1558-1748 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information. This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JSEN.2019.2941273, IEEE Sensors Journal 1 MEMS Based IMU Drift Minimization: Sage Husa Adaptive Robust Kalman Filtering Mundla Narasimhappa, Arun D. Mahindrakar, Vitor C. Guizilini, Marco H Terra and Samrat L Sabat Abstract—The Attitude Heading Reference System (AHRS) has been widely used to provide the position and orientation of a rigid body. A low cost MEMS based inertial sensor measurement unit (IMU) is a core device in AHRS. To improve the AHRS system performance, there is a need to develop (i) stochastic IMU error models and (ii) random noise minimization techniques. In this paper, we modify the Sage-Husa Adaptive Kalman Filter (SHAKF) to incorporate time-varying noise estimator and robustifier, termed as Modified Sage-Husa Adaptive Robust Kalman Filter (MSHARKF). In the proposed algorithm, a three segment approach is used to evaluate the adaptive scale factor followed by the learning statistics. The scale factor is iteratively updated in the MSHARKF equations. In addition, angle random walk (ARW) and bias instability (BI) errors are represented by state-space models. The proposed algorithm is applied to restrain the drift error and random noise in the MEMS IMUs signals. The perfor- mance of this algorithm is analyzed using Allan variance (AV) analysis for static signals whereas the Root Mean Square Error (RMSE) values are evaluated for dynamic signals. Experimental results demonstrate the effectiveness of MSHARKF in reducing the drift and random noise in static and dynamic conditions as compared with other ex- isting algorithms. Finally, we present sufficient conditions for convergence proof of the MSHARKF algorithm. Index Terms—MEMS IMU, bias drift, Sage-Husa robust Kalman Filter, Sliding Average Allan Variance. I. I NTRODUCTION Mundla Narasimhappa and Arun D. Mahindrakar are with the Department of Electrical Engineering, Indian Institute of Technol- ogy Madras, Chennai-600036, India, mr.narasimha08@gmail.com, arun dm@iitm.ac.in. Vitor C. Guizilini is with the School of Information Technologies, University of Sydney, Sydney, Australia, vitor.guizilini@sydney.edu.au Marco H Terra is with the School of Electrical Engineering, University of S˜ ao Paulo (USP), S˜ ao Carlos, SP-Brazil, terrac- rob@gmail.com Samrat L Sabat is with the Centre for Advanced Studies in Elec- tronics Science and Technology (CASEST), University of Hyderabad, Hyderabad-50046, India, slssp@uohyd.ernet.in I N modern navigation, the Attitude Heading Ref- erence System (AHRS) is an integral and im- portant part of navigation and has been widely used for providing attitude of a rigid body [1]. In general, AHRS comprises an inertial measure- ment unit (IMU) and a processing computer. The IMU includes three single-axis gyroscopes and ac- celerometers, which can measure the body angular velocity and inertial accelerations [2], [3]. Over recent years, there has been a surge in the use of micro electro mechanical system (MEMS) sensor based IMU in view of its small size, low cost and low power consumption. Due to these advantages, the MEMS IMU is applicable in robotics, driver- less cars, unmanned underwater and aerial vehicles and other fields [4]. The performance of the AHRS depends on the quality of IMU sensors. Since the MEMS based IMU operates for long duration, the stochastic errors due to external disturbances and internal device operation get accumulated in the measurements and lead to their drift. In addition, the dynamic performance of the MEMS based IMU degrades due to temperature, pressure and mechan- ical stresses [5]. In order to obtain relative attitude information accurately, the IMU errors must be corrected using sensor fusion, which is the focus of current research [6]. The performance of the AHRS can be restricted by the MEMS IMU errors, that can be characterized as (a) deterministic and (b) stochastic. In general, the bias, scale factor and mis-alignments errors are of deterministic nature, whereas the angle random walk (ARW), quantization noise (QN), bias insta- bility (BI), rate ramp (RR) and rate random walk (RRW) are stochastic errors [7]. The deterministic errors can be eliminated easily by using calibration techniques in the laboratory environment. However, the elimination of stochastic errors is not possible, because of its randomness. Moreover, these errors are present in position, velocity and attitude infor-