예제 #1
0
/**
 * correction - perform a correction of the EKF
 * @params[in] self
 * @params[in] args
 *  - Z - vector of measurements (position, velocity, mag, baro)
 *  - sensors - binary flags for which sensors should be used
 * @return state
 */
static PyObject*
correction(PyObject* self, PyObject* args)
{
	PyArrayObject *vec_z;
	float z[10];
	int sensors;

	if (!PyArg_ParseTuple(args, "O!i", &PyArray_Type, &vec_z,
				   &sensors))  return NULL;
	if (NULL == vec_z)  return NULL;

	if (!parseFloatVecN(vec_z, z, 10))
		return NULL;

	INSCorrection(&z[6], &z[0], &z[3], z[9], sensors);

	return pack_state(self);
}
예제 #2
0
파일: imu.c 프로젝트: Laurenceb/Dactyl
/**
  * @brief  This function runs the EKF, talks to aux sensors with a state machine and applys control loops to servos (main control task loop)
  * @param  None
  * @retval None
  */
void run_imu(void) {
	//Static variables
	static uint32_t Iterations=0,millis_pitot;//Number of ekf iterations, pitot sample timestamp
	static Control_type control=DEFAULT_CONTROL;//The control structure
	static Ubx_Gps_Type gps;		//This is our local copy - theres is also a global, be careful with copying
	static float ac[3],Wind[3];		//The accel is not always avaliable - 100hz update
	static float AirSpeed=0,Baro_Alt,Body_x_To_x=0,Body_x_To_y=0,Mean_Baro_Offset;
	static uint32_t Baro_Pressure;		//Baro pressure is static for use in air density calculations
	static uint8_t PWM_Disabled;		//Keeps track of PWM hardware passthrough state
	//Non Static
	float ma[3],gy[3],gps_velocity[3],gps_position[3],target_vector[3],waypoint[3]={0,0,0},old_density;
	float Delta_Time=DELTA_TIME,x_down,y_down,h_offset,N_t_x,N_t_y,time_to_waypoint,Horiz_t,GPS_Errors[4];
	uint16_t SensorsUsed=0;			//We by default have no sensors
	int32_t Baro_Temperature;		//In units of 0.1C
	//Now read the gyro buffer, convert to float from uint16_t and apply the calibration
	Calibrate_3(gy,&(Gyro_Data_Buffer[1]),&Gyr_Cal_Dat);//Read gyro data buffer - skip the temperature
	//Now read the accel downconvertor buffer and apply calibration
	Accel_Access_Flag=LOCKED;		//Lock the data
	/*LOCKED*/Calibrate_3(ac,Accel_Data_Vector,&Acc_Cal_Dat);//Grab the data from the accel downsampler/DSP filter
	Accel_Access_Flag=UNLOCKED;		//Unlock the data
	//Run the EKF - we feed it float pointers but it expects float arrays - alignment has to be correct
	INSStatePrediction(gy,ac,Delta_Time);	//Run the EKF and incorporate the avaliable sensors
	INSCovariancePrediction(Delta_Time);
	//Process the GPS data
	while(Bytes_In_Buffer(&Gps_Buffer))	//Dump all the data from DMA
		Gps_Process_Byte((uint8_t)(Pop_From_Buffer(&Gps_Buffer)),&gps);
	if(gps.packetflag==REQUIRED_DATA){	
		gps_position[0]=(float)(gps.latitude-Home_Position.Latitude)*LAT_TO_METERS;//Remember, NED frame, and gps altitude needs *=-1
		gps_position[1]=(float)(gps.longitude-Home_Position.Longitude)*Long_To_Meters_Home;//This is a global set with home position
		gps_position[2]=-((float)gps.mslaltitude*0.001)+Home_Position.Altitude;//Home is in raw gps coordinates - apart from altitude in m NED
		gps_velocity[0]=(float)gps.vnorth*0.01;//Ublox velocity is in cm/s
		gps_velocity[1]=(float)gps.veast*0.01;
		gps_velocity[2]=(float)gps.vdown*0.01;
		GPS_Errors[0]=(float)gps.horizontal_error*(float)gps.horizontal_error*1e-6*GPS_SPECTRAL_FUDGE_FACTOR/2.0;//Fudge factor for spectral noise density 
		GPS_Errors[1]=(float)gps.vertical_error*(float)gps.vertical_error*1e-6*GPS_SPECTRAL_FUDGE_FACTOR;//Fudge factor is defined in ubx.h/gps header file
		GPS_Errors[2]=(float)gps.speedacc*(float)gps.speedacc*1e-4*GPS_SPECTRAL_FUDGE_FACTOR*GPS_Errors[0]/(GPS_Errors[0]+GPS_Errors[1]);
		GPS_Errors[3]=GPS_Errors[2]*GPS_Errors[1]/GPS_Errors[0];//Ublox5 only gives us 3D speed accuracy? Weight with position errors
		if(OUTDOOR==Operating_Mode)
			INSResetRGPS(GPS_Errors);//Adjust the measurement covariance matrix with reported gps error
		SensorsUsed|=POS_SENSORS|HORIZ_SENSORS|VERT_SENSORS;//Set the flagbits for the gps update
		//Correct baro pressure offset - average the gps altitude over first 100 seconds and apply correction when filter initialised
