No abstract

No abstract

No abstract

No abstract

A nonlinear cost function is defined for field calibration of the accelerometer, using the rule that the norm of the measured vector in a static state is equal to the magnitude of the gravity vector. To solve this cost function, various optimization methods like Newton and Levenberg–Marquardt have been presented in different references. However, these methods are complicated, time-consuming, and require an initial value. This study presents a method that simplifies the cost function and obtains the error coefficients, including bias, scale factor, and non-orthogonality using the linear least-squares method which is simpler and faster than other optimization methods and does not need initial values. Also, the output of the low-cost MEMS accelerometer depends on temperature due to its silicon property. Thus, by finding the dependency of the error coefficients on temperature, they can be compensated. This paper models dependency of error coefficients on temperature using cubic spline interpolation and minimizes the temperature effect. Simulation results of MATLAB and the proposed field calibration method and temperature compensation on the low-cost MPU6050 sensor show its good performance.