cyrilli / robot_pick_place_simulator

A Gazebo pick & place simulation used to test and collect training data for deep learning based grasp detection algorithms

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

robot_pick_place_simulator

A Gazebo pick & place simulation used to test and collect training data for deep learning based grasp detection algorithms

This simulation contains models from multiple resources:

  1. Baxter Simulator from Rethink Robotics
  2. Test object models from YCB dataset
  3. Dropbox from Udacity RoboND-Perception-Project

Below are videos for testing a custom grasp detection algorithm in one-object and multiple-object scenarios.

Video: Planar Grasp Detection with One Object

Video: Planar Grasp Detection with Multiple Objects

About

A Gazebo pick & place simulation used to test and collect training data for deep learning based grasp detection algorithms

License:MIT License