Comprehension and decision based modeling of cooperation behavior between drivers in dynamic traffic situations and its methodological foundations - CoMove

动态交通情况下驾驶员之间合作行为的基于理解和决策的建模及其方法基础 - CoMove

• 批准号: 272996369
• 负责人: Professor Dr. Martin Baumann
• 金额: --
• 依托单位: Institut für Psychologie und Pädagogik
• 依托单位国家: 德国
• 项目类别: Priority Programmes
• 财政年份: 2015
• 资助国家: 德国
• 起止时间: 2014-12-31 至 2023-12-31
• 项目状态: 已结题
• 来源: https://gepris.dfg.de/gepris/projekt/272996369?language=en
• 关键词: Comprehension; decision; based; modeling; cooperation

In the first phase of the Priority Program "Cooperatively interacting automobiles" the predecessor project focussed on the cognitive processes of situation comprehension that represent the basis for the initiation of cooperation between drivers in dynamic, potentially critical traffic situations. The goal was to create an empirically validated basis for the creation of a trustworthy, comprehensible, and acceptable automation on the basis of a deep understanding of these cognitive processes that determine how drivers identify conflicts in traffic and how they come to a solution for these conflicts. The major goals of the follow-up project CoMove are to investigate and model the course of action when drivers cooperate with each other in more complex traffic situations and the development and validation of new methods for describing and evaluating dynamic cooperation behaviour. Accordingly, the investigation of decision making processes in the course of communication and cooperation processes when several drivers coordinate their actions while cooperatively executing manoeuvers are now in the focus of both empirical investigations and modelling activities. Starting from results on accepted and non-accepted cooperative solution strategies gained in the predecessor project we will investigate how action coordination between drivers develops as a function of situational characteristics, which decisions drivers make as reaction to observed behavioural communication signals from other drivers, which actions drivers execute to signal willingness to cooperate and need for cooperation, how driver characteristics, such as personality and driver state, influence these processes, and with which methods these coordination processes can be described. We will also consider situations where the coordination between partners is not successful or optimal. The goal of these empirical investigations is to gain detailed knowledge about the coordination processes as a basis for the creation of action planning strategies of cooperatively interacting automobiles that are both comprehensible and predictable for the surrounding human drivers as they reflect familiar human behaviour and interaction strategies, and that allow automated vehicles to solve traffic conflicts with human drivers safely and efficiently.

在优先计划的第一阶段“合作互动的汽车”的前身项目集中在情况理解的认知过程，代表了在动态的，潜在的关键交通情况下，司机之间的合作启动的基础。我们的目标是在深入理解这些认知过程的基础上，创建一个经过经验验证的基础，以创建一个值得信赖的，可理解的和可接受的自动化系统，这些认知过程决定了驾驶员如何识别交通冲突，以及他们如何解决这些冲突。后续项目CoMove的主要目标是调查和模拟驾驶员在更复杂的交通情况下相互合作的行动过程，并制定和验证描述和评估动态合作行为的新方法。因此，调查的决策过程中的沟通和合作过程中，当几个司机协调他们的行动，同时合作执行机动的过程中，现在的重点是实证调查和建模活动。从先前项目中获得的接受和不接受的合作解决方案策略的结果开始，我们将研究驾驶员之间的行动协调如何发展为情境特征的函数，驾驶员对观察到的其他驾驶员的行为沟通信号做出反应，驾驶员执行哪些行动以表示合作意愿和合作需求，驾驶员特征如何，如个性和驾驶员状态，影响这些过程，以及用什么方法可以描述这些协调过程。我们还将考虑合作伙伴之间的协调不成功或不理想的情况。这些实证研究的目标是获得有关协调过程的详细知识，作为创建协作交互汽车的行动规划策略的基础，这些策略对于周围的人类驾驶员来说是可理解和可预测的，因为它们反映了熟悉的人类行为和交互策略，并且允许自动驾驶汽车安全有效地解决与人类驾驶员的交通冲突。