		old_density=((((float)gps.mslaltitude*0.001-Home_Position.Altitude)*Air_Density*Home_Position.g_e)+(float)Baro_Pressure-Baro_Offset\
		-Mean_Baro_Offset)*(0.01/(float)GPS_RATE);//reuse variable here
		if(Iterations++>100*GPS_RATE)
			Baro_Offset+=old_density;//fixed tau: 0.01s^-1, use the averaged offset to take out gps altitude error in the home pos
		else
			Mean_Baro_Offset+=old_density;
		if(!Gps.packetflag)Gps=gps;	//Copy the data over to the main 'thread' if the global unlocked
		gps.packetflag=0x00;		//Reset the flag
	}
	//Now the state dependant I2C stuff
	if(Completed_Jobs&(1<<MAGNO_READ)) {	//This I2C job will run whilst the prediction runs
		Completed_Jobs&=~(1<<MAGNO_READ);//Wipe the job completed bit
		Calibrate_3(ma,Magno_Data_Buffer,&Mag_Cal_Dat);//Calibrate - the EKF can take a magnetometer input in any units
		SensorsUsed|=MAG_SENSORS;	//Let the EKF know what we used
	}
	if(Completed_Jobs&(1<<BMP_24BIT)) {
		Completed_Jobs&=~(1<<BMP_24BIT);//Wipe the job completed bit
		Baro_Pressure=Bmp_Press_Buffer;	//Copy over from the read buffer
		flip_adc24(&Baro_Pressure);	//Fix endianess
		Bmp_Simp_Conv(&Baro_Temperature,&Baro_Pressure);//Convert to a pressure in Pa
		Baro_Alt=(Baro_Offset-(float)Baro_Pressure)/(Home_Position.g_e*Air_Density);//Use the air density calculation (also used in pitot), linear approx
		SensorsUsed|=BARO_SENSOR;	//We have used the baro sensor
		Balt=Baro_Alt;
		if(READ==UAVtalk_conf.semaphores[BARO_ACTUAL]) {//If this data has been read, we can update it (avoids risk of overwriting data mid packet tx)
			UAVtalk_Altitude_Array[0]=Baro_Alt;//+Home_Position.Altitude;//Populate Baro_altitude UAVtalk packet here - Altitude is MSL altitude in m
			UAVtalk_Altitude_Array[1]=(float)Baro_Temperature/10.0;//Note that as baro_actual has three independant 32bit
			UAVtalk_Altitude_Array[2]=(float)Baro_Pressure*1e-3;//Variables, can be set directly here without passing data back - note pressure in kPa
			UAVtalk_conf.semaphores[BARO_ACTUAL]=WRITE;//Set the semaphore to indicate this has been written
		}
	}
	if(Completed_Jobs&(1<<BMP_16BIT)) {
		Completed_Jobs&=~(1<<BMP_16BIT);//Wipe the job completed bit
		Bmp_Copy_Temp();		//Copy the 16 bit temperature out of its buffer into the temperature global
		old_density=Air_Density;	//Save old air density so we can find the delta
		Air_Density=Calc_Air_Density((float)gps.mslaltitude*1e-3,(float)Baro_Pressure);//Find air density using atmospheric model (Kgm^-3)
		Baro_Offset+=Baro_Alt*(Air_Density-old_density)*Home_Position.g_e;//Correct the baro offset term to account for changing density
	}
	if(Completed_Jobs&(1<<PITOT_READ)) {
		Completed_Jobs&=~(1<<PITOT_READ);//Wipe the job completed bit
		if((Millis-millis_pitot)<2*PITOT_PERIOD) {//An uncorrupted pitot data sample - the bmp085 has same address as ltc2481 global config
			Pitot_Pressure=Pitot_Conv((uint32_t)Pitot_Pressure);//Align and sign the adc value - 1lsb=~0.24Pa
			AirSpeed=Pitot_convert_Airspeed(Pitot_Pressure, Air_Density);//Windspeed
			Wind[0]*=WIND_TAU;Wind[1]*=WIND_TAU;//Low pass filter
			Wind[0]+=(1.0-WIND_TAU)*(Nav.Vel[0]-AirSpeed*Body_x_To_x);//This assumes horizontal wind and neglidgible slip
			Wind[1]+=(1.0-WIND_TAU)*(Nav.Vel[1]-AirSpeed*Body_x_To_y);
			Balt=(float)AirSpeed;	//Pitot_Pressure;//Note debug
		}
		millis_pitot=Millis;		//Save a timestamp so we can monitor conversion time, BMP085 corruption will double this
	}
	INSCorrection(ma,gps_position,gps_velocity,-Baro_Alt,SensorsUsed);//Note that Baro has to be in NED frame
	if(!Nav_Flag) {Nav_Global=Nav; Nav_Flag=(uint32_t)0x01;}//Copy over Nav state if its been unlocked
	//EKF is finished, time to run the guidance
	if(New_Waypoint_Flagged) {memcpy(waypoint,Waypoint_Global,sizeof(waypoint)); New_Waypoint_Flagged=0;}//Check for any new waypoints set
	target_vector[0]=waypoint[0]-Nav.Pos[0];//A float vector to the waypoint in NED space
	target_vector[1]=waypoint[1]-Nav.Pos[1];
	target_vector[2]=waypoint[2]-Nav.Pos[2];//Now work out the eta at the waypoint (just the x,y/North,East position)
	Horiz_t=sqrtf(target_vector[0]*target_vector[0]+target_vector[1]*target_vector[1]);//Horizontal distance to target
