Correlation Coefficients And Confidence Intervals And Applications In The Process Of Robot Active Olfaction

Robot active olfaction of complex and dangerous environment has importantacademic value and application prospects in industrial production, social security,military affairs and so on. Robot odor source search is an active olfaction. Manyscholars at home and abroad have done research on the algorithm of robot activeolfaction. With the development of society and technology and the increase of robotresearchers (providers) and robot types, various and specific robots’ search strategiesor methods will be inevitably more and more. It is a tend to do research on theevaluation methods of robot search strategy.The main contents of this paper and the results are as follows:(1) In this thesis, a quartic radical function has been proposed to replace Fisher ztransformation to approximate the cumulative distribution function of standard normaldistribution. It is simpler and has18.5%of the error compared with Fisher ztransformation by optimizing parameters. The calculating speed of the quartic radicalfunction is faster than Fisher z transformation. The quartic radical function is a goodtrade-off between the calculating accuracy and speed.(2) Based on the form of quartic radical function, another better approximationhas been found by optimizing the parameters in radical function. The error betweenthe better approximation and the cumulative distribution function of standard normaldistribution is much less than that by the quartic radical function.(3) This paper proposes a way to evaluate odor source search strategy ofmulti-robot collaborative work by using multilevel fuzzy integrated evaluationmethod of fuzzy mathematics. Evaluate the performance of different multi-robot odorsource search algorithms by analyzing the relation of correlation coefficients andconfidence intervals between each two trajectories of multi-robot.