Cross-Platform, Multi-Use Motion Profiling and Trajectory Generation.

Pathfinder is a library for generating Motion Profiles, a way to smoothly fit and follow a trajectory based upon given waypoints. Currently, both a C and Java API are available, but can be applied to almost any application.

An example profile is given below, with the waypoints:

1. X = -4, Y = -1, Angle = -45 degrees
2. X = -2, Y = -2, Angle = 0
3. X = 0, Y = 0, Angle = 0

The Graph on top is the X/Y position, and the Graph on the bottom is the Velocity.

Pathfinder supports Modifiers. Modifiers are a way to manipulate a trajectory with a given rule.
Pathfinder supports Tank and Swerve Drive modifiers.

Tank Drive:

Swerve Drive:

Add the following lines to your build.gradle if you're using GradleRIO (2018.01.11 or higher):

dependencies {
    compile pathfinder()
}

model {
    frcUserProgram(NativeExecutableSpec) {
        lib library: "pathfinder"
    }
}

If you're not using GradleRIO, do the following:

If you're not using GradleRIO, you must download this manually and copy into ~/wpilib/user/java/lib: http://dev.imjac.in/maven/jaci/pathfinder/Pathfinder-Java/1.8/Pathfinder-Java-1.8.jar
You also have to download this, extract it, and place libpathfinder.so into ~/wpilib/user/java/lib: http://dev.imjac.in/maven/jaci/pathfinder/Pathfinder-JNI/1.8/Pathfinder-JNI-1.8-athena.zip

Download this and put libpathfinder.a in ~/wpilib/user/cpp/lib: http://dev.imjac.in/maven/jaci/pathfinder/Pathfinder-Core/1.8/Pathfinder-Core-1.8-athena.zip
Download this and extract all of its contents to ~/wpilib/user/cpp/include: http://dev.imjac.in/maven/jaci/pathfinder/Pathfinder-Core/1.8/Pathfinder-Core-1.8-headers.zip

If you want to use Pathfinder on your system, you will have to build the library for your platform. You can do this with the following:

./gradlew build

The native libraries will be located under ./Pathfinder-Core/build/libs/pathfinder/ (shared and static libs, platform=any64) and ./Pathfinder-Java/build/libs/pathfinderjava/shared/any64. If you're using native shared (or java), you must put these somewhere on your system load path (e.g. Mac/Linux = /usr/local, Windows = Somwhere on PATH)

To see the usage for each language variation of the API, see the README in their folder.

The RobotPy project has created python bindings around the Pathfinder libraries, and can be found at https://github.com/robotpy/robotpy-pathfinder

The releases on Maven (mentioned above) are built for the NI RoboRIO (v16/7 2018 image). If you go to the github releases, you can download the 1.5 version for Windows/Linux/Mac releases.

If you want to know more about how Pathfinder works, I highly suggest watching the seminar on Motion Profiling by FRC Team 254, which inspired and provided a lot of guidance for this project. Both use the same generation procedure, with some logistical differences.