Abstract: In order to enrich the cognitive ability of artificial fish in virtual marine environment, this paper proposes a memory-based cognitive model of artificial fish according to the biological principle and memory mechanism of natural fish. Information is transferred and stored during the three stages: ultra short-term memory(USTM), short-term memory(STM) and long-term memory(LTM). Artificial fish uses the focuser to extract the interested sensory information, and uses the decision-maker and STM to select behaviors and stores the experience knowledge to LTM which is called by STM to optimize current decision-making. Animation results show that artificial fish with this model can have memory-directed behaviors and display more vivid and intelligent life properties.

Key words: artificial fish, memory, cognitive model, focus, decision-making

摘要： 为丰富虚拟海洋环境中人工鱼的认知能力，以自然鱼的生物原理和记忆机制为理论依据，提出了一种基于记忆的人工鱼认知模型。信息在瞬时记忆、短时记忆和长时记忆3个阶段传输和存储，人工鱼通过聚焦器提取所关注的感知信息，通过决策器和短时记忆进行行为选择，将经验知识存储在长时记忆中，被短时记忆调用以优化当前决策。动画结果表明，采用此模型后人工鱼能产生记忆指导下的行为，表现出更真实更智能的生命特征。

关键词: 人工鱼, 记忆, 认知模型, 聚焦, 决策

CLC Number: 

• TP18