Martin Haueis

Making Bertha Drive?An Autonomous Journey on a Historic Route

125 years after Bertha Benz completed the first overland journey in automotive history, the Mercedes Benz S-Class S 500 INTELLIGENT DRIVE followed the same route from Mannheim to Pforzheim, Germany, in fully autonomous manner. The autonomous vehicle was equipped with close-to-production sensor hardware and relied solely on vision and radar sensors in combination with accurate digital maps to obtain a comprehensive understanding of complex traffic situations. The historic Bertha Benz Memorial Route is particularly challenging for autonomous driving. The course taken by the autonomous vehicle had a length of 103 km and covered rural roads, 23 small villages and major cities (e.g. downtown Mannheim and Heidelberg). The route posed a large variety of difficult traffic scenarios including intersections with and without traffic lights, roundabouts, and narrow passages with oncoming traffic. This paper gives an overview of the autonomous vehicle and presents details on vision and radar-based perception, digital road maps and video-based self-localization, as well as motion planning in complex urban scenarios.

Video based localization for Bertha

In August 2013, the modified Mercedes-Benz SClass S500 Intelligent Drive (“Bertha”) completed the historic Bertha-Benz-Memorial-Route fully autonomously. The self-driving 103 km journey passed through urban and rural areas. The system used detailed geometric maps to supplement its online perception systems. A map based approach is only feasible if a precise, map relative localization is provided. The purpose of this paper is to give a survey on this corner stone of the system architecture. Two supplementary vision based localization methods have been developed. One of them is based on the detection of lane markings and similar road elements, the other exploits descriptors for point shaped features. A final filter step combines both estimates while handling out-of-sequence measurements correctly.

System Monitoring for Lifetime Prediction in Automotive Industry

System monitoring in automotive industry has multiple demanding facets. Increasing complexity of control tasks in modern automobiles has caused a shift in the way we think about sensor and actuator functions. Support systems for controlling a highly efficient engine, handling difficult driving conditions and realizing outstanding comfort functions are what usually is understood as system monitoring in vehicles. However, a new aspect is gaining significant importance: system monitoring for life time prediction. This paper outlines the fundamentals of system monitoring for life time prediction in vehicles. The vision of “Car Health Management” is introduced. The particular example of oil condition monitoring is chosen to outline the concept of predictive maintenance. Technological challenges encountered with this new philosophy are discussed. As an example an autonomously operated oil sensing system is presented.

Robust localization based on radar signal clustering

Significant advances have been achieved in mobile robot localization and mapping in dynamic environments, however these are mostly incapable of dealing with the physical properties of automotive radar sensors. In this paper we present an accurate and robust solution to this problem, by introducing a memory efficient cluster map representation. Our approach is validated by experiments that took place on a public parking space with pedestrians, moving cars, as well as different parking configurations to provide a challenging dynamic environment. The results prove its ability to reproducibly localize our vehicle within an error margin of below 1% with respect to ground truth using only point based radar targets. A decay process enables our map representation to support local updates.

Landmark based radar SLAM using graph optimization

On the way to achieving higher degrees of autonomy for vehicles in complicated, ever changing scenarios, the localization problem poses a very important role. Especially the Simultaneous Localization and Mapping (SLAM) problem has been studied greatly in the past. For an autonomous system in the real world, we present a very cost-efficient, robust and very precise localization approach based on GraphSLAM and graph optimization using radar sensors. We are able to prove on a dynamically changing parking lot layout that both mapping and localization accuracy are very high. To evaluate the performance of the mapping algorithm, a highly accurate ground truth map generated from a total station was used. Localization results are compared to a high precision DGPS/INS system. Utilizing these methods, we can show the strong performance of our algorithm.

Integrity for autonomous driving: A survey

Autonomous driving has become a focus for many universities and automotive companies alike, aiming for public introduction within the next few years. However, this requires assurance that the inherent risk in critical systems such as localization is sufficiently low. In aviation, the concept of integrity was established for this purpose. It relies upon a localization systems ability to provide timely and correct alerts notifying when the system must not be used. For autonomous vehicles, alert generation is still a topic of research because the environment is more complex. This work contributes by identifying two alert generation approaches applicable to autonomous driving and comparing their conceptual merits and drawbacks. The focus lies on methods for fault detection and isolation as well as those based on a bounded-error assumption. Both methods addressed here are guaranteed to either detect out-of-tolerance errors or determine a valid solution. However, they differ in how risk is dealt with.

