Marcel Baltzer

Uncanny and Unsafe Valley of Assistance and Automation: First Sketch and Application to Vehicle Automation

Progress in sensors, computer power and increasing connectivity allow to build and operate more and more powerful assistance and automation systems, e.g. in aviation, cars and manufacturing. Besides many benefits, new problems occur e.g. in human-machine-interaction. In the field of automation, e.g. vehicle automation, a comparable, metaphorical design correlation is implied, an unsafe valley e.g. between partially- and highly-automated automation levels, in which due to misperceptions a loss of safety could occur. This contribution sketches the concept of the (uncanny and) unsafe valley of automation, summarizes early affirmative studies, gives first hints towards an explanation of the valley, outlines the design space how to secure the borders of the valley, and how to bridge the valley.

Joint Decision Making and Cooperative Driver-Vehicle Interaction during Critical Driving Situations

Abstract In automated driving, the human driver and an automation form a joint human-machine system. In this system, each partner has her own individual cognitive as well as perceptual processes, which enable them to perform the complex task of driving. On different layers of the driving task, both, drivers and automation systems, assess the situation and derive action decisions. Although the processes can be divided between human and machine, and are sometimes very elaborate, the outcome should be a joint one because it affects the entire driver-vehicle system. In this paper, the individual processes for decision-making are defined and a framework for joint decision-making is proposed. Joint decision-making relies on common goals and norms of the two subsystems, human and automation, and evolves with experience.

Behaviour Adaptation Using Interaction Patterns with Augmented Reality Elements

This publication describes a systematic approach for behaviour adaptations of humans, based on interaction patterns as a fundamental way to design and describe human machine interaction, and on image schemas as the basic elements of the resulting interaction. The natural learning path since childhood involves getting knowledge by experience; it is during this process that image schemas are built. The approach described in this paper was developed in close interplay with the concepts of cooperative guidance and control (CGC), where a cooperative automation and a human control a machine together, and of augmented reality (AR), where a natural representation of the world, e.g. in form of a video stream, is enriched with dynamic symbology. The concept was instantiated as interaction patterns “longitudinal and lateral collision avoidance”, implemented in a fix based simulator, and tested with professional operators whether driving performance and safety in a vehicle with restricted vision could be improved. Furthermore, it was tested whether interaction patterns could be used to adapt the current driver behaviour towards better performance while reducing the task load. Using interaction patterns that escalated according to the drivers actions and the current environmental state, lead to a reduction of temporal demand, effort and frustration. Furthermore less collisions were counted and the overall lateral displacement of the vehicle was reduced. The results were a good mix of encouragement and lessons learned, both for the methodical approach of pattern based human machine interaction, and for the application of AR-based cooperative guidance and control.

Vision and Driving Support for Shielded Vehicles – Implementation and Test of an Electronic Vision Replacement System with Augmented Reality

The compromises encountered in the design of shielded and armored vehicles regarding the trade-offs between outside visibility and crew protection often result in the decrease of situational awareness. Drivers using busy roads face the challenge of maneuvering large vehicles, which, due to their safety requirements, have large blind spots that make it difficult to detect elements that could pose a threat to the safety of the crew and the vehicle. This paper presents the implementation of a camera-monitor system used as an electronic vision replacement and its evaluation results from a test trial with soldiers from the German Bundeswehr. Additionally, elements of Augmented Reality were incorporated to further assist the driver while navigating through the simulated urban scenario.

Das (unheimliche und) unsichere Tal der Assistenz und Automation - Beschreibung und Absicherungsmöglichkeiten

Zusammenfassung Fortschritte in der Automatisierungstechnik ermöglichen es technischen Systemen immer mehr Aufgaben zu übernehmen. Für die Gestaltung der Interaktion und Kooperation von Mensch und Maschine ist die Einschätzung des Menschen bzgl. der Fähigkeiten der Maschine sicherheitsrelevant. Das Uncanny Valley der Robotik beschreibt, wie Roboter mit hoher, aber nicht perfekter Menschenähnlichkeit als unheimlich wahrgenommen werden. Bestehende Studien, z. B. für Luft- und Bodenfahrzeuge, deuten an, dass es einen ähnlichen Zusammenhang zwischen Automationsgrad und Sicherheit geben könnte, da sich zwischen gut funktionierenden teil- und hochautomatisierten Modi ein unsicheres Tal bildet, in welcher die Sicherheit des Systems stark reduziert ist. Es werden Gestaltungsoptionen beschrieben, um diesen Bereich abzusichern.

H-Mode 2D

Vor dem Hintergrund wachsender technischer Moglichkeiten im Bereich der Assistenz und Automation entstehen vielfaltige Herausforderungen, Risiken und Chancen in der Gestaltung des assistierten, teil- und hochautomatisierten Fahrens. Eine der grosten Herausforderungen besteht darin, eine Vielzahl von komplexen technischen Funktionen so zu integrieren und dem Menschen anzubieten, dass sie intuitiv als ein zusammenhangendes, mit dem Fahrer kooperierendes System verstanden und jederzeit zuverlassig, sicher und angenehm bedient werden konnen. Dabei verschwimmen die Grenzen zwischen Assistenz und Automation zunehmend und es wird notwendig, einander erganzende Assistenz- und Automationsgrade zu definieren [1]. Somit ist es sinnvoll, einen starkeren Fokus auf die Einbeziehung des Menschen im Sinne einer kognitiven Kompatibilitat und im Hinblick auf das Vertrauen zwischen Mensch und Automation bzw. Assistenz (vgl. [2, 3] und Kap. 58) sowie auch dem Menschen im Entwicklungsprozess zu legen [4].