Malfunction condition judgment apparatus, control method, automobile and program

Abstract

A malfunction condition judgment apparatus (MCJA) that judges malfunction condition (MC) of an observed object based on a change of observed values. The MCJA acquires time series data for values of each of a plurality of variables; calculates, with respect to each of the variables, a statistic defining a probability density function of that variable at T1, based on the value of that variable at T1 and that statistic at a point of time prior to T1; calculates dissimilarity showing an extent of variation between the statistic calculated for each variable and a statistic of a criterial probability density function predetermined corresponding to that variable; and picks, out of the plurality of variables, a variable for which the calculated dissimilarity is larger than a predetermined reference value, as the variable by which MC of the observation object is detected.

1 - 16 . (canceled) 
   
   
       17 . A malfunction condition judgment apparatus that judges a malfunction condition of an observed object based on a change of observed values observed from the observation object, comprising:
 an acquiring unit operable to acquire time series data of values for each of a plurality of variables that has a plurality of observed values observed from the observation object as a value;   a statistic computing unit operable to calculate, with respect to each of the variables, a statistic defining a probability density function of that variable at a point of time (T1), by adding into a value where the statistic at the prior time to T1 is subtracted by a value obtained by multiplying a forgetting rate set corresponding to the variable at the prior time to said point of time, a value where a value of the variable at T1 is multiplied by said forgetting rate;   a dissimilarity computing unit operable to calculate, with respect to each of the variables, dissimilarity showing an extent of variation between the statistic calculated for that variable and a statistic of a criterial probability density function predetermined corresponding to that variable;   an extracting unit operable to pick up, out of the plurality of variables, a variable in which the calculated dissimilarity is larger than a predetermined reference value, as the variable by which malfunction condition of the observed object is detected; and   a forgetting rate setting unit operable to set a forgetting rate with respect to each of the variables in a manner that, in a case where an updating frequency at which that variable is updated in a unit time is higher, the forgetting rate is set lower than in a case where the updating frequency is lower.   
   
   
       18 . The malfunction condition judgment apparatus according to  claim 17 , further comprising:
 a judging unit operable to judge that a malfunction condition has occurred in a portion that is an observed object of observed values held by the picked up variable, among the observation objects.   
   
   
       19 . The malfunction condition judgment apparatus according to  claim 18 , further comprising:
 a change-point detecting unit operable to detect a change-point indicating a point of time when a changing pattern of values of at least one of the picked up variables changes; and   an associated variable selecting unit operable to select a set of variables having change-points, detected by the change-point detecting unit, which are similar to one another, wherein   the judge unit judges the set of variables selected by the associated variable selecting unit to be a set of variables indicating a portion that is a cause of malfunction condition.   
   
   
       20 . The malfunction condition judgment apparatus according to  claim 17 , wherein
 the statistic computing unit calculates, as a statistic of each of the variables, at least one cumulant of the probability density function at T1; and   the dissimilarity computing unit calculates, as the dissimilarity, an extent of variation between the calculated cumulant and a cumulant predetermined with respect to the criterial probability density function.   
   
   
       21 . The malfunction condition judgment apparatus according to  claim 20 , wherein
 the statistic computing unit calculates, with respect to each of the variables, as the statistic, kurtosis and skew of the probability density function at T1; and   the dissimilarity computing unit calculates, as the dissimilarity, at least the extent of variation between the calculated kurtosis or skew and a kurtosis or a skew predetermined with respect to the criterial probability density function.   
   
   
       22 . The malfunction condition judgment apparatus according to  claim 21 , wherein
 the dissimilarity computing unit calculates, as the dissimilarity, differences of the kurtosis and skew calculated by the statistic computing unit respectively from kurtosis and skew in a normal distribution function; and   the extracting unit picks up a variable in which at least any one of the differences of the kurtosis and skew is larger than a predetermined criterion.   
   
   
       23 . The malfunction condition judgment apparatus according to  claim 22 , further comprising:
 a reference statistic recording unit operable to record therein predetermined criterial kurtosis and skew in a manner that the criterial kurtosis and skew correspond to each of the plurality of variables, wherein   the dissimilarity computing unit calculates, as the dissimilarity, difference values of the kurtosis and skew calculated by the statistic computing unit respectively from the kurtosis and skew recorded in the reference statistic recording unit.   
   