	N_t_x=target_vector[0]/Horiz_t;
	N_t_y=target_vector[1]/Horiz_t;		//Normalised horizontal target vector
	time_to_waypoint=Horiz_t/(control.airframe.airspeed+Wind[0]*N_t_x+Wind[1]*N_t_y);//Time if we travel in a straight line - ignores vertical offset
	N_t_x=target_vector[0]-Wind[0]*time_to_waypoint;
	N_t_y=target_vector[1]-Wind[1]*time_to_waypoint;//Modify the aiming point to account for cross track error
	AirSpeed=Nav.Vel[0]-Wind[0];AirSpeed*=AirSpeed;//Do this here to decrease runtime a little
	AirSpeed=sqrtf(AirSpeed+(Nav.Vel[1]-Wind[1])*(Nav.Vel[1]-Wind[1])+Nav.Vel[2]*Nav.Vel[2]);//Work out true airspeed assume horizontal wind
	//Move the body x,y axes into earth NED frame using Reb, looking at z=down components of body x and y
	x_down=2 * (Nav.q[1] * Nav.q[3] - Nav.q[0] * Nav.q[2]);
	y_down=2 * (Nav.q[2] * Nav.q[3] + Nav.q[0] * Nav.q[1]);
	//Work out the heading offset - z component of target vector cross body x axis
	Body_x_To_x=Nav.q[0] * Nav.q[0] + Nav.q[1] * Nav.q[1] - Nav.q[2] * Nav.q[2] - Nav.q[3] * Nav.q[3];
	Body_x_To_y=2*Nav.q[1] * Nav.q[2] + Nav.q[0] * Nav.q[3];//These are saved for subsequent iterations as static variables
	Horiz_t=sqrtf(N_t_x*N_t_x+N_t_y*N_t_y);	//Normalise the wind corrected target vector
	N_t_x/Horiz_t;
	N_t_y/Horiz_t;
	h_offset=Body_x_To_x * N_t_x + Body_x_To_y * N_t_y -1;//We do dot product to get cos(angle), then subtract 1
	if(Body_x_To_x * N_t_y - Body_x_To_y * N_t_x<0)//Multiply by -1 depending on the direction - cos(-angle)==cos(angle)
		h_offset*=-1;
	//Run the control loops if we arent controlled from the ground
	if(Ground_Flag&0x0F) {//TODO implement ground control using PWM input capture and sanity check, as well as PWM passthrough?
		if(PWM_Disabled) {//If PWM isnt already enabled, enable it
			Timer_GPIO_Enable();
			PWM_Disabled=0;
		}
		//Pitch setpoint control pid (actually a PI) -- Note- uses dynamic pressure 
		Run_PID(&(control.pitch_setpoint),&(control.airframe.pitch_setpoint),control.airframe.airspeed-AirSpeed,0);
		//Roll setpoint set by heading error
		Run_PID(&(control.roll_setpoint),&(control.airframe.roll_setpoint),h_offset,0);
		if(Ground_Flag&0xF0) {//If we are Armed, the motor can be run - defaults to disarmed
			//Throttle set according to altitude error
			Run_PID(&(control.throttle),&(control.airframe.throttle),\
			target_vector[2]-((AirSpeed*AirSpeed)-(control.airframe.airspeed*control.airframe.airspeed))/(2*Home_Position.g_e),Nav.Vel[2]);
			control.throttle.out+=control.airframe.throttle_optimal;
		}
		else
			control.throttle.out=-1.0;//Apply zero throttle to stop motor spinning
		//Elevator, remember x_down is reversed sign
		Run_PID(&(control.elevator),&(control.airframe.elevator),control.pitch_setpoint.out+x_down,gy[1]-Nav.gyro_bias[1]);
		//Ailerons, TODO - work out if signs sane
		Run_PID(&(control.ailerons),&(control.airframe.ailerons),control.roll_setpoint.out-y_down,gy[0]-Nav.gyro_bias[0]);
		//Rudder, D term takes out turbulence, and I term for roll-bank (no P?)
		Run_PID(&(control.rudder),&(control.airframe.rudder),ac[1],gy[2]-Nav.gyro_bias[2]);
		//Apply the feedforward, linking rudder to roll offset
		control.rudder.out+=control.airframe.rudder_feedforward*control.roll_setpoint.out;
		Apply_Servos(&control);//Servo driver function called here using pwm
	}
	else {//any control code to run whilst in ground mode goes here	
		if(!PWM_Disabled) {//If PWM was previously enabled, disable it
			Timer_GPIO_Enable();
			PWM_Disabled=1;
		}
	}	
}
예제 #3
0
파일: attitude.c 프로젝트: mirasanti/vbrain
/**
 * @brief Use the INSGPS fusion algorithm in either indoor or outdoor mode (use GPS)
 * @params[in] first_run This is the first run so trigger reinitialization
 * @params[in] outdoor_mode If true use the GPS for position, if false weakly pull to (0,0)
 * @return 0 for success, -1 for failure
 */
static int32_t updateAttitudeINSGPS(bool first_run, bool outdoor_mode)
{
	UAVObjEvent ev;
	GyrosData gyrosData;
	AccelsData accelsData;
	MagnetometerData magData;
	BaroAltitudeData baroData;
	GPSPositionData gpsData;
	GPSVelocityData gpsVelData;
	GyrosBiasData gyrosBias;

