Gain-compensation Methodology for a Sinusoidal Scan of a Galvanometer Mirror in Proportional-Integral-Differential Control Using Pre-emphasis Techniques

The overall goal of this procedure is to demonstrate a technique to compensate for the gain reduction of galvanometer mirrors in sine-wave path tracking. Using proportional-integral-differential control for motion blur compensation. This method can be applied in the field of optical engineering to such problems as target tracking, scanning control and motion blur compensation.

The main advantage of this technique is that it relies on pre-configured parameters and the input-to-output ratio, and does not require additional active contour models and hardware. Perform the experiment at an optical bench. And prepare a galvanometer mirror to mount on the bench.

The mirror should be in a custom-made jig, to prevent the mirror and its body from moving during oscillations. Place the jig and mirror in an optical mount. Then, fix the mount to the optical bench to continue.

Control the mirror with a computer equipped with an A/D D/A Board. Extend properly connected cables from the board to the mirror. There, connect the cables to the input and position sockets of the galvanometer mirror.

At the computer, employ a graphical user interface for a sine-wave function generator. Program the interface to perform multiple trials while varying Frequency and Amplitude. For this experiment, use an outer loop over a frequency range.

Set the Minimum frequency to 100 Hertz. Set the Maximum frequency to 500 Hertz. Increment the frequency by 100 Hertz.

The inner loop is over the voltage amplitude. Use a minimum voltage of 10 millivolts. For a maximum voltage, use 500 millivolts.

Increment the voltage by 10 millivolts. Enter the duration of the sine-wave path signal as 2000 samplings. Drive the mirror with the computer and simultaneously record its position signal and determine its angle.

Use a separate file to save the data for each frequency, and each voltage amplitude. Once all of the data is collected, plot the output amplitude as a function of the input amplitude for each of the frequencies. Note that the data for the low frequencies show greater output amplitude as a function of input amplitude than the high frequencies.

This demonstrates the reduction in gain as a function of frequency, for which this technique compensates. Exclude the non-linear behavior and focus on the approximately linear portion of the plots to obtain the pre-emphasis coefficients. This experiment requires a moving image in view of the mirror.

To achieve this, employ a conveyor belt that can move at 30 kilometers per hour. The image for this experiment, a fine texture pattern, has been printed onto tape and applied to the belt. Now, move on to set up the optical elements.

Arrange a camera with a telephoto lens, so that it can use the galvanometer mirror to track the motion of the conveyor belt. Ensure that the galvanometer mirror is connected to the control computer with pre-emphasis software. In order to capture decent images, illuminate the conveyor belt with an appropriate light source.

At the bench, set the appropriate lens focus for the computer display. Next, work to synchronize the mirror and exposure time of the camera. Then, input the camera settings for the expected images.

This image illustrates the timing used by the trigger software. As a function of time, the mirror angle is given by the blue sinusoidal curve. The Red triangle function represents the ideal mirror angle for a scan as a function of time.

Center the exposure time when the angle is at zero degrees and moving with a positive angular velocity. This is where the two curves intersect. Set the software to trigger the camera half of the exposure time earlier and end half of the exposure time later.

Using the pre-emphasis software, input the conveyor belt speed in kilometers per hour. Then, enter the distance from the camera to the belt in meters. The software will calculate the required angular velocity.

Next, enter the frequency at which the mirror is driven. The software calculates the input amplitude. Use the amplitude in the pre-emphasis software.

Once all the parameters have been entered, return to the conveyor belt. Start the belt moving at the speed used in the software to create a moving image. Start the control software and begin recording images from the camera.

the controller compensates for the inertia of galvanometer mirror to improve image blur. Before the pre-emphasis technique, this was the plot of the output signal as a function of the input signal. After applying the technique, this is the plot.

Almost all of the output plots are on the line y equals x. Similarly, before the pre-emphasis technique, this was the plot of gain as a function of the input signal. After applying the technique, almost all of the amplitude plots are on the line y equals zero dB.

This is an image that was placed on the conveyor belt, shown when the belt is at rest. The blue line indicates where the profile at the right was taken. This is the same image when the conveyor belt moves at 30 kilometers per hour and no motion blur compensation is active.

Here, there is motion blur compensation, but pre-emphasis is inactive. In this final version, both motion blur compensation and the pre-emphasis technique are active. It represents a great improvement over the two other options.

Once mastered, this technique can be done in one hour if it is performed properly. After its development, this technique paved the way for researchers in the field of optical engineering to explore motion blur compensation in high-speed scanning.