Autonomes Fahren auf der historischen Bertha-Benz-Route

Zusammenfassung Im Jahre 1888 trat Bertha Benz die erste Überlandfahrt in der Geschichte des Automobils an. 125 Jahre später wiederholte die Mercedes Benz S-Klasse S 500 Intelligent Drive diese historische Fahrt von Mannheim nach Pforzheim – selbständig, ohne Fahrereingriff und im realen Verkehr. Die Bertha-Benz-Route ist 103 km lang und zeichnet sich durch eine breite Vielfalt von zu bewältigenden Fahrsituationen aus, die repräsentativ für den heutigen Alltagsverkehr sind. Die Strecke beinhaltet die Innenstädte von Mannheim und Heidelberg sowie die Durchfahrung von 23 Ortschaften und kleineren Städten. Zu den Situationen, die ein autonomes Fahrzeug auf der Bertha-Benz-Route beherrschen muss, gehören z. B. Kreisverkehre, Kreuzungen mit und ohne Ampelanlagen, Zebrastreifen, Überholen von Radfahrern oder enge Ortsdurchfahrten mit entgegenkommendem Verkehr. Eine Besonderheit des vorgestellten Projektes war die ausschließliche Verwendung seriennaher Sensorik. Kameras und Radarsensoren in Verbindung mit einer präzisen digitalen Karte ermöglichten die Erfassung des Fahrzeugumfelds auch in komplexen Situationen. Dieser Artikel liefert eine Systemübersicht des Fahrzeugs. Er beschreibt die kamerabasierte Umgebungswahrnehmung, die verwendeten digitalen Karten und die kartenrelative Selbstlokalisierung sowie die Manöverplanung in komplexen Verkehrsszenarien.

Gestaltung erfolgreicher Lernumgebungen in der manuellen Montage

Kurzfassung Die erfolgreiche Gestaltung von Lernumgebungen ist, insbesondere bei zunehmender Komplexität und immer kürzer werdenden Produktlebenszyklen, ein wichtiger Schlüssel für hohe Produktqualität. Unter Nutzung der Komponenten der Selbstbestimmungstheorie und einem sozio-technischen Regelkreismodell können kritische Parameter zum erfolgreichen Lernen im Arbeitskontext der manuellen Montage identifiziert werden. Im Ergebnis wurde die Designregel E3 für die erfolgreiche Gestaltung von Lernumgebungen entwickelt. Eine standardisierte Verwendung der E3-Matrix wird empfohlen.

Deriving HD maps for highly automated driving from vehicular probe data

High definition (HD) map data is a key feature to enable highly automated driving. With the advent of highly automated vehicles, car makers and map suppliers investigate new approaches to create and maintain HD maps by using on-board sensor data of series vehicles. While state-of-the-art-approaches focus on position and speed data analysis, the consideration of additional vehicle sensor data allows for novel approaches in the context of HD maps. By 2020, more than 30 million connected vehicles are expected to be sold per year, which will generate millions of terabytes of vehicular probe data. One of the major upcoming research issues is to find methods to exploit that probe data to generate and maintain HD maps. In this paper, we address how to develop such methods. We introduce a scalable infrastructure, which supports the ingestion, management and analysis of huge amounts of probe data. It supports an iterative process to develop, assess and tune methods for generating HD maps from probe data. We present a metric to assess methods regarding resulting map precision. As a proof of concept, we present an approach to derive road geometry of highways from location and sensor information.

Autonomous Sensor Systems for Car Applications

The development of new powertrain functions has many goals: more power and torque, low fuel consumption and emission, best driving performance and a brand specific acoustics. In the future the accuracy of control in Motor Control Units (MCU) will increase more and more. For this reason open loop control will replaced by closed loop control. New and/or other values are needed for better algorithms in powertrain control. Autonomous sensor systems for automobile industry offer increased reliability due to the reduction of cable connections, novel sensing applications made possible by wireless data transfer and cost advantages resulting from the aforementioned aspects. These systems consist of a power generator, energy storage and data transceiver. The technological challenge for realizing such system is most of all the construction and fabrication of a mini power generator that matches the dimensions of standard sensor modules. In this paper different power generators for autonomous sensor systems are reviewed and systematically discussed. It is shown that low power thermoelectric generators can supply autonomous sensor systems with up to 7 mW utilizing waste energy of the engine.