Real time self-calibration algorithm of pressure sensor for robotic hand glove system

This study investigates the use of a novel Proposed Self-Calibration Algorithm (PSCA) of multi pressure sensors in real time on robotic hand glove system. The PSCA should be able to fix major problems in the pressure sensor including hysteresis, variation in gain and lack of linearity with high accuracy. The traditional calibration process for FlexiForce sensor used is a time-consuming task because it is usually done through manual and repetitive identification [1, 2]. Furthermore, a traditional computational method is inadequate for solving the problem since it is extremely difficult to resolve the mathematical formula among multiple confounding pressure variables [1, 3, 4]. In traditional calibration method of FlexiForce sensor, the maximum calibration time is 20 min and the signal decrease is equal to 83% for sinusoidal excitations of frequency, amplitude, and mean value as reported in [5]. Accordingly, this study proposed a new method to predicting self-calibration in a pressure sensor using Levenberg Marquardt Back Propagation Artificial Neural Network (LMBP-ANN) model. The proposed method was achieved using one-hour calibration data set of pressure sensor in real time. The collected measurements were shown to lead to lack of linearity and fluctuation in output pressure sensor over time which should be compensated. The proposed method was validated by comparing the output force of PSCA with the experimental target force from load cell (reference). This work shows that the Proposed model exhibited a remarkable performance than traditional methods with a max MSE of 0.17325 and R value over 0.99 for the total response of training, testing and validation. The model was tested using an untrained input data set in order to verify the Proposed model’s capability for implementing a self-calibration algorithm. We find that, the Proposed LMBP-ANN model for self-calibration purposes is able to successfully predict the desired pressure over time, even the uncertain behaviour of the pressure sensors due to its material creep. Developing an intelligent wearable robotic hand glove system based on PSCA with self-calibration feature of grasping mechanism is implemented. Then grasping sampled objects (plastic bottle and sponge ball) with different weights based on the developed robotic hand glove system were successfully performed. The results proved that the PSCA has the ability to successfully and accurately estimate the desired grasping forces in real time even the decrease and fluctuation of the forces pattern in sensor response. Afterwards, the PSCA was implemented and tested in real time based on MCU and software (MATLAB). For validity and performance purpose of PSCA, the MSE and MAPE are calculated of 0.30 and 1.21% base on MCU and 0.08 and 0.6 base on MATLAB respectively. Overall, the PSCA presented here ensure that the problems of hysteresis, variation in gain and lack of linearity over time have overcome. Furthermore, we have obtained accurate measurements of grasping mechanism. This provides a useful methodology for the user to evaluate the performance of any measurement system in a real-time environment.

Text FK 2019 79 - IR.pdf Download (1MB)

Actions (login required)

View Item