	static bool mag_updated = false;
	static bool baro_updated;
	static bool gps_updated;
	static bool gps_vel_updated;

	static float baroOffset = 0;

	static uint32_t ins_last_time = 0;
	static bool inited;

	float NED[3] = {0.0f, 0.0f, 0.0f};
	float vel[3] = {0.0f, 0.0f, 0.0f};
	float zeros[3] = {0.0f, 0.0f, 0.0f};

	// Perform the update
	uint16_t sensors = 0;
	float dT;

	// Wait until the gyro and accel object is updated, if a timeout then go to failsafe
	if ( (xQueueReceive(gyroQueue, &ev, FAILSAFE_TIMEOUT_MS / portTICK_RATE_MS) != pdTRUE) ||
	     (xQueueReceive(accelQueue, &ev, 1 / portTICK_RATE_MS) != pdTRUE) )
	{
		// Do not set attitude timeout warnings in simulation mode
		if (!AttitudeActualReadOnly()){
			AlarmsSet(SYSTEMALARMS_ALARM_ATTITUDE,SYSTEMALARMS_ALARM_WARNING);
			return -1;
		}
	}

	if (inited) {
		mag_updated = 0;
		baro_updated = 0;
		gps_updated = 0;
		gps_vel_updated = 0;
	}

	if (first_run) {
		inited = false;
		init_stage = 0;

