# Introduction to Robot Characterization¶

The characterization tools consist of a python application that runs on the user’s PC and matching robot code that runs on the user’s robot. The PC application will send control signals to the robot over NetworkTables, while the robot sends data back to the application. The application then processes the data and determines characterization parameters for the user’s robot mechanism, as well as producing diagnostic plots. Data can be saved (in JSON format) for future use, if desired.

## What is ”Characterization?“¶

”Characterization“ - or, more formally, system identification - is the process of determining a mathematical model for the behavior of a system through statistical analysis of its inputs and outputs.

In FRC, the most common system that we’re interested in characterizing is the permanent-magnet DC motor. In particular, we’re interested in figuring out which motor *input* (i.e. voltage from the motor controller) is required to achieve our desired *outputs* (i.e. velocity and acceleration of the motor).

Fortunately, it is not so difficult to do this. A permanent-magnet DC motor (with no load other than friction and inertia) will obey the following ”voltage-balance equation“ (for more information, see this paper):

where \(V\) is the applied voltage, \(d\) is the displacement (position) of the motor, \(\dot{d}\) is its velocity, and \(\ddot{d}\) is its acceleration (the ”overdot“ notation traditionally denotes the derivative with respect to time).

Heuristically, we can interpret the coefficients in the above equation as follows:

`kS`

is the voltage needed to overcome the motor’s static friction, or in other words to just barely get it moving; it turns out that this static friction (because it’s, well, static) has the same effect regardless of velocity or acceleration. That is, no matter what speed you’re going or how fast you’re accelerating, some constant portion of the voltage you’ve applied to your motor (depending on the specific mechanism assembly) will be going towards overcoming the static friction in your gears, bearings, etc; this value is your kS. Note the presence of the signum function, because friction force always opposes the direction-of-motion.

`kV`

describes how much voltage is needed to hold (or ”cruise“) at a given constant velocity while overcoming the electromagnetic resistance in the motor and any additional friction that increases with speed (known as viscous drag). The relationship between speed and voltage (at constant acceleration) is almost entirely linear (with FRC® components, anyway) because of how permanent-magnet DC motors work.

`kA`

describes the voltage needed to induce a given acceleration in the motor shaft. As with `kV`

, the relationship between voltage and acceleration (at constant velocity) is almost perfectly linear for FRC components.

Once these coefficients have been determined (here accomplished by a multiple linear regression), we can then take a given desired velocity and acceleration for the motor and calculate the voltage that should be applied to achieve it. This is very useful - not only for, say, following motion profiles, but also for making mechanisms more controllable in open-loop control, because your joystick inputs will more closely match the actual mechanism motion.

Some of the tools in this toolsuite introduce additional terms into the above equation to account for known differences from the simple case described above - details for each tool can be found below:

## Included Characterization Tools¶

הערה

Many other types of mechanisms can be characterized by simply adapting the existing code in this library.

The robot characterization toolsuite currently supports characterization for:

Simple Motor Setups

Drivetrains

Arms

Elevators

### Simple Motor Characterization¶

The simple motor characterization tool determines the best-fit parameters for the equation:

where \(V\) is the applied voltage, \(d\) is the displacement (position) of the drive, \(\dot{d}\) is its velocity, and \(\ddot{d}\) is its acceleration. This is the model for a permanent-magnet dc motor with no loading other than friction and inertia, as mentioned above, and is an accurate model for flywheels, turrets, and horizontal linear sliders.

### Drivetrain Characterization¶

The drivetrain characterization tool determines the best-fit parameters for the equation:

where \(V\) is the applied voltage, \(d\) is the displacement (position) of the drive, \(\dot{d}\) is its velocity, and \(\ddot{d}\) is its acceleration. This is the same modeling equation as is used in the simple motor characterization - however, the drivetrain characterizer is specifically set up to run on differential drives, and will characterize each side of the drive independently if desired.

The drivetrain characterizer can also determine the effective trackwidth of your robot using a gyro. More information on how to run the characterization is available in the track width characterization article.

### Arm Characterization¶

The arm characterization tool determines the best-fit parameters for the equation:

where \(V\) is the applied voltage, \(\theta\) is the angular displacement (position) of the arm, \(\dot{\theta}\) is its angular velocity, and \(\ddot{\theta}\) is its angular acceleration. The cosine term (\(kCos\)) is added to correctly account for the effect of gravity.

### Elevator Characterization¶

The elevator characterization tool determines the best-fit parameters for the equation:

where \(V\) is the applied voltage, \(d\) is the displacement (position) of the drive, \(\dot{d}\) is its velocity, and \(\ddot{d}\) is its acceleration. The constant term (\(kG\)) is added to correctly account for the effect of gravity.

## דרישות מקדימות¶

To use the Robot Characterization Toolsuite, you must have Python 3.7 installed on your computer, as well as the standard WPILib programming toolsuite.

אזהרה

Do not install Python from the Microsoft Store. Please use the link above to download and install Python.

## Installing and Launching the Toolsuite¶

To install the Robot Characterization Toolsuite, open a console and enter the following command

```
pip install frc-characterization
```

The toolsuite, and all of its dependencies, should be automatically downloaded and installed. If you are using a Windows machine and the command pip is not recognized, ensure that your python scripts folder has been added to the PATH.

הערה

If you are on Ubuntu, you will have to manually install tkinter with `sudo apt-get install python3-tk`

. You will also have to use the `pip3`

command instead of `pip`

as `pip`

refers to Python 2 on Ubuntu distributions.

If you already have the toolsuite installed, be sure to update it regularly to benefit from bugfixes and new features additions:

```
pip install --upgrade frc-characterization
```

הערה

If you would like to use the beta version of this tool, you must type `pip install --pre --upgrade frc-characterization`

instead.

Once the toolsuite has been installed, launch a new drive characterization project to ensure that it works by running the following command from powershell or a terminal window.

```
frc-characterization drive new
```

The new project GUI should open momentarily. To launch other characterization projects, simply replace `drive`

with the desired characterization type (`arm`

, `elevator`

, `simple-motor`

).

While the new project GUI has buttons for launching both the logging tool and the analyzer tool, these can also be launched directly from the CLI by replacing `new`

with `logger`

or `analyzer`

.

חשוב

It is highly recommended that you utilize the new project GUI to launch the logger and analyzer tools for more effective unit conversions rather than launching the logger and analyzer from the CLI.

For more information on CLI usage, enter `frc-characterization -h`

.