Decreasing Influence of the Error Due to Acquired Inhomogeneity of Sensors by the Means of Artificial Intelligence

Jotsov, Vladimir; Кочан, Орест; Su, Jun

doi:10.1007/978-3-319-78437-3_5

Cited by 2 publications

(2 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Poor understanding of sensor characteristics such as on papers [33][34][35] affect the application of the system in the application of appropriate algorithms to produce better accuracy. In paper [35] the use of certain algorithmic methods in identifying sensor measurement data errors affects the performance of a system such as the use of Bayesian approaches to evaluate navigation sensors, in paper [33] the use of artificial intelligence neural network methods to improve the measurement accuracy of temperature sensors in thermocouple systems, and the use of software algorithms such as wavelet signal-based approaches used to detect errors Sensor measurements on an industrial scale [34]. Based on the papers [33][34][35] it is possible to state that every use of sensors in the measurement of a particular system requires an algorithm to correct the readings by the sensor, to produce accurate output.…”

Section: Literature Review and Problem Statementmentioning

confidence: 99%

“…In paper [35] the use of certain algorithmic methods in identifying sensor measurement data errors affects the performance of a system such as the use of Bayesian approaches to evaluate navigation sensors, in paper [33] the use of artificial intelligence neural network methods to improve the measurement accuracy of temperature sensors in thermocouple systems, and the use of software algorithms such as wavelet signal-based approaches used to detect errors Sensor measurements on an industrial scale [34]. Based on the papers [33][34][35] it is possible to state that every use of sensors in the measurement of a particular system requires an algorithm to correct the readings by the sensor, to produce accurate output. The application of the Kalman filter algorithm is required in Hooke's law demonstration system to correct ultrasonic sensor readings to produce more accurate distance parameters.…”

Section: Literature Review and Problem Statementmentioning

confidence: 99%

See 1 more Smart Citation

Implementation of Kalman filter algorithm to optimize the calculation of ultrasonic sensor distance value in Hooke law props system

Pratiwi,

Fadli,

Cahyanto

et al. 2024

EEJET

View full text Add to dashboard Cite

The Kalman filter algorithm is very important as a recursive algorithm method to optimize sensor output from physical parameter measurement systems, especially physics practicum demonstration systems. One of the distance parameter measurement demonstration systems used in Hooke’s law demonstration system is applied in physics practicum, the system has problems related to fluctuating or unstable sensor output. This research implements the Kalman filter algorithm on the Arduino IDE sketch to reduce noise that appears at the ultrasonic sensor output. The methodology used in this study includes the application of the Kalman filter algorithm to the Arduino IDE sketch with the variable value of the Kalman filter algorithm equation modified with a value of R=10, H=1, and Q=1, and returns the filtered Kalman out value. The Arduino output results are exported to Ms. Excel for further analysis and generate a filtered ultrasonic sensor output signal graph compared without using the Kalman filter. The ultrasonic sensor output noise filtration effectively reduces noise by showing a decrease in the mean squared error (MSE) value and obtaining the best performance of up to 89.23 %. The accuracy of Kalman filter filtration results can be seen from the calculation that the spring constant of filtered metal materials is smaller than the conventional measurement spring constant. Accurate and effective results with the implementation of the Kalman filter algorithm can be developed for the variation values of distance parameters and Kalman filter algorithm variables (R, Q, and H) with other value variations, especially variables that produce filtering curves close to straight lines. It was concluded that the Kalman filter algorithm was able to improve the performance of Hooke’s law prop system

show abstract