Human system patterns for interaction and cooperation of automated vehicles and humans

Abstract

Abstract As the first automated driving functions are now finding their way into serial production vehicles, the focus of research and development has shifted from purely automated capabilities to cooperative systems, i.e. cooperation between vehicles, and vehicle automation with drivers. Especially in partially and highly automated cooperative driving the driver should be able to take over the driving task or adapt the driving behavior. This paper presents the pattern approach to cooperation as a method to recognize and solve reoccurring problems. As an example, the pattern approach is applied to the use case of a takeover request on a highway. The concept of Confidence Horizons, which balance the capabilities of the driver and the automation based on cooperative interaction patterns, is introduced. To estimate the human capabilities for this Confidence Horizon, a Diagnostic Takeover Request is used, in which the automation analyzes the driver’s orientation reaction to a takeover request. This allows the early detection of potentially unsafe takeovers reducing possible transitions to a Minimum Risk Maneuver (MRM).