KUKA YouBot navigation using PRM algorithm on MATLAB - CoppeliaSim

Binary occupancy map method is used to generate a map of the scene. The map is created by using the corresponding functions in MATLAB.

After the map is created, mapInflated function is used in order to inflate the objects in the map. By doing so, a more effective path is generated. As a result the movement of the robot becomes more reliable.

Afterwards by applying the PRM function in MATLAB, the path is generated automatically.

The waypoint number of the path is counted and the x,y coordinates of each waypoint is stored in a variable. Also, the velocity vector for both x,y directions is calculated by subtracting the current waypoint from the next waypoint.

The connection between the MATLAB and CoppeliaSim is established. The current coordinates of the robot is obtained from CoppeliaSim using simxGetObjectPosition function.

Then the algorithm first checks the number of waypoints and according to the result it executes the commands that is being sent to the robots wheels.

After the waypoint number check, the algorithm checks the x,y position of the robot and the x,y position of the next waypoint. If the robot reaches the next waypoint, the velocity vector gets updated so that the robot starts to move towards the next waypoint. The velocity vector is updated in x,y velocities.

However in the code, the inverse kinematic calculations are performed. So after the inverse kinematic calculations are done, the power is sent to wheels using simxSetJointTargetVelocity function.

Finally, when the robot reaches the final waypoint, the velocity vector is updated to zero and the power to the wheel actuators are cut.

remoteApi files are not added which are necessary for the connection of MATLAB and CoppeliaSim