Nur Aqilah Othman

Examining the eigenvalues effect to the computational cost in mobile robot simultaneous localization and mapping

Covariance update in an extended Kalman filter-based simultaneous localization and mapping contributes the most to the computing time.Simplification of the covariance structure could decrease the computational cost and thereby shorten the processing time.Diagonalization of the covariance matrix by means of eigenvalues approach was introduced. One of the biggest factors that contributes to the computational cost of extended Kalman filter-based simultaneous localization and mapping is the computation of the covariance update. This results from the multiplications of the covariance matrix with other parameters along with the increment of its dimension, which is twice the number of landmarks. This study attempts to look for an optimal solution to decrease the computational complexity of the covariance matrix without compromising the accuracy of the state estimation through eigenvalue approach. This paper presents a study on the matrix-diagonalization technique, which is applied to the covariance matrix in extended Kalman filter-based simultaneous localization and mapping to simplify the multiplication process. The behavior of estimation and covariance were observed based on four case studies to analyze the performance of the proposed technique.

Sufficient condition for estimation in designing H ∞ filter-based slam

Extended Kalman filter (EKF) is often employed in determining the position of mobile robot and landmarks in simultaneous localization and mapping (SLAM). Nonetheless, there are some disadvantages of using EKF, namely, the requirement of Gaussian distribution for the state and noises, as well as the fact that it requires the smallest possible initial state covariance. This has led researchers to find alternative ways to mitigate the aforementioned shortcomings. Therefore, this study is conducted to propose an alternative technique by implementing filter in SLAM instead of EKF. In implementing filter in SLAM, the parameters of the filter especially γ need to be properly defined to prevent finite escape time problem. Hence, this study proposes a sufficient condition for the estimation purposes. Two distinct cases of initial state covariance are analysed considering an indoor environment to ensure the best solution for SLAM problem exists along with considerations of process and measurement noises statistical behaviour. If the prescribed conditions are not satisfied, then the estimation would exhibit unbounded uncertainties and consequently results in erroneous inference about the robot and landmarks estimation. The simulation results have shown the reliability and consistency as suggested by the theoretical analysis and our previous findings.

Analyzing the mobile robot localization performance in partially observable conditions

This paper deals with theoretical investigation of robot localization considering partial observability conditions. The problem is very important as most of the mobile robot applications are controllable but not observable due to some aspects. The paper investigate the importance of correlation to the mobile robot estimation through the technique of decorrelating some of the elements of state covariance of the updated state covariance. To demonstrate the effect of correlation, two cases of partial observability are examined which are the unstable and stable partial observability. The cases are build based on the configurations of the state covariance for mobile robot references when doing observations. Our preliminary results suggest that the system with stable partial observability shows very good and consistent estimation while the unstable case lead to inconsistency and erroneous estimation.

Impact of processing parameters on low-voltage power MOSFET threshold voltage considering defect generation

Power MOSFET is the most commonly used power device due to its low gate drive power and fast switching speed compared to the existing power bipolar transistor. In this study, the vital parameters in manufacturing the power VD-MOSFET, such as P-base and N-drift doping concentrations, thickness of N-drift region, and gate width are investigated to overcome MOSFET defects like NBTI and HCI; as degradation due to NBTI and HCI results shift in threshold voltage. The impact of those parameters on the threshold voltage is recorded and analysed in this study. The VD-MOSFET model used in this study is capable to withstand the breakdown voltage less than 30V. The model is simulated using 2D device and process simulation software from SILVACO; ATLAS and ATHENA. It is shown that, concentration of P-base doping is significant in determining threshold voltage and threshold voltage is proportional to the P-base doping concentration.

The effect of intermittent measurement in Simultaneous Localization and Mapping

Intermittent measurement is defined as a situation where a mobile robot experiences loss of measurement data during observation due to sensor failure or imperfection of the system. The impact of intermittent measurement on the Simultaneous Localization and Mapping (SLAM) of a mobile robot is the subject of this paper. The analysis is important since SLAM requires recursive measurement data update throughout the process. In this paper, the effect of intermittent measurement on the state error covariance matrix was analyzed on two basic conditions, which are when the mobile robot is stationary and when it is in motion. We have observed the impact on the determinant of covariance matrix. It is proven by our analysis that intermittent measurement could cause inaccurate estimation of robots position and might increase the state error covariance matrix.

A Solution to Partial Observability in Extended Kalman Filter Mobile Robot Navigation

Partial observability in EKF based mobile robot navigation is investigated in this paper to find a solution that can prevent erroneous estimation. By only considering certain landmarks in an environment, the computational cost in mobile robot can be reduced but with an increase of uncertainties to the system. This is known as suboptimal condition of the system. Fuzzy Logic technique is proposed to ensure that the estimation achieved desired performance even though some of the landmarks were excluded for references. The Fuzzy Logic is applied to the measurement innovation of Kalman Filter to correct the positions of both mobile robot and any observed landmarks during observations. The simulation results shown that the proposed method is capable to secure reliab le estimation results even a number of landmarks being excluded from Kalman Filter update process in both Gaussian and non -Gaussian noise conditions.

Investigation on intermittent observation in mobile robot localization with fuzzy logic technique

This paper deals with an analysis of intermittent observations for mobile robot localization with Fuzzy Logic approach. Mobile robot can easily lost its sight during environment observations due to several factors such as sensor faulty, and dynamic conditions. This can lead to erroneous estimation and the mobile robot become uncertain about its position. As a solution to this issue, this paper proposed a study on Fuzzy Logic technique to overcome such problem considering the Extended Kalman Filter(EKF) measurement innovation characteristic. The rules and fuzzy sets are designed such that it preserved good estimation whenever the relative angle and its relative distance measurements suddenly becomes larger than the previous measurements. The simulation results discusses two different cases observing the performance of the proposed technique. The results show that EKF with Fuzzy Logic technique is able to deal with intermittent observations if the design takes proper analysis and consideration on the measurement innovations.

Estimation behavior of intermittent measurement in EKF-based SLAM

Simultaneous localization and mapping (SLAM) of mobile robot by means of extended Kalman filter requires the availability of continuous data measurement along the process to ensure successful estimation of the state vector. Extended Kalman filter is a recursive algorithm that uses previous data in completing the iteration. Therefore the availability of measurement data is crucial in the estimation process. However, due to some failures of the sensors or network, measurement data may not be available at a certain period of time throughout the estimation process. Such situation is known as intermittent. In this paper, a theoretical study on the EKF-based SLAM with intermittent measurement is conducted to examine the estimation behavior of this nonlinear process. From the analysis, it is observed that the estimation of robot position is still possible even when the measurement data are unavailable. However the estimation possesses high uncertainties and produce abnormal covariance behavior. It has been proven that EKF is able to correct the estimation upon the availability of measurement data. Simulation results prove the consistency of the proposed analysis.