We define, in terms of noncommutative algebra, a transfer function matrix of a meromorphic nonlinear time-varying system, which algebraically characterizes the input-output relation of the system. Although the transfer function matrix represents the input-output relation of the system, the matrix derived from the state-space representation can depend on the state variables. By exploiting the results of noncommutative algebra, it is shown that the state variables can always be eliminated from the transfer function matrix.
In this paper, we discuss some phenomena of obstacle clustering by distributed autonomous robots, in the light of space-discretization (or cellular automata) approach. This work was motivated by Swiss Robots which collect scattered obstacles into some clusters without any global information nor intelligent concentrated controller. In order to evaluate these phenomena from quantitative and statistical points of view, we propose an analysis platform using discretized state space, i.e., a hexagonal cellular space where the robots’ direction and velocity are discretized as well. We then introduce two types of local rule, Sense & Avoid rule (which resembles the Swiss Robot’s action) and Push & Turn rule and compare the results focusing on size of resulting clusters, transient/steady-state behaviors and density of obstacles and robots.
To identify multiattribute utility functions with the mutual utility independence, decision makers must specify indifference points and subjective probabilities precisely. By relaxing this exact evaluation, multiattribute value models with incomplete information have been developed. In this paper, we present a new method finding a best alternative through strict preference relations derived by questions that are not difficult to answer. For given strict preference relations, in our method alternatives satisfying the relations can be obtained by solving a mathematical programming problem with constraints representing the preference relations. If there are multiple best alternatives, we can eventually find a unique best one by solving the mathematical programming problem with additional preference constraints.
This paper proposes an improved learning scheme for DTFEL (Discrete time feedback error learning) . Although conventional DTFEL schemes based on the gradient method are easily implemented, these schemes are applied to only SISO systems and have drawbacks that assumptions of positive realness and PE condition for the convergence of the tracking error are required. Also, a class of feedforward (FF) controller is confined to biproper case, and the prefilter parameters, which is known, have to be learned as well as the plant parameters. On the other hand, the proposed scheme can relax two assumptions above, be applied to MIMO systems and improve the problems of both degree-of-freedom and redundancy of FF controller by using the desired value of the future. The paper verifies its effectiveness in terms of simulation.