Can MpNet with baxter be used in a changing environment?
Closed this issue · 0 comments
watashihageteru commented
Excuse me, but please ask me about the meaning of the phrase In the paper(https://arxiv.org/pdf/1907.06013.pdf):
"For Baxter robot, we created an L-shaped table top on which 5 different objects representing obstacles were randomly placed to constitute 10 different challenging environments in ROS Gazebo.".
Does this mean that one neural network can learn the plausible root for 10 different changing environments or do I have to prepare 10 different neural networks for 10 environments?