Progress - April, 19

The ping pong ball fall out problem is solved by using another initial condition. If we apply 90 degree (vertical position) as initial value of angle. Then it won't be a problem.

But then the robot will be start in the singular configuration, so we increase make the judgment of near singular more strict, to ensure the inverse kinematic works well

Progress - April, 16

After finish all the code and simulation on the MATLAB, we start to integrate the program with the robot.

The first issue we are dealing is using MATLAB to connect with robot. Previously, the program structure didn't include the GUI part, so it works well. But after the GUI part is combined, where should we placed the connection code to prevent the frequent COM port connection for causing the program crash. Eventually, we place the port connection in the background drawing subroutine, and it works just fine. After the connection is done, the rest of work would be insert the motor control function in to each corresponding button.

New problem:

1. The inverse kinematic works well, but we neglect that ping pong ball will fall out if the barrel angle Phi is less than 0. Solving the inverse kinematic with integral Jacobian matrix will converge only to the nearest solution. Therefore, we need to find a alternative solution for fixing this situation.
2. The trajectory projection function works fine, but the problem is to express it with a better way. This also need to be fixed. 
3. Parameter adjusting 
• lambda for the depth of view
• initial velocity of ping pong ball
• the coordinate scaling for task space to background picture.