In this paper we present a top-down methodology aimed at evaluating the scalability of Spiking Neural Network (SNN) simulations on massively many-core and densely interconnected platforms. Spiking neural networks mimi...
详细信息
In this paper we present a top-down methodology aimed at evaluating the scalability of Spiking Neural Network (SNN) simulations on massively many-core and densely interconnected platforms. Spiking neural networks mimic brain activity by emulating spikes sent among neurons populations. Many-core platforms are emerging computing targets to achieve real-time simulation SNN. Neurons are mapped to parallel cores and spikes are sent over the on-chip and off-chip network. However, due to the heterogeneity and complexity of neuron population activity, achieving an efficient exploitation of platforms resources is a challenging problem, often impacting simulation reliability and limiting the biological network size. To address this challenge, the proposed methodology, based on customized SNN configurations, extracts detailed profiling information about network usage of on-chip and off-chip resources. We first show the results of the application of our methodology to the SpiNNaker neuromorphic many-core platform. Then, we demonstrate how the profiling information can be used to improve the reliability of biologically plausible SNN simulations.
The droop controller is widely used for the parallel operation of inverters. However, the conventional droop controller has a trade-off between the power sharing and the regulations of the output voltage amplitude and...
详细信息
This paper presents an interface system display which is conceived to improve pilot situation awareness with respect to a flight envelope protection system developed for a mid-sized transport aircraft. The new display...
详细信息
Many robot planning approaches use cell decomposition methods for finding a sequence of regions that the robot should traverse. When a fully-actuated robot is used, the robot reference trajectory is simply constructed...
详细信息
Many robot planning approaches use cell decomposition methods for finding a sequence of regions that the robot should traverse. When a fully-actuated robot is used, the robot reference trajectory is simply constructed by linking through line segments the middle points of common facets of successive traversed cells. This paper improves this approach by proposing three different optimizations that yield the waypoints through which the robot piecewise linear route is passing. The optimizations use different metrics defined as sums of norms for the linear segments that compose the route. The norms L 1 , L 2 squared and L ∞ are used and standard optimization problems result in each case. Examples are included for showing the usefulness of these optimizations, since shorter routes can be obtained under a negligible computational overhead.
作者:
Iocchi, LucaDept. of Computer
Control and Management Engineering Sapienza University of Rome Via Ariosto 25 Rome00185 Italy
Learning maps from sensor data has been addressed since more than two decades by Simultaneous Localization and Mapping (SLAM) systems. Modern state-of-the-art SLAM approaches exhibit excellent performances and are abl...
详细信息
Point cloud registration is an essential part for many robotics applications and this problem is usually addressed using some of the existing variants of the Iterative Closest Point (ICP) algorithm. In this paper we p...
详细信息
Vehicular Ad Hoc Networks (VANETs) represent promising technologies for comfort driving and entertainment applications which rely heavily on the data downloading. Due to the rapid change of network topology and interm...
详细信息
Vehicular Ad Hoc Networks (VANETs) represent promising technologies for comfort driving and entertainment applications which rely heavily on the data downloading. Due to the rapid change of network topology and intermittent connection, it is a big challenge to satisfy the download requirements from multi-vehicles at the same time. This paper proposes a Bus-Based Content Downloading (BBCD) which aims to maximize the volume of downloaded data from bus to vehicles while the download opportunity fairness of each vehicle is taken into consideration. By predicting the number of buses which the vehicle would encounter in its future path and estimating the connection duration that the vehicle stays in the coverage of bus, the proposed BBCD schedules the download service for vehicles slot by slot such that the volume of downloaded data can be guaranteed while achieving the download opportunity fairness. The effectiveness of the proposed algorithm is evaluated by extensive simulations.
In the field of nuclear engineering, national colleges have been carrying out important roles with about 10 % of graduates obtaining jobs in nuclear related companies. Although national colleges do not have special co...
详细信息
In the field of nuclear engineering, national colleges have been carrying out important roles with about 10 % of graduates obtaining jobs in nuclear related companies. Although national colleges do not have special courses and classes for nuclear engineering, we have been trying to educate our students in cooperation with universities and industries through the activities of the Japan Nuclear Human Resource Development Network. With aide from the Ministry of Education, Culture, Sports, Science and Technology, we have been developing a short term curriculum for nuclear education consisting of lectures, experiments and facility tours. Moreover, we have developed a textbook for nuclear engineering basics and texts for experiments. In order to let the students understand the background of radiation, radiation dose measurements have been conducted by the students of more than 33 colleges. In the fiscal year of 2015, a new trial of remote lecture connecting 51 national colleges began. Through these education programs, we aim to help our students to understand radiation and nuclear engineering correctly, and to have their own opinions on this important matter.
The frequency domain analysis of dynamically perturbed Higher Order Sliding Mode (HOSM) systems is tackled using Describing Function (DF) and Harmonic Balance (HB) techniques. The goal of this analysis is to study pos...
详细信息
ISBN:
(纸本)9781479978878
The frequency domain analysis of dynamically perturbed Higher Order Sliding Mode (HOSM) systems is tackled using Describing Function (DF) and Harmonic Balance (HB) techniques. The goal of this analysis is to study possible limit cycles in such systems. DFs of Nested 3rd and 4th order algorithms are obtained for the first time. Then, HB equation is used to analyze the real sliding motion in the HOSM system, where the sliding set converges to a limit cycle. Chattering (limit cycle) in the HOSM systems is studied, and the chattering parameters (amplitude and frequency) are computed. A definition of Tolerance Limits, which characterizes the acceptable performance in the real HOSM system, is applied to verify if the chattering parameters fit the amplitude and frequency limits. Next, Performance Phase Margin and Performance Gain Margin definitions, which give the metrics for robustness of real HOSM to unmodeled dynamics, are applied to assess the robustness of the limit cycle emerged in the real HOSM system. Examples and simulations that validate the obtained results are presented.
Intermittent control uses open-loop control punctuated with feedback at times determined by error-driven events. The open-loop trajectories are based on an underlying closed-loop strategy and are generated by a system...
详细信息
ISBN:
(纸本)9781479977970
Intermittent control uses open-loop control punctuated with feedback at times determined by error-driven events. The open-loop trajectories are based on an underlying closed-loop strategy and are generated by a system-matched hold. The single-loop event-driven intermittent control method is extended to the multi-loop decentralised control situation. This decentralised intermittent controller is based on an underlying continuous-time decentralised design which is suitable for systems with both input and state interactions. This extension is achieved by using local models of the remote interacting subsystems. These models are used for control signal generation and they are only updated with remote information at discrete event-driven sample times thus reducing information flow. The approach is illustrated using a simulation of a live-pool irrigation channel model previously examined in the literature.
暂无评论