		mag_updated = 0;
		baro_updated = 0;
		gps_updated = 0;
		gps_vel_updated = 0;

		ins_last_time = PIOS_DELAY_GetRaw();

		return 0;
	}

	mag_updated |= (xQueueReceive(magQueue, &ev, 0 / portTICK_RATE_MS) == pdTRUE);
	baro_updated |= xQueueReceive(baroQueue, &ev, 0 / portTICK_RATE_MS) == pdTRUE;
	gps_updated |= (xQueueReceive(gpsQueue, &ev, 0 / portTICK_RATE_MS) == pdTRUE) && outdoor_mode;
	gps_vel_updated |= (xQueueReceive(gpsVelQueue, &ev, 0 / portTICK_RATE_MS) == pdTRUE) && outdoor_mode;

	// Get most recent data
	GyrosGet(&gyrosData);
	AccelsGet(&accelsData);
	MagnetometerGet(&magData);
	BaroAltitudeGet(&baroData);
	GPSPositionGet(&gpsData);
	GPSVelocityGet(&gpsVelData);
	GyrosBiasGet(&gyrosBias);

	// Discard mag if it has NAN (normally from bad calibration)
	mag_updated &= (magData.x == magData.x && magData.y == magData.y && magData.z == magData.z);
	// Don't require HomeLocation.Set to be true but at least require a mag configuration (allows easily
	// switching between indoor and outdoor mode with Set = false)
	mag_updated &= (homeLocation.Be[0] != 0 || homeLocation.Be[1] != 0 || homeLocation.Be[2]);

	// Have a minimum requirement for gps usage
	gps_updated &= (gpsData.Satellites >= 7) && (gpsData.PDOP <= 4.0f) && (homeLocation.Set == HOMELOCATION_SET_TRUE);

	if (!inited)
		AlarmsSet(SYSTEMALARMS_ALARM_ATTITUDE,SYSTEMALARMS_ALARM_ERROR);
	else if (outdoor_mode && gpsData.Satellites < 7)
		AlarmsSet(SYSTEMALARMS_ALARM_ATTITUDE,SYSTEMALARMS_ALARM_ERROR);
	else
		AlarmsClear(SYSTEMALARMS_ALARM_ATTITUDE);
			
	if (!inited && mag_updated && baro_updated && (gps_updated || !outdoor_mode)) {
		// Don't initialize until all sensors are read
		if (init_stage == 0 && !outdoor_mode) {
			float Pdiag[16]={25.0f,25.0f,25.0f,5.0f,5.0f,5.0f,1e-5f,1e-5f,1e-5f,1e-5f,1e-5f,1e-5f,1e-5f,1e-4f,1e-4f,1e-4f};
			float q[4];
			float pos[3] = {0.0f, 0.0f, 0.0f};

			// Initialize barometric offset to homelocation altitude
			baroOffset = -baroData.Altitude;
			pos[2] = -(baroData.Altitude + baroOffset);

			// Reset the INS algorithm
			INSGPSInit();
			INSSetMagVar(revoCalibration.mag_var);
			INSSetAccelVar(revoCalibration.accel_var);
			INSSetGyroVar(revoCalibration.gyro_var);
			INSSetBaroVar(revoCalibration.baro_var);

			// Initialize the gyro bias from the settings
			float gyro_bias[3] = {gyrosBias.x * F_PI / 180.0f, gyrosBias.y * F_PI / 180.0f, gyrosBias.z * F_PI / 180.0f};
			INSSetGyroBias(gyro_bias);

			xQueueReceive(magQueue, &ev, 100 / portTICK_RATE_MS);
			MagnetometerGet(&magData);

			// Set initial attitude
			AttitudeActualData attitudeActual;
			attitudeActual.Roll = atan2f(-accelsData.y, -accelsData.z) * 180.0f / F_PI;
			attitudeActual.Pitch = atan2f(accelsData.x, -accelsData.z) * 180.0f / F_PI;
			attitudeActual.Yaw = atan2f(-magData.y, magData.x) * 180.0f / F_PI;
			RPY2Quaternion(&attitudeActual.Roll,&attitudeActual.q1);
			AttitudeActualSet(&attitudeActual);

			q[0] = attitudeActual.q1;
			q[1] = attitudeActual.q2;
			q[2] = attitudeActual.q3;
			q[3] = attitudeActual.q4;
			INSSetState(pos, zeros, q, zeros, zeros);
			INSResetP(Pdiag);
		} else if (init_stage == 0 && outdoor_mode) {
			float Pdiag[16]={25.0f,25.0f,25.0f,5.0f,5.0f,5.0f,1e-5f,1e-5f,1e-5f,1e-5f,1e-5f,1e-5f,1e-5f,1e-4f,1e-4f,1e-4f};
			float q[4];
			float NED[3];

