In this work, we solve the multifingered grasping problem using multimodal reinforcement learning. The simulation and real robot experiments with dedicated initial grasping poses show that our method outperforms hard-coded closing finger motion and the agent with fewer modalities in the grasp success rate for seen and unseen objects.

PDF file can be downloaded from here.



  author  = {Liang, Hongzhuo and Cong, Lin and Hendrich, Norman and Li, Shuang and Sun, Fuchun and Zhang, Jianwei},
  journal = {IEEE Robotics and Automation Letters (RA-L)},
  title   = {Multifingered Grasping Based on Multimodal Reinforcement Learning},
  year    = {2022},
  volume  = {7},
  number  = {2},
  pages   = {1174-1181},
  doi     = {10.1109/LRA.2021.3138545}