[2M6-GS-13] AI application: Artificial General Intelligence

[2M6-GS-13-01] Probabilistic Model of Spatial Concepts Integrating Generative Adversarial Networks for Semantic Mapping

〇Yuki Katsumata¹, Akira Taniguchi¹, Lotfi El Hafi¹, Yoshinobu Hagiwara¹, Tadahiro Taniguchi¹ (1. Ritsumeikan university)

Keywords:semantic mapping, symbol emergence in robotics, generative adversarial networks

This paper proposes SpCoMapGAN, a method to generate the semantic map in a newly visited environment by training an inference model using previously estimated semantic maps. SpCoMapGAN uses generative adversarial networks (GANs) to transfer semantic information based on room arrangements to the newly visited environment. We experimentally show in simulation that SpCoMapGAN can use global information for estimating the semantic map and is superior to previous related methods.