The stochastic approximation algorithm (SAA), starting from the pioneer work by Robbins and Monro in 1950s, has been successfully applied in systems and control, statistics, machine learning, and so forth. In this pap...
详细信息
The stochastic approximation algorithm (SAA), starting from the pioneer work by Robbins and Monro in 1950s, has been successfully applied in systems and control, statistics, machine learning, and so forth. In this paper, we will review the development of SAA in China, to be specific, the stochastic approximation algorithm with expanding truncations (SAAWET) developed by Han-Fu Chen and his colleagues during the past 35 years. We first review the historical development for the centralizedalgorithm including the probabilistic method (PM) and the ordinary differential equation (ODE) method for SAA and the trajectory-subsequence method for SAAWET. Then, we will give an application example of SAAWET to the recursive principal component analysis. We will also introduce the recent progress on SAAWET in a networked and distributed setting, named the distributed SAAWET (DSAAWET).
Given a set of events and a set of robots, the dispatch problem is to allocate one robot for each event to visit it. In a single round, each robot may be allowed to visit only one event (matching dispatch), or several...
详细信息
Given a set of events and a set of robots, the dispatch problem is to allocate one robot for each event to visit it. In a single round, each robot may be allowed to visit only one event (matching dispatch), or several events in a sequence (sequence dispatch). In a distributed setting, each event is discovered by a sensor and reported to a robot. Here, we present novel algorithms aimed at overcoming the shortcomings of several existing solutions. We propose pairwise distance based matching algorithm (PDM) to eliminate long edges by pairwise exchanges between matching pairs. Our sequence dispatch algorithm (SQD) iteratively finds the closest event-robot pair, includes the event in dispatch schedule of the selected robot and updates its position accordingly. When event-robot distances are multiplied by robot resistance (inverse of the remaining energy), the corresponding energy-balanced variants are obtained. We also present generalizations which handle multiple visits and timing constraints. Our localized algorithm MAD is based on information mesh infrastructure and local auctions within the robot network for obtaining the optimal dispatch schedule for each robot. The simulations conducted confirm the advantages of our algorithms over other existing solutions in terms of average robot-event distance and lifetime.
Localization is an active field of research in wireless sensor networks WSNs. The information of exact physical location of the sensor nodes in WSNs is useful for various application e. g. intrusion detection, target ...
详细信息
ISBN:
(纸本)9781467330947;9781467330961
Localization is an active field of research in wireless sensor networks WSNs. The information of exact physical location of the sensor nodes in WSNs is useful for various application e. g. intrusion detection, target tracking, environmental monitoring and network services etc. In this paper we present the classification and comparative study of localization algorithms. The goal of our consideration is to analyze, how these localization algorithms work in order to increase the life span of network nodes in harsh environments like oil fields, gas fields, forests, chemical factories and underground mines etc. and how to find the position of mobile node with distributed, Range-based and Beacon-based Localization technique in harsh environments. Furthermore this paper also highlights some issues experienced by these localization techniques.
暂无评论