			// Reset the INS algorithm
			INSGPSInit();
			INSSetMagVar(revoCalibration.mag_var);
			INSSetAccelVar(revoCalibration.accel_var);
			INSSetGyroVar(revoCalibration.gyro_var);
			INSSetBaroVar(revoCalibration.baro_var);

			INSSetMagNorth(homeLocation.Be);

			// Initialize the gyro bias from the settings
			float gyro_bias[3] = {gyrosBias.x * F_PI / 180.0f, gyrosBias.y * F_PI / 180.0f, gyrosBias.z * F_PI / 180.0f};
			INSSetGyroBias(gyro_bias);

			GPSPositionData gpsPosition;
			GPSPositionGet(&gpsPosition);

			// Transform the GPS position into NED coordinates
			getNED(&gpsPosition, NED);
			
			// Initialize barometric offset to cirrent GPS NED coordinate
			baroOffset = -NED[2] - baroData.Altitude;

			xQueueReceive(magQueue, &ev, 100 / portTICK_RATE_MS);
			MagnetometerGet(&magData);

			// Set initial attitude
			AttitudeActualData attitudeActual;
			attitudeActual.Roll = atan2f(-accelsData.y, -accelsData.z) * 180.0f / F_PI;
			attitudeActual.Pitch = atan2f(accelsData.x, -accelsData.z) * 180.0f / F_PI;
			attitudeActual.Yaw = atan2f(-magData.y, magData.x) * 180.0f / F_PI;
			RPY2Quaternion(&attitudeActual.Roll,&attitudeActual.q1);
			AttitudeActualSet(&attitudeActual);

			q[0] = attitudeActual.q1;
			q[1] = attitudeActual.q2;
			q[2] = attitudeActual.q3;
			q[3] = attitudeActual.q4;

			INSSetState(NED, zeros, q, zeros, zeros);
			INSResetP(Pdiag);
		} else if (init_stage > 0) {
			// Run prediction a bit before any corrections
			dT = PIOS_DELAY_DiffuS(ins_last_time) / 1.0e6f;

			GyrosBiasGet(&gyrosBias);
			float gyros[3] = {(gyrosData.x + gyrosBias.x) * F_PI / 180.0f, 
				(gyrosData.y + gyrosBias.y) * F_PI / 180.0f, 
				(gyrosData.z + gyrosBias.z) * F_PI / 180.0f};
			INSStatePrediction(gyros, &accelsData.x, dT);
			
			AttitudeActualData attitude;
			AttitudeActualGet(&attitude);
			attitude.q1 = Nav.q[0];
			attitude.q2 = Nav.q[1];
			attitude.q3 = Nav.q[2];
			attitude.q4 = Nav.q[3];
			Quaternion2RPY(&attitude.q1,&attitude.Roll);
			AttitudeActualSet(&attitude);
		}

		init_stage++;
		if(init_stage > 10)
			inited = true;

		ins_last_time = PIOS_DELAY_GetRaw();	

		return 0;
	}

	if (!inited)
		return 0;

	dT = PIOS_DELAY_DiffuS(ins_last_time) / 1.0e6f;
	ins_last_time = PIOS_DELAY_GetRaw();

	// This should only happen at start up or at mode switches
	if(dT > 0.01f)
		dT = 0.01f;
	else if(dT <= 0.001f)
		dT = 0.001f;

	// If the gyro bias setting was updated we should reset
	// the state estimate of the EKF
	if(gyroBiasSettingsUpdated) {
		float gyro_bias[3] = {gyrosBias.x * F_PI / 180.0f, gyrosBias.y * F_PI / 180.0f, gyrosBias.z * F_PI / 180.0f};
		INSSetGyroBias(gyro_bias);
		gyroBiasSettingsUpdated = false;
	}

