Shu Ting Goh

State-of-Charge Estimation of Lithium-Ion Battery Using Square Root Spherical Unscented Kalman Filter (Sqrt-UKFST) in Nanosatellite

State-of-charge (SOC) estimation is an important aspect for modern battery management system. Dynamic and closed loop model-based methods such as extended Kalman filter (EKF) have been extensively used in SOC estimation. However, the EKF suffers from drawbacks such as Jacobian matrix derivation and linearization accuracy. In this paper, a new SOC estimation method based on square root unscented Kalman filter using spherical transform (Sqrt-UKFST) with unit hyper sphere is proposed. The Sqrt-UKFST does not require the linearization for nonlinear model and uses fewer sigma points with spherical transform, which reduces the computational requirement of traditional unscented transform. The square root characteristics improve the numerical properties of state covariance. The proposed method has been experimentally validated. The results are compared with existing SOC estimation methods such as Coulomb counting, portable fuel gauge, and EKF. The proposed method has an absolute root mean square error (RMSE) of 1.42% and an absolute maximum error of 4.96%. These errors are lower than the other three methods. When compared with EKF, it represents 37% and 44% improvement in RMSE and maximum error respectively. Furthermore, the Sqrt-UKFST is less sensitive to parameter variation than EKF and it requires 32% less computational requirement than the regular UKF.

Gain-scheduled extended kalman filter for nanosatellite attitude determination system

The extended Kalman filter (EKF) has been widely used for attitude determination in various satellite missions. However, it requires an extensive computational power that is not suitable for nanosatellite application. This paper proposes a gain-scheduled EKF (GSEKF) to reduce the computational requirement in the nanosatellite attitude determination process. The proposed GSEKF eliminates the online recursive Kalman gain computation by analytically determining the Kalman gain based on the sensor parameters, such as the gyroscope noise variance, the quaternion variance, the observation matrix, and the satellite rotational speed. Two GSEKF Kalman gains for two satellite operating modes are presented: the Sun-pointing and nadir-pointing modes. The simulation and experimental results show that the proposed method has comparable attitude estimation accuracy to the conventional EKF. In addition, the proposed GSEKF reduces 86.29% and 89.45% of the computation load compared with the multiplicative EKF and Murrells version.

A low complexity Kalman filter for improving MEMS based gyroscope performance

Due to the mass, power and computational constraint of nano-satellite, high performance gyroscope is typically not available. In our previous nano-satellite mission named VELOX-I which was launched in June 2014, the sun tracking algorithm used an observer free quaternion error correction algorithm, but its performance is highly susceptible to the microelectromechanical systems (MEMS) based gyroscope noise. This paper presents a low complexity Kalman Alter (LCKF) based gyro drift filtering approach which utilizes the present states of the MEMS gyroscope. Low complexity was achieved by expressing the state transition matrix and the observation matrix into sparse matrices with non-zero diagonal elements. The performance of the proposed approach has been evaluated experimentally using the hardware of VELOX-I, a two axes rotary table and a sun simulator. Besides a 40.81% reduction in computational time, the experimental results show that the LCKF is capable of reducing the gyroscope noise in all axes. Overall, the experimental results agreed well with the simulation results and it has validated the improvement in the sun tracking performance.

Modeling and state of charge estimation of a lithium ion battery using unscented Kalman filter in a nanosatellite

State of charge (SOC) estimation is an essential part of battery management system. Dynamic and closed loop model-based methods such as extended Kalman filter (EKF) have been extensively used in SOC estimation. However, the EKF suffers from drawbacks such as requiring Jacobian matrix derivation and linearization accuracy. In this paper, a new SOC estimation method based on square root unscented Kalman filter (Sqrt-UKF) is proposed. With the proposed method, Jacobian matrix calculation is not needed and higher linearization order (2nd order) can be achieved. The proposed approach has been validated with the experimental data and has been benchmarked with the Coulomb counting method in terms of accuracy and performance. The experimental results have shown that the proposed method has a mean error of 1.19% and a maximum error of 4.96% and has performed better than the Coulomb counting method.