor suite, control units, processing units), ‘Connectivity’ (e.g. software updates, over-the-air, infotainment), ‘Electrification’ (e.g. battery cells, drive train), and ‘Shared Mobility’ (e.g. changing customers behaviour, changing markets, different usage of the car), vehicle data gains in importance. Thus, this article focuses on data and data sets generated during the usage of vehicle fleets in general and individual vehicles in particular. In addition to the general relevance of vehicle data for several stakeholders and interested parties, further complexity arises due to the development of vehicles from driver operated mobility devices towards system enhanced automated driving systems (according to the five level SAE approach).

Autonom fahrende Autos erobern mehr und mehr unseren Alltag. Bereits jetzt sind viele Fahrzeuge mit Assistenzsystemen des Level 1 ausgestattet. Diese unterstützen den Fahrer, indem sie den Abstand zum vorausfahrenden Fahrzeug regeln, die Geschwindigkeit halten oder beim Spurwechsel vor einem Fahrzeug im toten Winkel warnen.

The progressive technical development in the field of advanced driving assistance systems (ADAS) and automated driving makes it feasible that motor vehicles with a high level of automation will be ready for market within the next decade. The vision of automated vehicles on public roads not only depends on technical progress but also requires an adequate liability framework to ensure legal certainty for the relevant stakeholders on both the demand and supply side.

Deep Learning and AI has become the standard model for object detection and recognition such as situation understanding, prediction and planning. In this chapter we explore how AI can be used to improve parts of the classical ADAS algorithm chain. Based on this, we investigate the full Automated Driving pipeline. Firstly, we describe the building blocks of the pipeline composed of standard computer vision tasks. We provide an overview of use cases for automated driving based on the authors’ experience in commercial deployment, e.g. Sensor-Fusion, Perception, SLAM or End-2-End Driving. Finally, we discuss the opportunities of using AI and Deep Learning to improve upon state-of-the-art classical methods.

Making Automated Driving (AD) a reality is a challenging task and still subject of intensive research and development. The core component for realizing AD beside the actual vehicle platform and the sensors for perceiving the environment is the AD system, i.e., the software that enables a vehicle to perform distinct dynamic driving tasks in a distinct operational design domain. Briefly speaking, an operational design domain defines environmental and time-of-delay conditions and restrictions, respectively, the presence or absence of certain traffic, and roadway characteristics.

In recent years, artificial intelligence has increasingly become the focus of public attention. The automotive industry can profit greatly from this – especially in the rapid development of autonomous driving functions and driver assistance systems. In addition to established OEMs, newcomers to the industry are also investing large sums of money and are putting enormous effort into advancing this technology.

It is now generally accepted that it is – to a reasonable extent – impossible to test and validate automated driving functions exclusively with the help of real driving tests. It certainly makes sense to feed measurement data from the field test into an SIL environment in order to test new software versions. However, the usable amount of data is by far not sufficient for a statistical proof of safety [1].

Driver assistance and automated driving functions are already changing driving as we know it, and not simply by removing the need for human interaction. They are also making driving more enjoyable – and vastly safer. Thanks to significant improvements in software, hardware, and especially Artificial Intelligence (AI), cars will soon be able to think and make life-saving decisions completely by themselves.

Current test and validation procedures for automated and connected driving functions use field tests on public roads primarily to identify unknown critical scenarios. These scenarios are then validated or varied within simulations or on closed proving grounds. In order to validate complex urban driving scenarios, this approach may need to be supplemented by the method of scenario-based field tests in real traffic environments. Scenario-based tests constitutes the methodical basis for an efficient and purposive test of automated and connected driving functions in public testbeds.

This paper focuses on concepts for the design, planning and implementation of scenario-based tests of automated and connected driving functions under real traffic conditions. The developed toolbox will be explained using the example of the “Digital Testbed Dresden / Saxony” as part of the initiative “Synchrone Mobilität 2023”.

To achieve the goal of an autonomous driving vehicle, more and more sensors are being integrated in the vehicle. For the last generations it was sufficient, that those sensors did send object data, which was then fused into an environment model. However, in order to have a more accurate model and to be able to navigate autonomously, the raw sensor data is needed.

Automated driving is a key technology for the future of transportation. There are several motivations to develop automated vehicles. First and foremost, it promises to reduce the number of traffic accidents. Figure 1 shows the accidents recorded by the German police over the past years ([1]) ranging back to 1960.

Today the level of confidence in many means of transport such as the car, bus, train, and airplane is in general very high, mainly due to the high reliability and safety, which is perceived during using those means of transport. Behind this are continuous improvements and advances in vehicle development and manufacturing, improved infrastructure, and ongoing adjustments of the legal framework worldwide.

In partly automated driving the driver has to both monitor the system and be able to take over driving immediately if necessary (SAE Level 2: L2) [SAE16]. The optimal monitoring performance is reached if the driver is neither under- nor overloaded in terms of strain [DeW96]. Advanced driver assistance systems may assist the driver in remaining in this optimal load during potentially long, monotonous automated drives. This is evaluated in a driver simulator study. The study assessed the impact of several secondary tasks (ST), naturalistic as well as standardized, during partly automated driving. 34 participants went through eight test conditions: Two baseline drives (manual driving (L0) and L2 without secondary activity) and six conditions with a L2 automation and different secondary activities. These activities included an auditory n-back task (1- and 2-back) (e.g. [Lor15]) and the surrogate reference task (SuRT) [ISO14198]. Furthermore, an activating task (stretching exercises) as well as a condition in which a video was played were integrated. Subjective strain was measured by using questionnaires (Subjective Experienced Stress: SEA; Stanford Sleepiness Scale: SSS), objective strain in terms of monitoring performance by using a detection task. The results suggest that visual secondary tasks lead to a decrease in drivers’ monitoring performance – not necessarily reflecting the drivers’ perceived strain. For example, subjectively demanding auditory tasks did not lead to substantial monitoring lapses but may induce positive effects in potentially monotonous automated driving situations.

Safety in automated vehicles is related to how the driver and the automation system cooperate in the driving task [1]. Intuitive and consistent operation of the vehicle up to SAE automation level 4 through elaborated human-machine interaction is a key enabler for safety, comfort and market acceptance of driving automation [2]. Sharing manual and automated lateral control provides a new driving experience where the driver can steer the vehicle over the automation system without deactivation.