[wpimath] Fix drivetrain system identification (#3406)

The units for angular Kv and Ka were inconsistent with the derivation. A
second factory function overload was added for angular units that uses a
trackwidth to convert to the other form.

Notice how section 15.2 of https://file.tavsys.net/control/controls-engineering-in-frc.pdf
defines the angular feedforward as u = Kv,angular v instead of u = Kv,angular + omega.
The units cancel for elements of A but not B, so just the B matrix was incorrect in our code.

This breaks existing C++ code since the units are part of the function
signature.
This commit is contained in:
Tyler Veness
2021-06-05 11:22:05 -07:00
committed by GitHub
parent 989de4a1bf
commit 15c521a7fe
4 changed files with 104 additions and 36 deletions

View File

@@ -98,8 +98,7 @@ class Drivetrain {
frc::Field2d m_fieldSim;
frc::LinearSystem<2, 2, 2> m_drivetrainSystem =
frc::LinearSystemId::IdentifyDrivetrainSystem(
1.98_V / 1_mps, 0.2_V / 1_mps_sq, 1.5_V / 1_rad_per_s,
0.3_V / 1_rad_per_s_sq);
1.98_V / 1_mps, 0.2_V / 1_mps_sq, 1.5_V / 1_mps, 0.3_V / 1_mps_sq);
frc::sim::DifferentialDrivetrainSim m_drivetrainSimulator{
m_drivetrainSystem, kTrackWidth, frc::DCMotor::CIM(2), 8, 2_in};
};

View File

@@ -52,11 +52,11 @@ constexpr double kEncoderDistancePerPulse =
// Toolsuite provides a convenient tool for obtaining these values for your
// robot.
constexpr auto ks = 0.22_V;
constexpr auto kv = 1.98 * 1_V * 1_s / 1_m;
constexpr auto ka = 0.2 * 1_V * 1_s * 1_s / 1_m;
constexpr auto kv = 1.98 * 1_V / 1_mps;
constexpr auto ka = 0.2 * 1_V / 1_mps_sq;
constexpr auto kvAngular = 1.5 * 1_V * 1_s / 1_rad;
constexpr auto kaAngular = 0.3 * 1_V * 1_s * 1_s / 1_rad;
constexpr auto kvAngular = 1.5 * 1_V / 1_mps;
constexpr auto kaAngular = 0.3 * 1_V / 1_mps_sq;
extern const frc::LinearSystem<2, 2, 2> kDrivetrainPlant;