This paper presents a 3D frontier-based exploration tool named 3D-FBET. Our runs onboard the MAV equipped with sensor. The map of environment explored is constructed incrementally from two consecutive point clouds obtained. Considering computation and memory limitations MAVs, OctoMap utilized to represent models. A novel approach designed extract frontiers OctoMap. Different existing extraction method, only state-changed space in processed each iteration. We implement our on top well-known...

This work reports a novel and standard metric of plot to evaluate the electrostatic generator's output performance under air breakdown limitation proposes strategy extend for achieving ultrahigh power generation.

Abstract Solar panels often suffer from dust accumulation, significantly reducing their output, especially in desert regions where many of the world's largest solar plants are located. Here, an autonomous removal system for panels, powered by a wind‐driven rotary electret generator is proposed. The applies high voltage between one panel's output electrode and upper mesh to generate strong electrostatic field. It discovered that particles on insulative glass cover panel can be charged under...

Geographic routing has been studied as an attractive approach due to its simplicity and scalability properties in for wireless sensor networks. The most influential geographic protocol is GPSR (Greedy Perimeter Stateless Routing) many protocols are based on greedy forwarding routing. However, large overhead, less reliability long link average end-to-end delay still the problems this type of protocol. In paper, we start with another perspective information propose a novel protocol, RGRP...

Electrostatic generators as a kind of effective energy harvesters have attracted intensive attention. However, the output is highly dependent on charge density. Here, we demonstrate an all-in-one rotary electrostatic nanogenerator based pumping and voltage multiplying strategy (CV-ESG), which achieves high power in SF6 atmosphere. CV-ESG integrates electret generator, main stabilization circuit pair rotator stator. We analyze breakdown effect its influence insulating layer covered electrodes...

In recent years, weigh-in-motion systems based on embedded sensor networks have received a lot of attention. However, how to improve the accuracy multi-sensor (WIM) while keeping costs low remains challenge. this paper, numerical simulation method is presented analyze relationship between location and static weight estimation. The finite element model WIM system developed, which consists three parts: pavement model, moving load two types models. Analysis results shows that ability sensing...

A multimodal part face recognition method based on principal component analysis (MMP-PCA) is presented. Five kinds of parts are dealt with by PCA to obtain eigenface, eigeneyebrow, eigeneye, eigennose and eigenmouth. Thirty-one different models created using combinations parts. The rates 17 among them given.

This paper proposes a fast inter prediction mode decision method for H.264. By pre-encoding downsampled small image, the candidate block modes can be reduced to subset. The simulation result shows that our algorithm achieve up 50% complexity reduction with less than 0.2dB PSNR decrease.

This work presents a new kind of reference frame, i.e., the "background frame", to improve accuracy motion compensation in video compression. When an object moves scene sequence, certain part background is overlapped at first. After several frames, it begins reappear. In frames reappears, blocks contained may be poorly predicted by neighboring which have been previous frames. To solve this problem, frame" constructed, based on that kept unchanged number continuous The frame free from...

This work proposes a fast inter prediction mode decision method for H.264. By pre-encoding down-sampled small image, the candidate block modes can be reduced to subset. The experimental result shows that our algorithm reduce nearly 50% of encoding time with PSNR decrease less than 0.2dB.

Electrostatic generators show great potential for powering widely distributed electronic devices in Internet of Things (IoT) applications. However, a critical issue limiting such is their high impedance mismatch when coupled to electronics, which results very low energy utilization efficiency. Here, we present high-performance management unit (EMU) based on spark-switch tube and buck converter with an RF inductor. By optimizing the elements parameters EMU, maximum direct current output power...

Abstract In traditional food industry, the assessment of mushy materials plays an important role in high‐quality production, which still relies heavily on human tactile perception. this work, to address issue for enhancing production efficiency, a flexible electret sensor that can mimic expert touch is developed. The consists pair electrodes, microstructural spacer, and pre‐charged electret. Benefiting from electrostatic induction‐based working mechanism, attains high sensitivity ideal...

The demand of transcoding MPEG-2 to H.264 will be raised as the promises higher coding efficiency over previous video compression standards, while MEPG-2 has been widely adopted in many existing systems. This paper presents a rate control scheme for transcoder with consideration peak signal-to-noise ratio (PSNR) fluctuation and re-quantized quality loss reduction. Our main idea is reuse information extracted from incoming stream efficiently possible. An analytic formula constructed set an...

Multirotor unmanned aerial vehicles (MAVs) with the capability of interaction their surroundings have recently received increasing attention. However, rapid change parameters during flight present a challenge for stable control MAVs. In this paper, baseline controller quadrotor is developed based on classical model reference adaptive (MRAC) and an identifier introduced to cope uncertainty in parameters. Specifically, recursive least squares method forgetting factor (FFRLS) used design...

In recent years, many clustering algorithms have been proposed. Among them, LEACH is the most famous one. However, in LEACH, within each 1/p rounds, once a node has selected as cluster-head (CH), its threshold will be set to 0, and thus it lose chance participate selection, even if still enough energy. this paper, we present novel selection algorithm. Instead of changing 0 directly, proposed algorithm adjusts gradually according roles they played last round, so more nodes could opportunity...

Optical flow plays an important role in vision-based navigation. A lot of optical based methods have been developed for autonomous robots. In this paper, we evaluate obstacle avoidance on synthetic and real-world scenes. Specifically, the balance strategy is chosen five representative algorithms are used. The new metrics performance introduced. Our experiments demonstrate effectiveness avoidance. Furthermore, some factors contributing to ability studied.

Since multi-radio multi-channel wireless mesh networks are widely brought into use, the channel assignment scheme plays a critical role in determining performance of networks. Nowadays many schemes have been proposed to enhance However, most existing assume that all interference links induce same level on target link, which will lead poor network as assumption deviates from actual situation. In this paper, we propose novel clique-based clustering (CCCA) takes different levels interfering...

Network addressing is a traditional problem in communications network. IP which can fully reflect the characteristics of Wireless Sensor will provide good support for design network routing and management protocols. The location information important most WSN application be used to simplify policy, also supported by IPv6. In this paper, we propose location-based protocol. We assume that nodes know their current physical location; thus, exploited facilitate generation Location-based address....