System model the processing of heterogeneous sensory information in robotized complex

Analyzed the scope and the types of robotic systems consisting of subsystems of the form "a heterogeneous sensors data processing subsystem". On the basis of the Queuing theory model is developed taking into account the unevenness of the intensity of information flow from the sensors to the subsystem of information processing. Analytical solution to assess the relationship of subsystem performance and uneven flows. The research of the obtained solution in the range of parameter values of practical interest.

. Graphs of the functioning of the LVS RK a closed-type sensors having different intensity of information flows: a) N=1; b) N=2; a) N=3.
Note that the solution for the model of type M/M/1//N, according to the classification of Kendall, ie, with the same intensity of service is known and the difference of the proposed model consists of recording the fact of non-uniformity of generation of the information flow sensors of the lower level.
In the first stage, first decide the task for the special cases N=1, N=2, N=3 and try to identify patterns, which can be used to find a common solution.
Based on the graphs depicted in Figure 1 and describing the process of functional planning of such a system, for different occasions make a system of linear differential equations [10] that will move on to the differential. These equations have the following form. 1.
For the stationary regime we get the following solutions:  3  2  1  2  3  1  3  1  2  3  2  3  1  2  1   1  2  3  2  1  3  1  3  2  3  1  2  2  3  1  3  2  1  1  2  3   15   10   2  1  3  1  3  2  3  1  2  2  3  1  3  2  1 3 where Si is a discrete state Markov chain describing the functioning of the regulation of LVS RK; RK group of States, which determines the probability of finding a system To the requests; The solution of the problem for cases N > 3 becomes much more complicated, however, analysis of the structural expression (1) allows to obtain a General formula for determining RK for any number of sensors having different intensity of the appearance of the queries [11]: To assess the impact of the uneven information processing sensors the lower level to the time parameters of the processing subsystems will conduct a research of the solution. Thus comparable values for the utilization rate of system resources subsystem processing options with identical and different treatment times, but with equal total intensity of the stream of queries processed, i.e., λоб = N λ -for RK with the same sensors; Analyze in the framework of the approach [12,13] will change as the parameter The value of the function δ(N, ρоб) for different N is shown in Figure 2.
, it δ(N, ρоб) a good illustration of the change a L . For different N it is possible to accurately find the maximum value of δ(N, Rob) by determining the first derivative of expression (6) and then solving the resulting equation for ρ. However, because of the cumbersome decision appropriate to use numerical methods.
In ρоб > 20 and ρоб ≤ 6•10 -3 the differences are minor and do not exceed 12%. The observed increase in the discrepancy function δ(N Rob) as the number of elements, while at the same time the maximum moves to the left.

Conclusion
The studies show that the value of the parameter a L in the two-phase model of mass service closedcircuit, originating in matter from sensors with different intensity, with different parameters N, ρ, may be different from the known models by more than 35 %. Given the fact that the parameter a L directly determines the bandwidth of this structure, it is obvious that for the analysis of processes of functioning of the internal LAN of Kazakhstan it is necessary to use the developed model.