Animals have certain cognitive competence about the environment so they can correct their navigation errors. Inspired by excellent navigational behavior of animals, this paper proposes a brain-like scheme to improve accuracy and intelligence Micro-Electro-Mechanical System based Inertial Navigation Systems (MEMS-INS). The proposed employs vision acquire external perception information as an absolute reference position errors INS, which is established analyzing error correction mechanism rat...

Automatic segmentation of salient objects in real-world images has gained increasing interests owing to its popularity diverse applications, such as autonomous driving, medical diagnosis, aviation security, and underwater surveillance. In this research, we propose Firefly Algorithm (FA)-enhanced evolving ensemble deep networks for semantic visual saliency prediction. An improved FA model is proposed optimize network hyper-parameters. Specifically, it employs mutation operators a neighbouring...

The recent breakthrough of deep learning based generative models has led to the escalated generation photo-realistic synthetic videos with significant visual quality. Automated reliable detection such forged requires extraction fine-grained discriminative spatial-temporal cues. To tackle challenges, we propose weighted and evolving ensemble comprising 3D Convolutional Neural Networks (CNNs) CNN-Recurrent (RNNs) Particle Swarm Optimization (PSO) network topology hyper-parameter optimization...

High-G MEMS accelerometers have been widely used in monitoring natural disasters and other fields. In order to improve the performance of accelerometers, a denoising method based on combination empirical mode decomposition (EMD) wavelet threshold is proposed. Firstly, EMD performed output main accelerometer obtain intrinsic function (IMF). Then, continuous mean square error rule find energy cut-off point, then corresponding high frequency IMF component denoised by threshold. Finally,...

Autonomous underwater vehicle (AUV) has been employed in oceanography applications based on a reliable navigation. The complex environment leads to more velocity measurement errors of AUV, so it is difficult determine the accurate navigation and positioning information. To solve problem, novel variational Bayesian-based filter for inaccurate input (VBFII) proposed state information under marine condition input. Firstly, velocities are assumed follow Gaussian distribution, which can better...

Accurate estimation of vehicle sideslip angle and attitude angles are essential for the safety control lateral behaviour driving performance. In this article, variation wheels cornering stiffness is considered addressed by introducing a recursive least squares approach. Based on nonlinear dynamic model investigated coupling effect between longitudinal velocity, an optimized moving horizon estimator proposed to obtain angle, in which iteration decent algorithm integrated. Furthermore,...

In the complex underwater environment, performance of microelectro-mechanical system sensors is degraded sharply and errors will become much larger. Especially when magnetic sensor disturbed by external interference, measurements unobservable so that navigation information estimated erroneously. To solve this problem, paper proposes a novel method fusing Quasi-Newton cubature Kalman filter (QNCKF). This takes full advantage computation efficiency estimation accuracy CKF in case nonlinearity....

The integration of micro-electro-mechanical system–inertial navigation systems (MEMS-INSs) with other autonomous sensors, such as polarization compasses (PCs) and geomagnetic compasses, has been widely used to improve the accuracy reliability vehicles in Internet Things (IoT) applications. However, a MEMS-INS/PC integrated system suffers from cumulative errors time-varying measurement noise covariance unknown, complex occlusion, dynamic environments. To overcome these problems system’s...

In underwater environments, the presence of multipath effects can cause measurement outliers in acoustic sensors, leading to reduced estimation accuracy for integrated navigation. To address this issue, paper proposes a sliding window variational Kalman filter based on Student’s t-distribution (SWVKF-ST) improve state accuracy. First, method makes use model heavy-tailed noise and adopts inverse Wishart distribution as prior covariance, thereby enhancing robustness against distributions. On...

The ocean environment is complex and changeable because of all kinds noise interferences, such as salt cliffs, ships around other electromagnetic so the measurement information prone to be lost. It difficult describe acquisition probability information. In this paper, a continuous discrete variational Bayesian filter (CD VBF) proposed solve problems heavy tailed noises random loss for state estimation. (VB) approach can effectively estimate vector, scale matrices, degree freedom (DOF)...

In order to solve the problem of strap-down inertial navigation system (SINS)/Doppler velocity log (DVL)/ultrashort baseline (USBL) interference in complex underwater environment, an asynchronous sequential robust filter method is proposed this article. The USBL original information azimuth, slant range and altitude are introduced as measurement information. large amount computation caused by high dimension multisensor fusion solved introducing technology. Meanwhile, a Kalman (KF) algorithm...

High-accuracy attitude estimation plays an important role in gliding with long endurance for underwater glider. Because microelectromechanical system (MEMS) inertial sensors have advantages, including small size and low power consumption, they are used as main to determine navigation information. However, the complicated harsh environment, performances of MEMS degrade errors will become larger. Moreover, acceleration or deceleration while gliders going up down, sudden vibration due...

The uncertainty, complexity, and variability of the marine environment inevitably lead to a change in measurement error resulting erroneous estimation navigation information. To solve this problem, paper proposes novel method integrating square-root cubature Kalman filter (SCKF) with expectation-maximization (EM) algorithm. proposed new SCKF (NSCKF) algorithm makes better use advantages EM online performance NSCKF is verified theoretically evaluated by experiments. results indicate that can...

In order to solve the problems of rapid path planning and effective obstacle avoidance for autonomous underwater vehicle (AUV) in 2D environment, this paper proposes a algorithm based on reinforcement learning mechanism particle swarm optimization (RMPSO). A feedback is embedded into (PSO) by using proposed RMPSO improve convergence speed adaptive ability PSO. Then, integrates velocity synthesis method with Bezier curve eliminate influence ocean currents save energy AUV. Finally, developed...

For the strap-down inertial navigation system (SINS)/Doppler velocity log (DVL) integrated system, installation error angle and DVL scale factor are key factors affecting accuracy of system. Aiming at this problem, based on traditional Kalman filter calibration method, article discusses estimation parameters under different models. On one hand, according to observability state variables, an improved model is proposed, corresponding equation measurement given. other 3-D information abandoned,...

Human action recognition can be applied in a multitude of fully diversified domains such as active large-scale surveillance, threat detection, personal safety hazardous environments, human assistance, health monitoring, and intelligent robotics. Owing to its high demands real-world applications, it has drawn significant attention. In this research, we propose hybrid deep neural networks, i.e. Convolutional Long Short-Term Memory (ConvLSTM) Networks, Long-term Recurrent Networks (LRCN), for...