Download contents


  • Multimodal Data for Learning Spatial Concept
    The data sets were collected for learning spatial concept. The data sets contain the robot's position estimated by Monte Carlo Localization, and image information captured by a Web camera at each position.
  • Virtual Home Environments for Collecting Multimodal Data
    The virtual home environments made by Sweet Home 3D can be downloaded by the following link for collecting multimodal data

Open source software

  • Spatial Concept Formation
    This is a model to learn spatial concept from images and robot position information. Autonomous robots, such as service robots, operating in the human living environment with humans have to be able to perform various tasks and language communication. To this end, robots are required to acquire novel concepts and vocabulary on the basis of the information obtained from their sensors, e.g., laser sensors, microphones, and cameras, and recognize a variety of objects, places, and situations in an ambient environment. Above all, we consider it important for the robot to learn the names that humans associate with places in the environment and the spatial areas corresponding to these names[ 1 ].
  • Nonparametric Bayesian Double Articulation Analyzer (NPB-DAA)
    This is a Python implementation for Nonparametric Bayesian Double Articulation Analyzer (NPB-DAA). The NPB-DAA can directly acquire language and acoustic models from observed continuous speech signals.

    This generative model is called hierarchical Dirichlet process hidden language model (HDP-HLM), which is obtained by extending the hierarchical Dirichlet process hidden semi-Markov model (HDP-HSMM) proposed by Johnson et al. An inference procedure for the HDP-HLM is derived using the blocked Gibbs sampler originally proposed for the HDP-HSMM.

  • SpCoMapping
    This model is a new statistical semantic mapping method called SpCoMapping, which integrates probabilistic spatial concept acquisition and simultaneous localization and mapping via Markov random field to yield semantic information. SpCoMapping can deal with multimodal sensor information to generate a semantic map. We also developed a nonparametric Bayesian extension of SpCoMapping using the Dirichlet process to enable the method to automatically estimate the adequate number of categories.
  • Active SLAM
    The source code for making the map by robot its self using the frontier approach based active SLAM algorithm. The source code was developed for working on Toyota HSR and its sensors. We use the active SLAM as preprocessing for spatial concept formation and spatial concept based semantic mapping.

 [ 1 ] Akira Taniguchi, Tadahiro Taniguchi, Tetsunari Inamura,Spatial Concept Acquisition for a Mobile Robot that Integrates Self-Localization and Unsupervised Word Discovery from Spoken Sentences,IEEE Transactions on Cognitive and Developmental Systems, Vol.8 (4), pp. 285-297 .(2016)DOI: 10.1109/TCDS.2016.2565542