Mobility: Human-Machine Interaction
Mobilität: Mensch-Maschine Interaktion
Duration: 2018 - 2022
Funding by: Federal Ministry of Education and Research Germany (BMBF)
Contact: Kristin Mühl, Martin Baumann
The overall goal of the project is the development of regional strategic cooperations between science, economy and society in order to drive innovations in the context of energy, mobility, health and biotechnology and transformation management. The focus of Ulm University in this project is the evaluation of a sensor system to assess reliable driver state data. These data can serve as basis to develop adaptive human-machine interaction concepts that enhance the acceptance and trust in automated driving. The transfer from theory into practice is facilitated by providing assessed data and gained knowledge to economy and society.
For more information visit the project website
AutoMate - Automation as accepted and trustful teamMate to enhance traffic safety and efficiency
Funding by: European Commission – Horizon 2020 research program
Cooperation Partners: Ulm University (Prof. Dietmayer), Offis, Broadbit Energy Technologies sro, Centro Ricerche FIAT, Continental Automotive France, Deutsches Zentrum für Luft- und Raumfahrt, Humatects, PSA, Re:Lab, Fondation Partenarial Moveotec
Contact: Juergen Pichen, Fei Yan
AutoMate is a project focusing on the development of a „TeamMate Car“ that combines the advantages from a highly automated driving vehicle with the abilities of a human being. The highly automated car cooperates with the driver in critical situations where the automation reaches its limitations. Our department evaluates this cooperation and improves the human-vehicle interaction by taking new situation comprehension concepts into account.
Further information: http://www.automate-project.eu/
Cooperative driver-vehicle interaction (KoFFI): safe, efficient and controllable interactions with automated vehicles
Development and evaluation of requirements, interaction concepts and context models for adaptive user interfaces
Kooperative Fahrer-Fahrzeug-Interaktion (KoFFI): Sichere, effiziente und kontrollierbare Interaktion mit autonomen Fahrzeugen
Funding by: Federal Ministry of Education and Research Germany (BMBF)
Cooperation Partners: Ulm University (Prof. Dr. Michael Weber) Robert Bosch GmbH, Daimler AG, EML European Media Laboratory GmbH, Hochschule Heilbronn (Prof. Dr. Gerrit Meixner), Hochschule der Medien (Prof. Dr. Petra Grimm, Prof. Dr. Tobias Keber
Contact: Marcel Woide, Kristin Muehl
The aim of the project is the development and demonstration of a holistic concept of driver-vehicle interaction. The driver and the automated vehicle will act like team players and work together in overcoming the ongoing challenges. It includes the cooperative realization of the handover of control in order to achieve a common aim in the best way. Ulm University focusses within that project on the development and evaluation of cooperative and adaptive interaction concepts as well as underlying requirements and context models.
Further information: https://www.technik-zum-menschen-bringen.de/projekte/koffi
Cooperatively Interacting Vehicles
Devoping and empirical testing of a comprehension based approach for driver-vehicle -cooperation in context of cooperatively interactive automobiles
Entwicklung und empirische Prüfung eines verstehensbasierten Modells der Fahrer-Fahrzeug-Kooperation für kooperativ interagierende Automobile
Funding by: German Research Foundation (DFG), Projektantrag im SPP 1835 „Kooperativ interagierende Automobile“
Contact: Tanja Stoll
Highly cooperative and automated vehicles are not just able to perceive the environment, classify and asses situations, select and execute actions but also to share information with other traffic users. Consequently, these vehicles might obtain further or additional information that a human driver could not achieve, which leads to new challenges for the interaction between drivers and their cars as well as among various drivers. First of all, it is central to find a solution how to avoid the lack of the system´s transparency status (e.g. a vehicle‘s environment model) and the absence of predictable vehicle behaviour. Therefore research in this projects focuses on modelling a comprehension based approach for driver-vehicle-cooperation in cooperatively interactive automobiles.
Further Information: http://www.coincar.de
Explanation of Adaptive Business Processes
Post Graduate School "Cognitive Computing in Socio-Technical Systems" of Ulm University and University of Applied Sciences Ulm.
