Neural network training for serial multisensor of autonomous vehicle system

Budisusila, Eka Nuryanto; Prasetyowati, Sri Arttini Dwi; Suprapto, Bhakti Yudho; Nawawi, Zainuddin

doi:10.11591/ijece.v12i5.pp5415-5426

Cited by 3 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Digital measurement requires automated tools that use sensors as innovative measuring instruments in measurement systems [2]. Digital distance measurement using ultrasonic sensors has replaced manual measurement without touching the object to be measured and its advantage [3] is resistance to light intensity, object color, and climate change. Ultrasonic sensor in paper [4][5][6] that utilizes ultrasonic sound waves with two transducers as transmitter (pin Trigger) and receiver (pin Echo).…”

Section: Literature Review and Problem Statementmentioning

confidence: 99%

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

Section: Literature Review and Problem Statementmentioning

confidence: 99%

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

show abstract

Real Time System Handling Using Multi Fixed Weight Artificial Neural Network

Budisusila,

Dwi Prasetyowati,

Arifin

et al. 2023

2023 10th International Conference on Electrical Engineering, Computer Science and Informatics (EECSI)

View full text Add to dashboard Cite

Advanced Artificial Neural Network for Steering and Braking Control of Autonomous Electric Vehicle

Budisusila,

Prasetyowati,

Arifin

et al. 2024

IJEER

View full text Add to dashboard Cite

Sensors are necessary for an autonomous electric vehicle (AEV) system to identify its environment and take appropriate action, such avoiding obstacles and crashes. Despite their limitations about color, light, and non-metallic items, cameras, radar, and lidar are widely employed to detect objects surrounding a vehicle. Ultrasonic sensors are weather and light-resistant. Thus, the goal of this work was to create object detectors by combining multiple long-range ultrasonic sensors into a multi-sensor circuit. The Arduino processor incorporates an artificial neural network that uses the advanced artificial neural network as a novel approach to control the sensors. There are two steps to this method: offline training and implementation test. The most ideal neural network weights are found offline using the adaptive back propagation algorithm, and the best fixed weight is then embedded into the neural network software on Arduino for implementation test. Because the system can sense more detail about the vehicle's surroundings and accurately avoid obstacles, the definition of actions by taking the object's distance into consideration is better. As a result, the training can produce an output that is closed to the target with 0.001 errors.

show abstract

Neural network training for serial multisensor of autonomous vehicle system

Cited by 3 publications

References 19 publications

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

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

Real Time System Handling Using Multi Fixed Weight Artificial Neural Network

Advanced Artificial Neural Network for Steering and Braking Control of Autonomous Electric Vehicle

Contact Info

Product

Resources

About