	// Because the sensor module remove the bias we need to add it
	// back in here so that the INS algorithm can track it correctly
	float gyros[3] = {gyrosData.x * F_PI / 180.0f, gyrosData.y * F_PI / 180.0f, gyrosData.z * F_PI / 180.0f};
	if (revoCalibration.BiasCorrectedRaw == REVOCALIBRATION_BIASCORRECTEDRAW_TRUE) {
		gyros[0] += gyrosBias.x * F_PI / 180.0f;
		gyros[1] += gyrosBias.y * F_PI / 180.0f;
		gyros[2] += gyrosBias.z * F_PI / 180.0f;
	}

	// Advance the state estimate
	INSStatePrediction(gyros, &accelsData.x, dT);

	// Copy the attitude into the UAVO
	AttitudeActualData attitude;
	AttitudeActualGet(&attitude);
	attitude.q1 = Nav.q[0];
	attitude.q2 = Nav.q[1];
	attitude.q3 = Nav.q[2];
	attitude.q4 = Nav.q[3];
	Quaternion2RPY(&attitude.q1,&attitude.Roll);
	AttitudeActualSet(&attitude);

	// Advance the covariance estimate
	INSCovariancePrediction(dT);

	if(mag_updated)
		sensors |= MAG_SENSORS;

	if(baro_updated)
		sensors |= BARO_SENSOR;

	INSSetMagNorth(homeLocation.Be);
	
	if (gps_updated && outdoor_mode)
	{
		INSSetPosVelVar(revoCalibration.gps_var[REVOCALIBRATION_GPS_VAR_POS], revoCalibration.gps_var[REVOCALIBRATION_GPS_VAR_VEL]);
		sensors |= POS_SENSORS;

		if (0) { // Old code to take horizontal velocity from GPS Position update
			sensors |= HORIZ_SENSORS;
			vel[0] = gpsData.Groundspeed * cosf(gpsData.Heading * F_PI / 180.0f);
			vel[1] = gpsData.Groundspeed * sinf(gpsData.Heading * F_PI / 180.0f);
			vel[2] = 0;
		}
		// Transform the GPS position into NED coordinates
		getNED(&gpsData, NED);

		// Track barometric altitude offset with a low pass filter
		baroOffset = BARO_OFFSET_LOWPASS_ALPHA * baroOffset +
		    (1.0f - BARO_OFFSET_LOWPASS_ALPHA )
		    * ( -NED[2] - baroData.Altitude );

	} else if (!outdoor_mode) {
		INSSetPosVelVar(1e2f, 1e2f);
		vel[0] = vel[1] = vel[2] = 0;
		NED[0] = NED[1] = 0;
		NED[2] = -(baroData.Altitude + baroOffset);
		sensors |= HORIZ_SENSORS | HORIZ_POS_SENSORS;
		sensors |= POS_SENSORS |VERT_SENSORS;
	}

	if (gps_vel_updated && outdoor_mode) {
		sensors |= HORIZ_SENSORS | VERT_SENSORS;
		vel[0] = gpsVelData.North;
		vel[1] = gpsVelData.East;
		vel[2] = gpsVelData.Down;
	}
	
	/*
	 * TODO: Need to add a general sanity check for all the inputs to make sure their kosher
	 * although probably should occur within INS itself
	 */
	if (sensors)
		INSCorrection(&magData.x, NED, vel, ( baroData.Altitude + baroOffset ), sensors);

	// Copy the position and velocity into the UAVO
	PositionActualData positionActual;
	PositionActualGet(&positionActual);
	positionActual.North = Nav.Pos[0];
	positionActual.East = Nav.Pos[1];
	positionActual.Down = Nav.Pos[2];
	PositionActualSet(&positionActual);
	
	VelocityActualData velocityActual;
	VelocityActualGet(&velocityActual);
	velocityActual.North = Nav.Vel[0];
	velocityActual.East = Nav.Vel[1];
	velocityActual.Down = Nav.Vel[2];
	VelocityActualSet(&velocityActual);

	if (revoCalibration.BiasCorrectedRaw == REVOCALIBRATION_BIASCORRECTEDRAW_TRUE && !gyroBiasSettingsUpdated) {
		// Copy the gyro bias into the UAVO except when it was updated
		// from the settings during the calculation, then consume it
		// next cycle
		gyrosBias.x = Nav.gyro_bias[0] * 180.0f / F_PI;
		gyrosBias.y = Nav.gyro_bias[1] * 180.0f / F_PI;
		gyrosBias.z = Nav.gyro_bias[2] * 180.0f / F_PI;
		GyrosBiasSet(&gyrosBias);
	}

	return 0;
}