We consider deterministic nonlinear discrete-time systems whose inputs are generated by policy iteration (PI) for undiscounted cost functions. We first assume that PI is recursively feasible, in the sense that the opt...
详细信息
We consider deterministic nonlinear discrete-time systems whose inputs are generated by policy iteration (PI) for undiscounted cost functions. We first assume that PI is recursively feasible, in the sense that the optimization problems solved at each iteration admit a solution. In this case, we provide novel conditions to establish recursive robust stability properties for a general attractor, meaning that the policies generated at each iteration ensure a robust KL -stability property with respect to a general state measure. We then derive novel explicit bounds on the mismatch between the (suboptimal) value function returned by PI at each iteration and the optimal one. However, we show by a counterexample that PI may fail to be recursively feasible, disallowing the mentioned stability and near-optimality guarantees. We therefore also present a modification of PI so that recursive feasibility is guaranteed a priori under mild conditions. This modified algorithm, called PI+, is shown to preserve the recursive robust stability when the attractor is compact. In addition, PI+ enjoys the same near-optimality properties as its PI counterpart under the same assumptions.
Analyzing the safety of automotive systems at the application level is -especially challenging as often very complex or even incomplete models are available. The authors combine formal reasoning with simulation data t...
详细信息
Analyzing the safety of automotive systems at the application level is -especially challenging as often very complex or even incomplete models are available. The authors combine formal reasoning with simulation data to effectively -validate safety or estimate worst case situations of automotive control systems.
This paper describes a heading and pitch determination algorithm using single-epoch measurements from two GPS receivers. The algorithm aims to provide accurate heading and pitch solutions with controlled quality. Sinc...
详细信息
This paper describes a heading and pitch determination algorithm using single-epoch measurements from two GPS receivers. The algorithm aims to provide accurate heading and pitch solutions with controlled quality. Since the separation of the two antennas (baseline length) is known, this length has been used in the algorithm to constrain the heading and pitch solutions in the final stage of the estimation process. Moreover, error bounds (EB) of heading and pitch solutions are calculated at each epoch in order to provide statistical bounds for their errors. The proposed algorithm has been tested with high (frequent turns with quick change in pitch angle to +/- 50 degrees and bank angle to +/- 80 degrees) and low dynamic flight test data collected in Germany and the solutions are compared with high-accuracy heading and pitch provided by a high-grade GPS/INS system. Test results of achieved accuracies and success rate of bounds on heading and pitch errors are described. This attitude determination work is part of the ANASTASIA (Airborne New Advanced Satellite techniques & Technologies in A System Integrated Approach) project, which was a 6th Framework programme of European Commission Project that was led by Thales Avionics, France. It has been found that about 96% of accepted solutions in the low dynamic data sets have heading and pitch errors less than the ANASTASIA required accuracy of 0.4 degrees. Moreover, the rms error and error bounding analyses of the results show that the performance of the heading and pitch algorithm on the low dynamic data sets is better than that of the high dynamic data sets. Extensive analysis of the relatively poor performance on the high dynamic data sets is described and the problems/error sources are identified and discussed.
We analyzed the delay failure induced in standard-cell ASICs by total-ionizing dose. We developed a novel cell-level fault model for delay failures. We used this fault model to identify worst-case test vectors (WCTV) ...
详细信息
We analyzed the delay failure induced in standard-cell ASICs by total-ionizing dose. We developed a novel cell-level fault model for delay failures. We used this fault model to identify worst-case test vectors (WCTV) for delay failures induced in ASIC devices exposed to total ionizing dose. The fault model was experimentally validated using SPICE simulation and total dose data. We introduced a fast search algorithm based on directed graph and genetic algorithms to help identify WCTV for large ASICs within reasonable search time. The methodology was validated using ASIC test chip and Cobalt 60 facility.
A held matching technique is used to analyze an N-way divider which includes a radial cavity with a coaxial central port and H-plane coupled rectangular waveguide output ports, Dielectric coated and disc ended probes ...
详细信息
A held matching technique is used to analyze an N-way divider which includes a radial cavity with a coaxial central port and H-plane coupled rectangular waveguide output ports, Dielectric coated and disc ended probes at the central port are considered. The analysis is divided into two parts, Part one is associated with the modeling of the central port and part two is concerned with the modeling of the waveguide peripheral ports. The analysis is simplified by assuming that the interaction between the central port and the peripheral ports is by means of axially symmetric radial waves which are uniform with cavity's height. Based on this analysis, algorithms are developed for calculating the scattering parameters of this structure using a PC. These algorithms show good agreement when tested against a finite element package (HFSS) and against measurements.
