Heinrich Gotzig

Correspondence between variational methods and Hidden Markov Models

This paper establishes a duality between the calculus of variations, an increasingly common method for trajectory planning, and Hidden Markov Models (HMMs), a common probabilistic graphical model with applications in artificial intelligence and machine learning. This duality allows findings from each field to be applied to the other, namely providing an efficient and robust global optimization tool and machine learning algorithms for variational problems, and fast local solution methods for large state-space HMMs.

Evaluation of an analytic model for car dynamics

This paper presents and evaluates a differential geometric model for single and double track vehicles, which allows to estimate internal state parameters (steering angles, velocities, wheel speeds, ...) from a given planar trajectory over time. Possible applications include accident reconstruction from video footage, constraining trajectory optimization by physical limits, and extracting control parameters a-priori from trajectories, for use in open-loop (non-feedback) controllers for short intervals. The model is easy to compute, yet the evaluation with IPG CarMaker suggests a good level of accuracy.

A tractable interaction model for trajectory planning in automated driving

This paper presents an efficient model for combining automotive trajectory planning with predicted environment interactions, named progressively interacting trajectories (PITRA). It allows to plan trajectories for fully-automated vehicles by actively considering how other traffic participants will react to the trajectory, while retaining many of the advantages of variational trajectory optimization methods, in particular expressiveness and ease of computation. This enables maneuvers such as proactively claiming a gap during a lane change in dense traffic, which is impossible to model in classical variational models. The PITRA approach does not rely on a specific prediction model, but can be used in combination with a wide range of existing models. Its model assumptions and limitations are derived theoretically and demonstrated in several realistic scenarios.

Lightweight, non-invasive collection of steering wheel angles and pedal positions

This paper presents a unit to measure steering wheel angles and pedal positions in cars where internal sensors cannot be installed or accessed. It provides the necessary information to build a version of the unit, to customize it for a given application and to understand its behavior. Circuit board layouts and CAD files for 3D printing are also provided online. The paper is mainly aimed at research institutions needing to collect this data in proving ground scenarios at a limited budged without requiring access to manufacturers IP. The measurement uncertainty of steering wheel angles is shown to be in the order of 2 ° at a rate of about 5 Hz, but can be improved in individual applications.

The Next Generation Park Assist Systems

Fundamental principles of park assist systems have not changed significantly since the first generation: They inform the driver about distances between close objects during the parking manoeuvre. Valeo has played the role of an outrider and introduced the first parking assistance system based on ultrasonic technology in 1991 in the BMW 7-Series. In the following the development steps, market studies, customer wishes for future parking systems and their technological realisation are being described.

Die nächste Generation Parkassistenzsysteme

An der Grundfunktionalitat von Parkhilfesystemen — der Information des Fahrers uber Abstande naher Objekte wahrend des Einparkvorganges — hat sich seit der ersten Generation nur wenig geandert. Valeo war Vorreiter auf diesem Gebiet und hat Anfang der 1990er-Jahre das erste auf Ultraschall basierende Parkassistenzsystem im BMW 7er eingefuhrt. Im Folgenden werden Entwicklungsstufen, Marktstudien, Wunsche der Endkunden an kunftige Parksysteme und deren technologische Umsetzung beschrieben.