Franz Winkler

Lateral trajectory tracking control for autonomous vehicles

In this contribution a structure for high level lateral vehicle tracking control is presented. It is based on the two degrees of freedom structure that allows to separately define the command response and disturbance attenuation. The application of the disturbance observer guarantees robust compensation of the disturbances. An advantage of the presented structure is its robustness against variable vehicle parameters. Only a considerably reduced model is necessary which significantly simplifies the application process. Moreover the presented approach is characterized by its extensibility and its modularity.

Kollisionsfreie Längs- und Quertrajektorienplanung unter Berücksichtigung fahrzeugspezifischer Potenziale

Zusammenfassung In diesem Beitrag wird ein Planungsansatz für teil- und hochautomatisierte Fahrfunktionen vorgestellt. Die Planungsaufgabe wird dabei von zwei Teilsystemen durchgeführt. Eine semi-globale Grobplanung gibt den Zielbereich vor, während eine lokale Trajektorienplanung den Verlauf der Trajektorie plant. Der vorgestellte Ansatz generiert Trajektorien, die Kollisionsfreiheit mit umliegenden Objekten gewährleisten. Gleichzeitig werden fahrdynamische Randbedingungen (wie ein begrenzter Reibwert) oder Aktuatorbegrenzungen berücksichtigt, sodass die geplante Trajektorie von der unterlagerten Regelung umsetzbar ist. Eine große Herausforderung in diesem Bereich stellt die Umsetzung auf Seriensteuergeräten dar, die in ihrer Rechenkapazität begrenzt sind.

Learning the “Language” of Road Users - How Shall a Self-driving Car Convey Its Intention to Cooperate to Other Human Drivers?

Communication between road users is ruled by road traffic regulations, but there are also implicit laws of communication. Especially lane changes in dense traffic scenarios require not only communicating one’s intention but also cooperating with other drivers. Self-driving vehicles will need to communicate with conventional vehicles on the road during the transition period to full automation. But how does a driver show his willingness to cooperate? A driving simulator study with N = 28 drivers in a dense traffic scenario on the highway was conducted. It was assumed that different lag vehicle reaction behavior on turn signals of the ego driver would influence the ego driver in his subjective evaluation of the situation. Three main effects, deceleration, the amount of velocity reduction and reaction time concerning perceived cooperation were found. The results of the study can be used to design cooperative driving strategies between self-driving and manually driven vehicles.

What Makes an Automated Vehicle a Good Driver

An automated vehicle needs to learn how human road users experience the intentions of other drivers and understand how they communicate with each other in order to avoid misunderstandings and prevent giving a negative external image during interactions. The aim of the present study is to identify a cooperative lane change indication which other drivers understand unambiguously and prefer when it comes to lane change announcements in a dense traffic situation on the highway. A fixed-base driving simulator study is conducted with N = 66 participants in Germany in a car-following scenario. Participants rated, from the lag drivers perspective, different lane change announcements of another driver which varied in lateral movements (i.e., duration, lateral offset). Main findings indicate that a medium offset and moderate duration of lateral movement is experienced as most cooperative. The results are crucial for the development of lane change strategies for automated vehicles.

Maneuver planning for highly automated vehicles

One important aspect of autonomous driving lies in the selection of maneuver sequences. Here the challenge is to optimize the driving comfort and travel-duration, while always keeping within the safety limits. Human drivers analyze and try to anticipate the traffic situation choosing their actions not only based on current information but also based on experience. The decision making process can be treated as a planning problem. Classical planning systems consider the autonomous driving task as a global numeric optimization problem, which in populated dynamic environments can become computationally intractable. In addition, purely numeric computations hamper the understanding of the decision making for the human user. We propose a planning system that presents a multi-level architecture, similar to the human reasoning process, which combines continuous planning with semantic information. This allows the planning system to deal with the complexity of the problem in a computationally efficient way and also provides an intuitive interface to communicate the decisions to the driver. We validate our approach in simulation and through a set of experiments carried out with a real vehicle and an integrated traffic simulation also known as vehicle in the loop (VIL).