Making an ESP32-based quadcopter from scratch
Python arduinoautopilotcontrol-theorydroneesp32flight-controlflight-controllermavlinkquadcopterquadcopter-firmwareuav
Add parameter for configuring gyro bias lpf + reset the filter on `reset` command
Oleg Kalachev7ad302279898
diff --git a/flix/cli.ino b/flix/cli.inoindex 2e4f9f8..bd67c30 100644--- a/flix/cli.ino+++ b/flix/cli.ino@@ -6,6 +6,7 @@#include "pid.h"#include "vector.h"#include "util.h"+#include "lpf.h"extern const int MOTOR_REAR_LEFT, MOTOR_REAR_RIGHT, MOTOR_FRONT_RIGHT, MOTOR_FRONT_LEFT;extern const int RAW, ACRO, STAB, AUTO;@@ -14,6 +15,7 @@ extern uint16_t channels[16];extern float controlTime;extern int mode;extern bool armed;+extern LowPassFilter<Vector> gyroBiasFilter;const char* motd ="\nWelcome to\n"@@ -178,6 +180,7 @@ void doCommand(String str, bool echo = false) {#endif} else if (command == "reset") {attitude = Quaternion();+ gyroBiasFilter.reset();} else if (command == "reboot") {ESP.restart();} else {diff --git a/flix/imu.ino b/flix/imu.inoindex 22cf3e3..ffdc1ff 100644--- a/flix/imu.ino+++ b/flix/imu.ino@@ -19,6 +19,8 @@ Vector acc; // accelerometer output, m/s/sVector accBias;Vector accScale(1, 1, 1);+LowPassFilter<Vector> gyroBiasFilter(0.001);+void setupIMU() {print("Setup IMU\n");imu.begin();@@ -50,8 +52,6 @@ void readIMU() {void calibrateGyroOnce() {static Delay landedDelay(2);if (!landedDelay.update(landed)) return; // calibrate only if definitely stationary-- static LowPassFilter<Vector> gyroBiasFilter(0.001);gyroBias = gyroBiasFilter.update(gyro);}diff --git a/flix/parameters.ino b/flix/parameters.inoindex e58abed..498bfb7 100644--- a/flix/parameters.ino+++ b/flix/parameters.ino@@ -59,6 +59,7 @@ Parameter parameters[] = {{"IMU_ACC_SCALE_X", &accScale.x},{"IMU_ACC_SCALE_Y", &accScale.y},{"IMU_ACC_SCALE_Z", &accScale.z},+ {"IMU_GYRO_BIAS_A", &gyroBiasFilter.alpha},// estimate{"EST_ACC_WEIGHT", &accWeight},{"EST_RATES_LPF_A", &ratesFilter.alpha},