Assessment support program and assessment support device for automatic braking system
View Patent ↗An assessment support program for an automatic braking system is executed by a computer having at least a CPU. The CPU functions as a part that constitutes an assessment unit that makes an assessment of the performance of the automatic braking system on the basis of a significance probability relating to a probability distribution. The probability distribution is obtained by a logistic regression analysis based on a test result. The test result indicates whether or not a vehicle equipped with the automatic braking system collides with an obstacle under actuation of the automatic braking system. An explanatory variable in the logistic regression analysis is a relative speed vi of the vehicle relative to the obstacle before the actuation of the automatic braking system. A response variable in the logistic regression analysis is an occurrence probability of an event in the test result.
1. An assessment support program for an automatic braking system executed by a computer having at least a CPU, the assessment support program executing the following steps:
a regression analysis step making the CPU obtain a probability distribution by a logistic regression analysis with reading out from a memory a result of a test of whether or not a vehicle equipped with the automatic braking system collides with an obstacle under actuation of the automatic braking system; and
an assessment step making the CPU function as a part that constitutes an assessment unit that makes an assessment of performance of the automatic braking system on the basis of the probability distribution;
wherein:
an explanatory variable in the logistic regression analysis is a relative speed of the vehicle relative to the obstacle before the actuation of the automatic braking system; and
a response variable in the logistic regression analysis is an occurrence probability of an event in the result of the test.
2. The assessment support program for an automatic braking system according to claim 1 , wherein
the assessment unit assesses that instability of the automatic braking system decreases with a decrease in a significance probability of the probability distribution.
3. The assessment support program for an automatic braking system according to claim 2 , wherein
by using the following regression formula 1 in which P(vi) represents an occurrence probability of an event relating to the vehicle having the relative speed, β0 represents a maximum likelihood estimate of a constant, and β1 represents a maximum likelihood estimate of a regression coefficient of the relative speed, the assessment unit assesses that instability of the automatic braking system decreases with a decrease in the significance probability of the maximum likelihood estimate β0 of the constant
ln
[
P
(
vi
)
1
-
P
(
vi
)
]
=
β
0
+
β
1
·
vi
.
Formula
1
4. The assessment support program for an automatic braking system according to claim 3 , wherein
the assessment unit makes an assessment of the performance of the automatic braking system on the basis of the maximum likelihood estimate β1 of the regression coefficient of the relative speed.
5. The assessment support program for an automatic braking system according to claim 4 , wherein
the assessment unit assesses that robustness of the automatic braking system increases with a decrease in the maximum likelihood estimate β1 of the regression coefficient of the relative speed.
6. The assessment support program for an automatic braking system according to claim 2 , wherein
the event includes a collision non-occurrence event in which the vehicle does not collide with the obstacle, and a collision occurrence event in which the vehicle collides with the obstacle, and
the assessment unit makes an assessment of the performance of the automatic braking system on the basis of a significance probability relating to the collision occurrence event in the probability distribution.
7. The assessment support program for an automatic braking system according to claim 6 , wherein
the collision occurrence event includes a first collision occurrence event and a second collision occurrence event, and
the assessment unit makes an assessment of the performance of the automatic braking system on the basis of the significance probability relating to the first collision occurrence event or the second collision occurrence event.
8. The assessment support program for an automatic braking system according to claim 7 , wherein
the first collision occurrence event is a light collision occurrence event in which a collision of the vehicle produces a light adverse effect, and the second collision occurrence event is a serious collision occurrence event in which a collision of the vehicle produces a serious adverse effect.
9. The assessment support program for an automatic braking system according to claim 1 , wherein
the CPU functions as a part that constitutes an assessment subject judgement unit that judges whether or not the probability distribution is included in subjects of the assessment on the basis of a significance probability relating to the probability distribution.
