Kalman Filter setpoint values

Projects Libraries
1

Dear folks,

How to adjust the below functions using the Kalman filter library.

SimpleKalmanFilter pressureKalmanFilter(1, 1, 0.01);

  // read a reference value from A0 and map it from 0 to 100
  float real_value = analogRead(A0)/1024.0 * 100.0;
  
  // add a noise to the reference value and use as the measured value
  float measured_value = real_value + random(-100,100)/100.0;

  // calculate the estimated value with Kalman Filter
  float estimated_value = simpleKalmanFilter.updateEstimate(measured_value);

Please refer the link: SimpleKalmanFilter/BasicKalmanFilterExample.ino at master · denyssene/SimpleKalmanFilter · GitHub

In my case sensor gives with noise value so real_value and measured_value are the same.

Thank you in advance.

2

The lines

  // add a noise to the reference value and use as the measured value
  float measured_value = real_value + random(-100,100)/100.0;

are in the demo code to generate additional noise. You don't want/need that in your code.

what's your exact question?

3

Dear J-M-L,

Thank you for your response.
How to tune this values in "SimpleKalmanFilter pressureKalmanFilter(1, 1, 0.01)"?

Could you please explain these below three terms with an example?

SimpleKalmanFilter(e_mea, e_est, q);
e_mea: Measurement Uncertainty
e_est: Estimation Uncertainty
q: Process Noise

Assume my sensor's output is a mixing of noise and Fluctuating between 1.1 to 1.7 values.

Thank you in advance

4

from the doc:

  • e_mea: Measurement Uncertainty - How much do we expect to our measurement vary
  • e_est: Estimation Uncertainty - Can be initialised with the same value as e_mea since the kalman filter will adjust its value.
  • q: Process Variance - usually a small number between 0.001 and 1 - how fast your measurement moves. Recommended 0.01. Should be tunned to your needs.
5

Dear J-M-L,

Thank you for your quick response,

My value fluctuates between 1.2 to 1.7, So It means

e_mea= 0.5
e_est= 0.5
q= 0.01

Please revert if I'm wrong.

Thank you in advance.

6

For best operations use the setProcessNoise() property to update the reading times.

void fReadBattery( void * parameter )
{
  const float r1 = 50500.0f; // R1 in ohm, 50K
  const float r2 = 10000.0f; // R2 in ohm, 10k potentiometer
  const TickType_t xFrequency = 1000; //delay for mS
  float    adcValue = 0.0f;
  float    Vbatt = 0.0f;
  int      printCount = 0;
  float    vRefScale = (3.3f / 4096.0f) * ((r1 + r2) / r2);
  uint64_t TimePastKalman  = esp_timer_get_time(); // used by the Kalman filter UpdateProcessNoise, time since last kalman calculation
  SimpleKalmanFilter KF_ADC_b( 1.0f, 1.0f, .01f );
  TickType_t xLastWakeTime = xTaskGetTickCount();
  for (;;)
  {
    adc1_get_raw(ADC1_CHANNEL_0); //read and discard
    adcValue = float( adc1_get_raw(ADC1_CHANNEL_0) ); //take a raw ADC reading
    KF_ADC_b.setProcessNoise( (esp_timer_get_time() - TimePastKalman) / 1000000.0f ); //get time, in microsecods, since last readings
    adcValue = KF_ADC_b.updateEstimate( adcValue ); // apply simple Kalman filter
    Vbatt = adcValue * vRefScale;
    xSemaphoreTake( sema_CalculatedVoltage, portMAX_DELAY );
    CalculatedVoltage = Vbatt;
    xSemaphoreGive( sema_CalculatedVoltage );
    
      printCount++;
      if ( printCount == 3 )
      {
      //log_i( "Vbatt %f", Vbatt );
      printCount = 0;
      }
    
    TimePastKalman = esp_timer_get_time(); // time of update complete
    xLastWakeTime = xTaskGetTickCount();
    vTaskDelayUntil( &xLastWakeTime, xFrequency );
    //log_i( "fReadBattery %d",  uxTaskGetStackHighWaterMark( NULL ) );
  }
  vTaskDelete( NULL );
}
7

That's the range of your data, it's not the Measurement Uncertainty. (or did you mean you get a varying value of 1.2 to 1.7 by measuring a known, unchanging reference??)

In Kalman's world, the measurement uncertainty is the variance of the measurement errors. this is often provided by the sensor's vendor or can be derived by a calibration procedure (you know the true value, you make say 100 measures and calculate the variance of the error between the measurement and true value)

8

Dear J-M-L,

I'm sensing an RH value with my Arduino setup is flutes between 1.2 to 1.7 every ten seconds once a 2.2 value shows.

I have an RH Master meter it is showing a constant value of 1.55.

Could you please give me some guidance on how to approach the Kalman?

Thank you in advance.

9

Dear Idahowalker,

Thank you for your response,

I'm not able to understand the logic of the code, My request could you please make a code for sensing the sensor voltage (0V to 1V/1000mV). The sensor gives noise value including actual value.

Thank you in advance.

10

so you are saying the true value si 1.55 and the Arduino sees between 1.20 and 1.70

run 100 measurements and calculate the variance of the measurement errors.

11

Dear J-M-L,
Thank you for your quick response.

Okay I got the variance of the measurement error.

May I know the Next step please?

Thank you in advance.

12

You use that for the first two parameters and something small for the last one

13

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.