   
       24 . The malfunction condition judgment apparatus according to  claim 21 , wherein
 the dissimilarity computing unit calculates, as dissimilarity, differences of the kurtosis and skew calculated by the statistic computing unit respectively from kurtosis and skew in a predetermined discrete value function that has discrete values as a value; and   the extracting unit picks up a variable in which at least any one of the differences of the kurtosis and skew calculated by the dissimilarity computing unit is larger than a predetermined criterion thereof.   
   
   
       25 . The malfunction condition judgment apparatus according to  claim 20 , wherein
 the statistic computing unit calculates a second-order cumulant;   the dissimilarity computing unit calculates, as the dissimilarity, differences of the second-order cumulant calculated by the statistic computing unit from a second-order cumulant in a predetermined constant function that has a constant as a value; and   the extracting unit picks up a variable in which the difference of the second-order cumulant calculated by the dissimilarity computing unit is larger than a predetermined criterion therefore.   
   
   
       26 . The malfunction condition judgment apparatus according to  claim 17 , wherein
 the acquiring unit acquires time series data for each of the variables by monitoring a bus that connects a plurality of observation sensors to one another and thereby sequentially acquiring observed values stored in communication packets transferred through the bus.   
   
   
       27 . The malfunction condition judgment apparatus according to  claim 17 , wherein
 each of the plurality of variable has, as a variable value, observed values observed from each different portion of one automobile; and   the extracting unit picks up, out of the plurality of variables, a variable in which the calculated dissimilarity is larger than the predetermined reference value, as the variable by which malfunction condition of the automobile is detected.   
   
   
       28 . The malfunction condition judgment apparatus according to  claim 17 , wherein
 each of the plurality of variables has, as a variable value, observed values observed from each different portion of a single individual that is an object of diagnosis; and   the extracting unit picks up, out of the plurality of variables, a variable in which the calculated dissimilarity is larger than the predetermined reference value, as the variable by which the malfunction condition of the individual is detected.   
   
   
       29 . An automobile including a plurality of observation sensors connected to one another through a single bus, and a malfunction condition judgment apparatus that judges a malfunction condition based on a change of observed values acquired from the plurality of observation sensors, comprising:
 an acquiring unit operable to acquire time series data of values for each of a plurality of variables that has a plurality of observed values observed from the observation object as a value;   a statistic computing unit operable to calculate, with respect to each of the variables, a statistic defining a probability density function of that variable at a point of time (T1), by adding into a value where the statistic at the prior time to T1 is subtracted by a value obtained by multiplying a forgetting rate set corresponding to the variable at the prior time to said point of time, a value where a value of the variable at T1 is multiplied by said forgetting rate;   a dissimilarity computing unit operable to calculate, with respect to each of the variables, dissimilarity showing an extent of variation between the statistic calculated for that variable and a statistic of a criterial probability density function predetermined corresponding to that variable;   an extracting unit operable to pick up, out of the plurality of variables, a variable in which the calculated dissimilarity is larger than a predetermined reference value, as the variable by which malfunction condition of the observed object is detected; and   a forgetting rate setting unit operable to set a forgetting rate with respect to each of the variables in a manner that, in a case where an updating frequency at which that variable is updated in a unit time is higher, the forgetting rate is set lower than in a case where the updating frequency is lower.   
   
   
       30 . The automobile according to  claim 29 , wherein
 the malfunction condition judgment apparatus is provided so as to be attachable to and detachable from the automobile.   
   
   
       31 . A program that causes an information processing apparatus to function as a malfunction condition judgment apparatus that judges a malfunction condition of an observed object based on a change of observed values observed from the observation object, causing the information processing apparatus to function as:
 an acquiring unit operable to acquire time series data of values for each of a plurality of variables that has a plurality of observed values observed from the observation object as a value;   a statistic computing unit operable to calculate, with respect to each of the variables, a statistic defining a probability density function of that variable at a point of time (T1), by adding into a value where the statistic at the prior time to T1 is subtracted by a value obtained by multiplying a forgetting rate set corresponding to the variable at the prior time to said point of time, a value where a value of the variable at T1 is multiplied by said forgetting rate;   a dissimilarity computing unit operable to calculate, with respect to each of the variables, dissimilarity showing an extent of variation between the statistic calculated for that variable and a statistic of a criterial probability density function predetermined corresponding to that variable;   an extracting unit operable to pick up, out of the plurality of variables, a variable in which the calculated dissimilarity is larger than a predetermined reference value, as the variable by which malfunction condition of the observed object is detected; and   a forgetting rate setting unit operable to set a forgetting rate with respect to each of the variables in a manner that, in a case where an updating frequency at which that variable is updated in a unit time is higher, the forgetting rate is set lower than in a case where the updating frequency is lower.