10. An assessment support program for an automatic braking system executed by a computer having at least a CPU, the assessment support program executing the following steps:
a regression analysis step making the CPU obtain a probability distribution by a logistic regression analysis with reading out from a memory a result of a test of whether or not a vehicle equipped with the automatic braking system collides with an obstacle under actuation of the automatic braking system; and
an assessment step making the CPU function as a part that constitutes an assessment unit that makes an assessment of performance of the automatic braking system on the basis of a logarithm of an odds ratio relating to the probability distribution;
wherein:
an explanatory variable in the logistic regression analysis is a relative speed of the vehicle relative to the obstacle before the actuation of the automatic braking system; and
a response variable in the logistic regression analysis is an occurrence probability of an event in the result of the test.
11. The assessment support program for an automatic braking system according to claim 10 , wherein
the assessment unit assesses that robustness of the automatic braking system relative to the relative speed vi increases with a decrease in the logarithm of the odds ratio.
12. The assessment support program for an automatic braking system according to claim 10 , wherein
the logarithm of the odds ratio is a maximum likelihood estimate of a regression coefficient of the relative speed vi with respect to the probability distribution obtained by the logistic regression analysis.
13. The assessment support program for an automatic braking system according to claim 10 , wherein
the CPU functions as a part that constitutes an assessment subject judgment unit that judges whether or not the probability distribution is included in subjects of the assessment on the basis of a significance probability relating to the probability distribution.
14. An assessment support program for an automatic braking system executed by a computer having at least a CPU, the assessment support program executing the following steps:
a regression analysis step making the CPU obtain a probability distribution by a logistic regression analysis with reading out from a memory a result of a test of whether or not a vehicle equipped with the automatic braking system collides with an obstacle under actuation of the automatic braking system; and
an assessment step making the CPU function as a part that constitutes an assessment unit that makes an assessment of performance of the automatic braking system on the basis of the probability distribution and an assessment subject judgment unit that judges whether or not the probability distribution is included in subjects of the assessment on the basis of a significance probability relating to the probability distribution;
wherein:
an explanatory variable in the logistic regression analysis is a relative speed of the vehicle relative to the obstacle before the actuation of the automatic braking system; and
a response variable in the logistic regression analysis is an occurrence probability of an event in the result of the test.
15. The assessment support program for an automatic braking system according to claim 14 , wherein
the assessment subject judgment unit compares the significance probability with a threshold value in magnitude, and excludes the automatic braking system from the subjects of the assessment in a case where the significance probability is larger than the threshold value.
16. The assessment support program for the automatic braking system according to claim 15 , wherein
in the following regression formula 1 in which P(vi) represents an occurrence probability of an event relating to the vehicle having the relative speed, β0 represents a maximum likelihood estimate of a constant, and β1 represents a maximum likelihood estimate of a regression coefficient of the relative speed, the assessment subject judgment unit makes a judgment by using the significance probability of the maximum likelihood estimate β0 of the constant
ln
[
P
(
vi
)
1
-
P
(
vi
)
]
=
β
0
+
β
1
·
vi
.
Formula
1
17. The assessment support program for the automatic braking system according to claim 15 , wherein
the assessment subject judgment unit adds magnitude comparison result information, which represents a result of the magnitude comparison, to information on the probability distribution, and
includes in subjects of the assessment the probability distributions belonging to the same group, out of the probability distributions sorted into groups on the basis of the magnitude comparison result information.
18. An assessment support device for an automatic braking system having a computer having at least a CPU, comprising:
an assessment unit for making an assessment of performance of the automatic braking system on the basis of a probability distribution, wherein
the probability distribution of the assessment unit is obtained by a logistic regression analysis based on a result of a test of whether or not a vehicle equipped with the automatic braking system collides with an obstacle under actuation of the automatic braking system;
an explanatory variable in the logistic regression analysis is a relative speed of the vehicle relative to the obstacle before the actuation of the automatic braking system; and
a response variable in the logistic regression analysis is an occurrence probability of an event in the result of the test.