🤖 Motion Planning for the KUKA LBR iiwa arm

The project implements Motion Planning for the KUKA LBR iiwa robotic arm.

Given a set of points the end effector must go through and an obstacle, the goal of the project is to implement a Motion Planning, performing null-space reconfiguration, using MATLAB and the CoppeliaSim simulator.

Prerequisites and how to run it

In order to run the project, MATLAB 2020 and CoppeliaSim robot simulator must be installed. Add the libraries necessary for the communication with CoppeliaSim copying them from the CoppeliaSim installation folder:

• remApi.m
• remoteApi.so
• remoteApiProto.m

in a subfolder vrep_lib in the current workspace.

Open the scene vrep_model.ttt defined in the vrep_sim sub-folder in the CoppeliaSim simulator and execute main.m from the MATLAB environment.

Results

Two possible motion controllers were used, one being a Cartesian Motion controller and the second being a Joint Motion controller. It's possible to configure which controller to use during the simulation by setting the controller variable in main.m (line 17) either to 'cartesian' or 'joint'.

The following videos show two versions of motion planning for the arm: on the left, the standard implementation without obstacle-avoidance, on the right, the obstacle-free motion implemented by exploiting 5 control points placed on the arm itself. The shown results are obtained using cartesian motion control, however the simulation can be tested using joint motion control as well.

For further information, please consult the project presentation or the youtube video