Using virtual reality, modelling, control, machine & deep learning to investigate behavioural dynamics of human machine interaction from a multiagent

使用虚拟现实、建模、控制、机器

• 批准号: 2296793
• 金额: --
• 依托单位: University of Bristol
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2018
• 资助国家: 英国
• 起止时间: 2018 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2296793
• 关键词: Using; virtual; reality; modelling; control

Starting point will be to identify cognitive and behavioural processes of groups of humans performing cooperative perceptual motor tasks, with the use of mathematical modelling, machine and deep learning. The project will also seek to unfold the degree to which mechanism lead to the emergence of robust multi-agent human coordination in complex multi-agent-environment tasks in general. The aim will be to design artificial agents able to perform joint herding and crowd control tasks in cooperation with humans and use observations from experiments to design feedback control algorithms and cognitive architectures able to drive such agents

起点将是通过使用数学建模，机器和深度学习来确定执行合作感知运动任务的人群的认知和行为过程。该项目还将寻求展开哪种机制导致在复杂的多代理环境任务中出现强大的多代理人协调的程度。目的是设计能够与人类合作执行共同放牧和人群控制任务的人造代理，并利用实验的观察来设计反馈控制算法和能够驱动此类代理的认知体系结构