Funding by: Minstry for Science, Research and Arts of the State of Baden-Württemberg
Cooperation Partners: Faculty of Computer Science, University of Applied Sciences Ulm (Prof. Dr. Christian Schlegel)
Contact: Christine Schnepf
Complex business processes, which may have to be dynamically adapted to changing circumstances during their enactment (as required in the context of Industry 4.0 processes), are neither easy to comprehend nor to trace. As an example, consider an order management process as required in the context of the production of customized or individualized goods. This project will empirically investigate both the psychological processes and structures underlying the understanding and anticipation of changing business processes. Furthermore, P5 will investigate how the need of respective process changes can be diagnosed and corresponding decisions be made. In particular, knowledge about these psychological processes will allow for the development of user-friendly algorithms and man-machine interfaces properly assisting end-users in changing business processes. In order to be able to reliably assess the current status of a business process, explanatory mechanisms will be provided. More precisely, based on formal process descriptions and their changes as well as the current process trace, simple inference procedures will be applied to create respective explanations and to present them in a user-friendly manner using different modalities.
Further information: here
Using persuasive technologies in highly automated driving to increase cooperation and driving safety
Persuasive Technologien in hochautomatisierten Fahrzeugen zur Erhöhung der Kooperation und Fahrsicherheit
Funding by: Carl Zeiss Foundation
Cooperation Partners: Institute of Media Informatics, Ulm University (Prof. Dr. Enrico Rukzio)
Contact: Philipp Hock
Highly automated driving aims to radically increase driving safety because human errors account for over 90 percent of severe traffic accidents. Even when automation is evolved enough to drive safer than human beings, it is assumed that the human driver could still request the control over the vehicle. Reasons for this may be a lack of trust, misunderstanding or a disagreement between the own driving style and the vehicle's driving style. In such cases, convincing the driver to keep the automation enabled increases traffic safety. Therefore, the aim of this research project is investigating technical systems to increase the usage of automation.
F3 Driver - Vehicle - Research
F3 brings together researchers from Ulm University who are all driven by the vision of future automated cars which move within an intelligent traffic infrastructure cooperatively with other automated and human-controlled vehicles. Those vehicles have to provide the highest degree of safety for own passengers and other traffic participants and at the same time contribute to global traffic efficiency.
F3 combines the competencies of participating researchers from the areas of advanced driver assistance, automated driving, and cooperative driving functions.
Driven by Trust
Contact: Johannes Kraus
In the process of automated driving becoming more and more part of every day’s reality, the role of the automated vehicle changes from a substitute for the human to a team player sharing the driving tasks on equal terms with the human driver.
Hereby, calibrated trust builds the basis for efficient and safe interaction with highly automated vehicles. In order to prevent both mis- and disuse of driving automation, this research project investigates the antecedents and process dynamics, in which trust towards a specific automated system is established. A process model integrating influencing factors for both, the driving automation and the human operator is developed, empirically tested and used to predict trust dynamics over time and in the face of system failures. Findings will provide insights for the design of driving automation and the associated interfaces.
Anticipation in Dynamic Driving Situations
Contact: Kristin Muehl
Anticipatory driving is an essential precondition for increasing the safety in traffic and avoiding conflicts and accidents. The anticipation of future behavior of the own vehicle and of other traffic participants enables an optimal preparation of the upcoming situation by increasing the time and space of action. In order to support drivers in anticipating relevant events, knowledge about the underlying cognitive processes of anticipation in dynamic traffic situations is required. Previous research predominantly focusses on visual cues which automatically activate knowledge structures and prior experiences of the long-term memory and trigger the prediction of prospective actions. Dynamic changes of the environment or motion extrapolation are not considered in previous theoretical assumptions. Therefore, a cognitive model of anticipation based on situation comprehension is going to be developed and evaluated. Pivotal findings will be the basis for designing anticipatory assistance systems.
Developing a Model-Based Lane Change Decision Aid System by Integrating Driver Uncertainty
Contact: Fei Yan
Driver uncertainty about the current lane change situation can substantially prolong the decision-making process, potentially leading to dangerous lane change maneuvers. In addition, the assistance that is not adaptive to drivers’ uncertainty states may decrease driver trust in assistance systems and further result in the disuse of such systems. Aiming to develop a Model-Based Lane Change Decision Aid System (MBLCDAS) integrating driver uncertainty during decision-making, driver uncertainty has been studied in a driving simulator for specific lane change scenarios on two-lane motorways. Then based on the empirical data, a probabilistic model of driver uncertainty that can classify diver’s uncertainty states during lane change situations as either “certain” or “uncertain” has been developed. After implementing the model of driver uncertainty and the corresponding HMI in the driving simulator, the developed MBLCDAS has been then evaluated.