A distributed computation consists of a set of processes chat cooperate and compete to achieve a common goal. These processes do not share a common global memory and communicately solely by passing messages communicat...
详细信息
A distributed computation consists of a set of processes chat cooperate and compete to achieve a common goal. These processes do not share a common global memory and communicately solely by passing messages communication network. The communication delay is finite but unpredictable. A process's actions are modeled as three types of events: internal, message send, and message receive. An internal event affects only the process at which it occurs, and the events at a process are linearly ordered by their order of occurrence. Send and receive events signify the flow of information between processes and establish causal dependency from the sender process to the receiver process. Consequently, the execution of a distributed application results in a set of distributed events produced by the process. The causal precedence relation induces a partial order on the events of a distributed computation. Causality among events, more formally the causal precedence relation, is a powerful concept for reasoning, analyzing, and drawing inferences about a distributed computation. Knowledge of the causal precedence relation between processes helps programmers, designers, and the system itself solve a variety of problems in distributed computing. The notion of time is basic to capturing the causality between events. Distributed systems have no built-in physical time and can only approximate it. However, in a distributed computation, both the progress and the interaction between processes occur in spurts. Consequently, logical clocks can be used to accurately capture the causality relation between events. This article presents a general framework of a system of logical clocks in distributed systems and discusses three methods-scalar, vector, and matrix-for implementing logical time in these systems.
The performance of gradient search adaptive filters, such as the least mean squares (LMS) algorithm, may degrade badly when the filter is subjected to input signals which are corrupted by impulsive interference. The m...
详细信息
The performance of gradient search adaptive filters, such as the least mean squares (LMS) algorithm, may degrade badly when the filter is subjected to input signals which are corrupted by impulsive interference. The median 1,MS (MLMS) adaptive filter is designed to alleviate this problem by protecting the filter coefficients from the impact of the impulses. MLMS is a modification of LMS, obtained by applying a median operation to the raw gradient estimates of the mean squared error performance surface. An analysis of the MLMS algorithm is provided for the class of independent and identically distributed inputs. For these inputs we establish exponential convergence of the MLMS algorithm. The rate of convergence is shown to depend on order statistics of the input, but unlike that for LMS shows little dependence on characteristics of the impulsive interference. The steady state performance of the LMS and MLMS is also examined. The average deviation of the parameter estimates from their optimal values caused by the arrival of an impulse is assessed. This analysis indicates a significantly improved performance for MLMS compared to LMS. Analytic predictions for both convergence and steady state behavior are supported by simulations.
Presents a heuristic search algorithm inspired by evolutionary methods to solve the all-terminal network design problem when considering cost and reliability. Network topology; Failure of links as mutually independent...
详细信息
Presents a heuristic search algorithm inspired by evolutionary methods to solve the all-terminal network design problem when considering cost and reliability. Network topology; Failure of links as mutually independent; Connectivity check for a spanning tree; Set of nodes and links.
Networks of linear units are the simplest kind of networks, where the basic questions related to learning, generalization, and self-organization can sometimes be answered analytically, We survey most of the known resu...
详细信息
Networks of linear units are the simplest kind of networks, where the basic questions related to learning, generalization, and self-organization can sometimes be answered analytically, We survey most of the known results on linear networks, including: 1) backpropagation learning and the structure of the error function landscape, 2) the temporal evolution of generalization, and 3) unsupervised learning algorithms and their properties, The connections to classical statistical ideas, such as principal component analysis (PCA), are emphasized as well as several simple but challenging open questions, A few new results are also spread across the paper, including an analysis of the effect of noise on backpropagation networks and a unified view of all unsupervised algorithms.
In this paper, we study the facility location problems on the real line. Given a set of n customers on the real line, each customer having a cost for setting up a facility at its position, and an integer k, we seek to...
详细信息
In this paper, we study the facility location problems on the real line. Given a set of n customers on the real line, each customer having a cost for setting up a facility at its position, and an integer k, we seek to find at most k of the customers to set up facilities for serving all n customers such that the total cost for facility set-up and service transportation is minimized. We consider several problem variations including the k-median, the k-coverage, and the linear model. The previously best algorithms for these problems all take O(nk) time. Our new algorithms break the O(nk) time bottleneck and solve these problems in sub-quadratic time. Our algorithms are based on a new problem modeling and interesting algorithmic techniques, which may find other applications as well.
暂无评论