基于新型联想记忆系统的机器人分布式智能控制

基于离散泰勒级数提出一种对多维函数可实现任意阶逼近的新型CMAC神经网络——DTS-CMAC,详细讨论了该系统的插值算法、训练规则,与传统CMAC相比,DTS-CMAC具有学习精度高、学习速度快及占用存储单元少等优点。基于DTS-CMAC设计了一种高性能的机械手轨迹跟踪分布式智能控制方案,并以肘关节为例,设计了机械手关节转矩控制器,仿真研究表明了该方案的可行性与有效性。     

Risk assessment for a video surveillance system based on Fuzzy Cognitive Maps

For various IT systems security is considered a key quality factor. In particular, it might be crucial for video surveillance systems, as their goal is to provide continuous protection of critical infrastructure and other facilities. Risk assessment is an important activity in security management; it aims at identifying assets, threats and vulnerabilities, analysis of implemented countermeasures and their effectiveness in mitigating risks. This paper discusses an application of a new risk assessment method, in which risk calculation is based on Fuzzy Cognitive Maps (FCMs) to a complex automated video surveillance system. FCMs are used to capture dependencies between assets and FCM based reasoning is applied to aggregate risks assigned to lower-level assets (e.g. cameras, hardware, software modules, communications, people) to such high level assets as services, maintained data and processes. Lessons learned indicate, that the proposed method is an efficient and low-cost approach, giving instantaneous feedback and enabling reasoning on effectiveness of security system.     

Signal shaper with a distributed delay: Spectral analysis and design

Input shapers with time delays have proved useful in many applications related to controls for various oscillatory devices, for example flexible manipulators and cranes. In the paper, a novel approach for designing a zero-vibration signal shaper based on equally distributed delay is proposed. The parameter assessment of the shaper is based on the spectral approach. Various characteristics of the shaper are analyzed and compared with the classical zero-vibration shaper with a lumped delay. The analysis shows that the novel shaper is a slower, but more robust alternative to the classical shaper. Besides, the discrete implementation of the shaper is proposed and tested. It includes zero placement based parameter adjustment with the objective to preserve full compensation of the oscillatory mode by the discrete algorithm.     

Analyzing probabilistic pushdown automata

The paper gives a summary of the existing results about algorithmic analysis of probabilistic pushdown automata and their subclasses.     

Evaluation of quadratic cost functionals for neutral systems: the frequency-domain approach

As shown in Grabowski (1983), the problem of evaluation of the quadratic performance index for delay-differential systems of neutral type can be effectively solved by the use of Lyapunov functionals. In this paper another solution to that problem, based on the frequency-domain approach, is proposed. With the use of Plancherel's theorem and other tools of complex analysis, this task is reduced to a Riemann-Hilbert problem, an explicit solution to which is found by elementary methods. Our results can be regarded as a generalization of those due to Walton and Marshall (1984, 1987). Two simple examples of application are also provided.     

A two-pronged approach for springback variability assessment using sparse polynomial chaos expansion and multi-level simulations

In this study, we show that stochastic analysis of metal forming process requires both a high precision and low cost numerical models in order to take into account very small perturbations on inputs (physical as well as process parameters) and to allow for numerous repeated analysis in a reasonable time. To this end, an original semi-analytical model dedicated to plain strain deep drawing based on a Bending-Under-Tension numerical model (B-U-T model) is used to accurately predict the influence of small random perturbations around a nominal solution estimated with a full scale Finite Element Model (FEM). We introduce a custom sparse variant of the Polynomial Chaos Expansion (PCE) to model the propagation of uncertainties through this model at low computational cost. Next, we apply this methodology to the deep drawing process of U-shaped metal sheet considering up to 8 random variables.     

Decision trees for uplift modeling with single and multiple treatments

Most classification approaches aim at achieving high prediction accuracy on a given dataset. However, in most practical cases, some action such as mailing an offer or treating a patient is to be taken on the classified objects, and we should model not the class probabilities themselves, but instead, the change in class probabilities caused by the action. The action should then be performed on those objects for which it will be most profitable. This problem is known as uplift modeling, differential response analysis, or true lift modeling, but has received very little attention in machine learning literature. An important modification of the problem involves several possible actions, when for each object, the model must also decide which action should be used in order to maximize profit. In this paper, we present tree-based classifiers designed for uplift modeling in both single and multiple treatment cases. To this end, we design new splitting criteria and pruning methods. The experiments confirm the usefulness of the proposed approaches and show significant improvement over previous uplift modeling techniques.