A hybrid learning-based stochastic noise eliminating method with attention-Conv-LSTM network for low-cost MEMS gyroscope.

Low-cost inertial measurement units (IMUs) based on microelectromechanical system (MEMS) have been widely used in self-localization for autonomous robots due to their small size and low power consumption. However, the low-cost MEMS IMUs often suffer from complex, non-linear, time-varying noise and errors. In order to improve the low-cost MEMS IMU gyroscope performance, a data-driven denoising method is proposed in this paper to reduce stochastic errors. Specifically, an attention-based learning architecture of convolutional neural network (CNN) and long short-term memory (LSTM) is employed to extract the local features and learn the temporal correlation from the MEMS IMU gyroscope raw signals. The attention mechanism is appropriately designed to distinguish the importance of the features at different times by automatically assigning different weights. Numerical real field, datasets and ablation experiments are performed to evaluate the effectiveness of the proposed algorithm. Compared to the raw gyroscope data, the experimental results demonstrate that the average errors of bias instability and angle random walk are reduced by 57.1 and 66.7%.

A Two-Stage Density-Aware Single Image Deraining Method.

Although advanced single image deraining methods have been proposed, one main challenge remains: the available methods usually perform well on specific rain patterns but can hardly deal with scenarios with dramatically different rain densities, especially when the impacts of rain streaks and the veiling effect caused by rain accumulation are heavily coupled. To tackle this challenge, we propose a two-stage density-aware single image deraining method with gated multi-scale feature fusion. In the first stage, a realistic physics model closer to real rain scenes is leveraged for initial deraining, and a network branch is also trained for rain density estimation to guide the subsequent refinement. The second stage of model-independent refinement is realized using conditional Generative Adversarial Network (cGAN), aiming to eliminate artifacts and improve the restoration quality. In particular, dilated convolutions are applied to extract rain features at multiple scales and gated feature fusion is exploited to better aggregate multi-level contextual information in both stages. Extensive experiments have been conducted on representative synthetic rain datasets and real rain scenes. Quantitative and qualitative results demonstrate the superiority of our method in terms of effectiveness and generalization ability, which outperforms the state-of-the-art.

Multi-UAV Area Coverage Based on Relative Localization: Algorithms and Optimal UAV Placement.

This paper is concerned with relative localization-based optimal area coverage placement using multiple unmanned aerial vehicles (UAVs). It is assumed that only one of the UAVs has its global position information before performing the area coverage task and that ranging measurements can be obtained among the UAVs by using ultra-wide band (UWB) sensors. In this case, multi-UAV relative localization and cooperative coverage control have to be run simultaneously, which is a quite challenging task. In this paper, we propose a single-landmark-based relative localization algorithm, combined with a distributed coverage control law. At the same time, the optimal multi-UAV placement problem was formulated as a quadratic programming problem by compromising between optimal relative localization and optimal coverage control and was solved by using Sequential Quadratic Programming (SQP) algorithms. Simulation results show that our proposed method can guarantee that a team of UAVs can efficiently localize themselves in a cooperative manner and, at the same time, complete the area coverage task.

Semantic Segmentation Leveraging Simultaneous Depth Estimation.

Semantic segmentation is one of the most widely studied problems in computer vision communities, which makes a great contribution to a variety of applications. A lot of learning-based approaches, such as Convolutional Neural Network (CNN), have made a vast contribution to this problem. While rich context information of the input images can be learned from multi-scale receptive fields by convolutions with deep layers, traditional CNNs have great difficulty in learning the geometrical relationship and distribution of objects in the RGB image due to the lack of depth information, which may lead to an inferior segmentation quality. To solve this problem, we propose a method that improves segmentation quality with depth estimation on RGB images. Specifically, we estimate depth information on RGB images via a depth estimation network, and then feed the depth map into the CNN which is able to guide the semantic segmentation. Furthermore, in order to parse the depth map and RGB images simultaneously, we construct a multi-branch encoder-decoder network and fuse the RGB and depth features step by step. Extensive experimental evaluation on four baseline networks demonstrates that our proposed method can enhance the segmentation quality considerably and obtain better performance compared to other segmentation networks.

Mesoporous carbon prepared by etching halloysite nanotubes (HNTs) with pyrrole as a precursor for a sulfur carrier of superior lithium-sulfur batteries.

Using pyrrole as a carbon precursor and halloysite nanotubes (HNT) as a templating agent, mesoporous carbon (MC) was prepared by template etching and combined with sulfur as a composite cathode for lithium-sulfur batteries. The mesoporous carbon/sulfur (MC/S) composite cathode exhibits a first cycle discharge specific capacity of 1355 mA h g-1 at 0.2C, and the utilization rate of active sulfur can reach 80.9%. Even after 500 cycles, the discharge specific capacity still remains at 496.9 mA h g-1 when tested at 0.5C. Furthermore, the hollow groove structure present in the MC/S electrode provides a large number of active sites for electrochemical reactions. The prepared MC/S composite cathode not only has a high discharge specific capacity and good cycle stability, but also increases the energy density of the lithium-sulfur battery. Therefore, this preparation process of MC is more conducive to practical application.