Videos

In order to better visualize the data generated from our Simulink model of the inverted pendulum, our team created a VPython program which reads in the data generated through the Matlab GUI and uses it to update the position of the cart and pendulum in a video. In the following videos, there is a horizontal axis with hashes marking every meter. The hash mark which extends above the axis marks 0 m.

The first video shows simulation data which our team thought was the most interesting visually. The pendulum continually oscillates throughout the video to demonstrate how the movement of the cart affects the angle of the pendulum. We found that a high proportional gain in the controller will cause the pendulum to be more likely to continually oscillate while larger integral and derivative gains will cause the oscillations to quickly damp over time. Fig. 1 shows the graphs used to generate the video.

Fig. 1: Data generated from P = 200, I = 30, D = 15, N = 10.5, Ts = 0.10, Angle = 3.4

 

 

The second video shows the pendulum oscillations quickly damping. The pendulum then moves in one direction at a constant rate after the pendulum stabilizes. The data used to generate this video is shown in Fig. 2.

Fig. 2: Data generated from P = 178.4, I = 4, D = 18.4, N = 19.32, Ts = 0.03981, Angle = 3.4

 

 

The third video shows a simulation in which the controller fails in keeping the pendulum upright. Fig. 3 shows the data used to generate this video. The video is at 25% normal speed to better show the motion of the system.

Fig. 3: Data generated from P = 232, I = 4, D = 18, N = 19.32, Ts = 0.12, Angle = 3.35