README

ROS / Gazebo based simulation

IRB 14000 YuMi control project, to grasp novel objects depending on neural network model to find the grasping
position with the highest quality of grasping.

Simulation is done in gazebo, simulation includes the robot, kinect depth camera and the object.

The package includes the inverse kinematics of YuMi robot.

- After pulling and building the project, use the following command to start YuMi control and simulation: 

$ roslaunch yumi_launch yumi_gazebo_traj_pos_control.launch

- To start planning excution with moveit, use:

$ roslaunch yumi_moveit_config demo_planning_excution.launch

- An example of controlling the robot to reach a desired position is included in python script in yumi_test_controllers/scripts/:

$ rosrun yumi_test_controllers test_moveit.py