Discrimination of dual-arm motions using a joint posterior probability neural network for human-robot interfaces

Taro Shibanoki, Toshio Tsuji

Research output: Chapter in Book/Report/Conference proceedingChapter


This chapter describes a novel dual-arm motion discrimination method that combines posterior probabilitiesestimated independently for left and right arm movements, and its application to control arobotic manipulator. The proposed method estimates the posterior probability of each single-arm motionthrough learning using recurrent probabilistic neural networks. The posterior probabilities outputfrom the networks are then combined based on motion dependency between arms, making it possible tocalculate a joint posterior probability of dual-arm motions. With this method, all the dual-arm motionsconsisting of each single-arm motion can be discriminated through leaning of single-arm motions only.In the experiments performed, the proposed method was applied to the discrimination of up to 50 dualarmmotions. The results showed that the method enables relatively high discrimination performance.In addition, the possibility of applying the proposed method for a human-robot interface was confirmedthrough operation experiments for the robotic manipulator using dual-arm motions.

Original languageEnglish
Title of host publicationHandbook of Research on Biomimetics and Biomedical Robotics
PublisherIGI Global
Number of pages129
ISBN (Electronic)9781522529941
ISBN (Print)1522529934, 9781522529934
Publication statusPublished - Dec 15 2017
Externally publishedYes

ASJC Scopus subject areas

  • Engineering(all)
  • Computer Science(all)
  • Chemical Engineering(all)
  • Biochemistry, Genetics and Molecular Biology(all)


Dive into the research topics of 'Discrimination of dual-arm motions using a joint posterior probability neural network for human-robot interfaces'. Together they form a unique fingerprint.

Cite this