The new paradigm of Industry 4.0 demand the collaboration between robot and humans.
They could help (human and robot) and collaborate each other without any additional
security, unlike other conventional manipulators. For this, the robot should have the
ability of acquire the environment and plan (or re-plan) on-the-fly the movement avoiding
the obstacles and people.
This work proposes a system that acquires the space of the environment, based on
a Kinect sensor, verifies the free spaces generated by a Point Cloud and executes the
trajectory of manipulators in these free spaces. The simulation system should perform
the path planning of a UR5 manipulator for pick-and-place tasks, while avoiding the
objects around it, based on the point cloud from Kinect. And due to the results obtained
in the simulation, it was possible to apply this system in real situations.
The basic structure of the system is the ROS software, which facilitates robotic applications
with a powerful set of libraries and tools. The MoveIt! and Rviz are examples
of these tools, with them it was possible to carry out simulations and obtain planning
results. The results are reported through logs files, indicating whether the robot motion
plain was successful and how many manipulator poses were needed to create the final
movement. This last step, allows to validate the proposed system, through the use of the
RRT and PRM algorithms. Which were chosen because they are most used in the field
of robot path planning.
