AUTO: Sensor Correction/Fusion

[schromya]

Description

Work with Dr. Heidari to figure out how to best correct the error in our sensors (encoders, rotational sensors, GPS). This could be a "sensor fusion".

• [x] Collect data w/ SpinUp drivetrain using encoder and GPS values (rpm, position, motor voltage, control command=angle/position to move to). Have the bot move to random positions around the field and collect data every second. Dr. Heidari will look at this data and give us suggestions.
• [ ] Determine best way to implement data correction.
• [ ] Program and test the data correction algorithm.

Possible Data Correction Tactics:

• Ramsete to track curves and correct for disturbances
• Kalman Filter

DOD

This task its considered done when it is thoroughly documented in the Engineering Notebook.

[schromya]

I finished the logger and gathered the data (7/11/2023 9:12pm AKST). The data is in csv format: positionData.txt

I ran the bot through 10 random points on the field within a 36inx36in square, starting at [0,0] with a heading of 0. Data was gathered about every second. The bot would know it’s current position (logged as initX/Y and initHeading), and then calculate the heading it needed to turn to and displacement it needed to move to to get to the next random position (logged as goal x/Y and goalHeading). While the bot moved to the next random position it continuously logged gps position, gps Heading, wheel displacement, wheel velocity, wheel voltage, and wheel power.

• Heading is in degrees from the positive x axis, rotated counterclockwise
• Wheel Displacement is in revolutions (FL = Front Left, FR = Front Right, BL = Back Left, BR = Back Front)
• Wheel Velocity is in RPM
• Wheel Voltage is in volts
• Wheel Power is in watts