Language acquisition is a crucial challenge in symbol emergence robotics. While cognitive science seeks to uncover mechanisms of language acquisition through human observation, symbol emergence robotics aims to elucidate the process by constructing mathematical models that replicate phenomena and validate their plausibility. This approach, which aims to clarify phenomena by constructing models, is called the constructive approach and strives to unravel the complex mechanisms underlying human language acquisition by developing mathematical models. Since word meanings are acquired through interactions with the environment, they are closely tied to sensory and motor experiences. In symbol emergence robotics, robots as embodied agents engage with their environments to acquire language, offering insights into the integration of multimodal information.