RoboCrab: An integrative approach to the natural ecology of decision making

RoboCrab：决策自然生态的综合方法

• 批准号: DP160102658
• 负责人: A/Prof Jan Hemmi
• 金额: $ 30.62万
• 依托单位: The University of Western Australia
• 依托单位国家: 澳大利亚
• 项目类别: Discovery Projects
• 财政年份: 2016
• 资助国家: 澳大利亚
• 起止时间: 2016-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/DP160102658
• 关键词: RoboCrab; integrative; approach; natural; ecology

The project aims to analyse and model the sophisticated and context-dependent escape behaviour of fiddler crabs under both natural conditions and in controlled laboratory settings. A crucial problem for biology is to understand how animals can make adaptive decisions in natural, complex sensory environments; such understanding also has direct application to robotics. The project plans to examine the effects of eye stabilisation and oscillation, record from key neural stages using naturalistic stimuli to derive precise algorithms, and integrate and test the results on a robot model – RoboCrab. This may provide new insight into the integration of low-level sensory input with behavioural decision making circuits and the evolution of escape behaviours.

该项目旨在分析和模拟招潮蟹在自然条件下和受控实验室环境中复杂的和依赖于环境的逃逸行为。生物学的一个关键问题是理解动物如何在自然的、复杂的感官环境中做出适应性决策;这种理解也直接应用于机器人技术。该项目计划检查眼睛稳定和振荡的影响，使用自然刺激从关键神经阶段记录以获得精确的算法，并在机器人模型RoboCrab上整合和测试结果。这可能为低水平感觉输入与行为决策回路的整合以及逃避行为的进化提供新的见解。