Adaptive Kalman filter for MEMS IMU data fusion using enhanced covariance scaling

MEMS(micro-electro-mechanical-system)IMU(inertial measurement unit)sensors are characteristically noisy and this presents a serious problem to their effective use.The Kalman filter assumes zero-mean Gaussian process and measurement noise variables,and then recursively computes optimal state estimates.However,establishing the exact noise statistics is a non-trivial task.Additionally,this noise often varies widely in operation.Addressing this challenge is the focus of adaptive Kalman filtering techniques.In the covariance scaling method,the process and measurement noise covariance matrices Q and R are uniformly scaled by a scalar-quantity attenuating window.This study proposes a new approach where individual elements of Q and R are scaled element-wise to ensure more granular adaptation of noise components and hence improve accuracy.In addition,the scaling is performed over a smoothly decreasing window to balance aggressiveness of response and stability in steady state.Experimental results show that the root mean square errors for both pith and roll axes are significantly reduced compared to the conventional noise adaptation method,albeit at a slightly higher computational cost.Specifically,the root mean square pitch errors are 1.1∘under acceleration and 2.1∘under rotation,which are significantly less than the corresponding errors of the adaptive complementary filter and conventional covariance scaling-based adaptive Kalman filter tested under the same conditions.

A Survey of Synthetic Data Augmentation Methods in Machine Vision

The standard approach to tackling computer vision problems is to train deep convolutional neural network(CNN)models using large-scale image datasets that are representative of the target task.However,in many scenarios,it is often challenging to obtain sufficient image data for the target task.Data augmentation is a way to mitigate this challenge.A common practice is to explicitly transform existing images in desired ways to create the required volume and variability of training data necessary to achieve good generalization performance.In situations where data for the target domain are not accessible,a viable workaround is to synthesize training data from scratch,i.e.,synthetic data augmentation.This paper presents an extensive review of synthetic data augmentation techniques.It covers data synthesis approaches based on realistic 3D graphics modelling,neural style transfer(NST),differential neural rendering,and generative modelling using generative adversarial networks(GANs)and variational autoencoders(VAEs).For each of these classes of methods,we focus on the important data generation and augmentation techniques,general scope of application and specific use-cases,as well as existing limitations and possible workarounds.Additionally,we provide a summary of common synthetic datasets for training computer vision models,highlighting the main features,application domains and supported tasks.Finally,we discuss the effectiveness of synthetic data augmentation methods.Since this is the first paper to explore synthetic data augmentation methods in great detail,we are hoping to equip readers with the necessary background information and in-depth knowledge of existing methods and their attendant issues.