This paper presents an effective path decision method to find a two-dimension path that is time optimal and obstacIe free using parametrization in Complex domain. Such a path is used only for moving forward vehicIes with constant velocity and non-holonomic kinematic constraints such as VA V (Vnmanned Aerial Vehicles). Each start and goal point of a vehicIe in a planar domain with randomly selected heading angle demonstrates this problem, wh ich is sometimes referred to as Dubins's circIe problems. Due to the nonlinearity of problems, researchers have recommended common approaches, incIuding optimal control using cost functions and numerical solutions of nonlinear functions. These approaches do involve additional restrictions that previous research has highlighted. However; the author here suggests a simpler, more effective and more generalized method, based on problems identified in complex domain using parameterization.Keywords-parametrization; complex domain; time optimal; unmanned aerial vehicle I. INTRODUCTlONPath-planning for the forward motion of non-holonomic vehicles is somewhat different from holonomic systems such as holonomic manipulators and ornni-directional mobile robots; in other words, simple control and decision schemes cannot be directly applied with success to vehicles with nonholonomic constraints. For example, two-wheel differential mobile robots or UA Vs work this way, because constraints such as minimum turning radius or curvature induced from constant velocity cannot be integrated to position level. In addition, its movement is only forward in the cases ofUA Vs. This geometrie problem is so called "piano-mover problem" in a configuration space such as a two-dimensional space. In this paper, we confine and solve this path-planning problem with simple kinematics of UA Vs using parametrization in Complex domain . B ACKGROUTNDPath-planning problems of non-holonomic vehicles could be categorized into three major groups according to approaches taken to solve them. The first one is calculusbased; the second, graph-based, and the last, uses numerical methods. The calculus-based path-planning method requires severe computationalloads [1]- [3]. The path-planning based
Abstmct-Background subtraction in an emcient technique in vision-based tratRc monitoring, it scgnlents t h e moving vohieic from t h e video sequences by comparing t h e incomframe to the model of background BCCne, The presented work i s a simple approach of adaptive background in which t h e Short Term Memory (STM) and L a w Term Memorv iLTMI are introduced t o construct t h e wlrolo baekto achieve high sensitivity in the detection or inoving vehicles while keeping low false alarm rate. Motion detectiorl is the common technique used fur vehicle det,ection in vidco streams mainly iiicluding three approaches: fmme differenciny, background subtructiori and optical flow. _ I Iground rncmary. T h e color cue is used t o build the model of pixel, U* and Y * chrominancy components arc carefully selected from modifled L'u'v' color space, they are perccptually uniform such t h a t color difference could be measured properly. Furthermore, object shadows are suppressed because thc luminancy effects are removed. A simple prot+ type cell is deflned t o characterize t h e background scene by i t s 'circular inRuence flald'. T h e match of prototype cell is measured by the Euclidean distance between t h e incoming pixel and prototype ecil. T h e m o s t recent prototype rolls are stored in S T M , they a d a p t quickly for t h e variations of background scene, b u t false detections easily occurs when t h e background has t h e high frequency variations. In LTM, prototype cells store t h e stable representation of t h e background scene, which are able to reduce the computation of S T M updatirig. T h e adaptive learning procedure is carried out in h o t b S T M a n d LTM, it ir able to deal with t h e scene ehangcs. This background model i s evaluated by t h e traffic video Stream, experimental results show t h a t t h e proposed approach i s feasible for t h e traffic monitoring. Kqmmrd-VideoSurveillance, Background Subtraction, Calor Pcrccption, Perceptual Uniformity, Short Term Memory, Long T e r m Memory, Adaptive Learning. I. IATRODUCTiOHRAFFIC monitoring is essential to improve the effb T . cieiicy of transportation system by providing the u s e id iniormution related to the trafic flow. Visioii-based traffic monitoring is distinguished from other traffic monitoring techniques for its rich pixel-based visual information about the traflir scene. The nranagemeut o i traflir flow is greatly benefited from it, because almost all important t r a f k features can be extracted from the imagc by veliiclc dclection, incident detection and vehicle queue measurement. In pace with the steady growth oi computation speed arid ~rrcmory capacity, vision-based traffic monitoring has provided a feasible solution for traffic monitoring using low cost video Cameras with increased handmidt.lis of video capturing, its research has become an increasingIn vision-bascd traffic monitoring, visiw systeirt contirruously pt:rforrns the task of video surveillancc i n all weath- PIS.\Jel~icle detection is the first shge; which is requiredYan Ciiye H...
-This paper presents a voice-activity detection (VAD) method for sound sequences with various SNRs. For real-time VAD applications, it is inadequate to employ a post-processing for the removal of burst clippings from the VAD output decision. To tackle this problem, building on the bilevel hidden Markov model, for which a state layer is inserted into a typical hidden Markov model (HMM), we formulated a robust method for VAD not requiring any additional post-processing. In the method, a forward-inference-ratio test was devised to detect the speech endpoints and Mel-frequency cepstral coefficients (MFCC) were used as the features. Our experiment results show that, regarding different SNRs, the performance of the proposed approach is more outstanding than those of the conventional methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.