Method for locating an impact on a surface and device for implementing such a method
View Patent ↗Method for locating an impact on a surface ( 9 ), in which acoustic sensors ( 6 ) pick up acoustic signals s ki (t) generated by the impact and the impact is located by calculating, for a number of reference points of index j, a validation parameter representative of a function: PROD kj i 1 i 2 . . . i 2P (ω)=φS ki1 (ω) φR ji1 (ω)*φS ki2 (ω)*φR ji2 (ω) . . . φS ki2p (ω)*φR ji2p (ω) where: φS ki (ω) and φR ji (ω)* are complex phases of S ki (ω) and of R ji (ω), for i=i 1 , i 2 , . . . , i 2p , indices denoting sensors, S ki (ω) and R ji (ω) being the Fourier transform of s ki (t) and r ji (t), r ji (t) being a reference signal corresponding to the sensor i for an impact at the reference point j, p being a non-zero integer no greater than N SENS /2.
1. Method for locating an impact on a surface belonging to an object provided with at least N SENS acoustic sensors, N SENS being a natural integer at least equal to 2, said method being implemented by a central processing unit linked to said acoustic sensors and in which method:
the sensors are made to pick up respectively N SENS signals S ki (t) from acoustic waves generated in the object by said impact,
i being an index between 1 and N SENS which denotes the corresponding sensor, where k is the area where the impact occurred
there is calculated, for each reference point of index j, at least one validation parameter representative of at least one intercorrelation of complex phases representative of said picked up signals S ki (t) and of reference signals r ji (t), each reference signal r ji (t) corresponding to the signal that would be received by the sensor i in case of impact at a reference point j out of N REF reference points belonging to said surface, N REF being a natural integer at least equal to 1 and j being an index between 1 and N REF , and
the impact is located by determining at least one reference point as close as possible to the point of impact, by applying at least one validation criterion to the validation parameter, characterized in that said validation parameter is representative of at least one intercorrelation:
PROD kj i 1 i 2 i 2p (ω)=φ S ki1 (ω)φ R ji1 (ω)*φ S ki2 (ω)*φ R ji2 (ω) . . . φ S ki2p (ω)*φ R ji2p (ω)
where:
φS ki (ω) is the complex phase of S ki (ω), for i=i 1 , i 2 , . . . , i 2p
φR ji (ω) is the complex phase of R ji (ω), for i=i 1 , i 2 , . . . , i 2p
* denotes the conjugate operator, applied to φS ki (ω) for i=i 2m and to φR ji (ω) for i=i 2m-1 , m being an integer between 1 and p;
S ki (ω) is the Fourier transform of S ki (t),
R ji (ω) is the Fourier transform of r ji (t),
p is a non-zero natural integer less than or equal to N SENS /2,
i 1 , i 2 , . . . i 2p are 2p indices denoting 2p sensors, each between 1 and N SENS .
2. Method according to claim 1 , in which the reference signals are predetermined theoretically.
3. Method according to claim 2 , in which the reference signals are used with said object without training phase.
4. Method according to claim 1 , wherein the reference signals are previously learned on a reference device identical to said object, then are used with said object without training phase.
5. Method according to claim 1 , including a step for calculating the intercorrelation PROD kj i 1 i 2 i 2p (ω), then a step for the inverse Fourier transform of this intercorrelation to obtain a time function prod kj i 1 i 2 i 2p (t), from which said validation parameter is then calculated.
6. Method according to claim 5 , in which the intercorrelation PROD kj i 1 i 2 i 2p (ω) is standardized before proceeding with the inverse Fourier transform.
7. Method according to claim 5 , in which, for each reference point j, a resemblance function V kj (t) is calculated, chosen from: prod kj i 1 i 2 i 2 p(t), and
a linear combination of a number of functions prod kj i 1 i 2 i 2p (t) corresponding to a number of subsets of 2p sensors out of N SENS .
8. Method according to claim 7 , in which N SENS is an even number and V kj (t) is proportional to prod kj i 1 i 2 i 2p (t), where p=N SENS /2.
9. Method according to claim 8 , in which N SENS =2 and V kj (t) is proportional to prod kj 12 (t).
10. Method according to claim 7 , in which N SENS =3 and, for each reference point j, a resemblance function V kj (t) is calculated, chosen from: −V kj (t)=a3·[prod kj 12 (t)+prod kj 23 (t)], and
V kj (t)=b3·{prod kj 12 (t)+prod kj 23 (t)+prod kj 13 (t)}, a3 and b3 being constants.
11. Method according to claim 10 , in which a3=½ and b3=⅓.
12. Method according to claim 7 , in which N SENS =4 and, for each reference point j, a resemblance function V kj (t) is calculated, chosen from:
V kj ( t )= a 4·prod kj 1234 ( t ),
V kj ( t )= b 4·[prod kj 12 ( t )+prod kj 34 ( t )],
V kj ( t )= c 4·[prod kj 12 ( t )+prod kj 23 ( t )+prod kj 34 ( t )+prod kj 14 ( t )],
a4, b4 and c4 being constants.
13. Method according to claim 12 , in which a4=1, b4=½ and c4=¼.
14. Method according claim 7 , in which the validation parameter is chosen from:
MAXIMUM
0
j
=
Vkj
(
t
=
0
)
,
MAXIMUM
1
j
=
Max
(
V
kj
(
t
)
)
,
CONTRAST
1
j
=
Max
(
MAXIMUM
0
j
)
(
Σ
j
MAXIMUM
1
j
-
Max
(
MAXIMUM
1
j
)
)
/
(
N
REF
-
1
)
ENERGY
=
Σ
i
(
Σ
t
[
ski
(
t
)
]
2
)
.
15. Method according to claim 14 , in which only V kj (t) is calculated for the time t=0 roughly corresponding to the impact, as being the real part of PROD kj i 1 i 2 i 2p (ω), and MAXIMUM 0j is used as validation parameter.
16. Method according to claim 14 , in which at least one validation criterion is used, chosen from the following criteria:
CONTRAST 1j >THRESHOLD 1 , with THRESHOLD 1 >1,
CONTRAST 1j >THRESHOLD 1 and
MAXIMUM 0j /MAXIMUM ij NOISE >THRESHOLD 2 , where
THRESHOLD 2 >1 and MAXIMUM 1j NOISE corresponds to the parameter MAXIMUM 1J of the signals processed previously and not having resulted in validation,
MAXIMUM 0j >THRESHOLD 3 , with THRESHOLD 3 >0 and
MAXIMUM 0j /MAXIMUM 0j NOISE >THRESHOLD 4 with THRESHOLD 4 >1, where MAXIMUM 0j NOISE corresponds to the parameter MAXIMUM 0j of the signals processed previously and not having resulted in validation,
MAXIMUM 0j /Average (MAXIMUM 0j NOISE )>THRESHOLD 5 , with THRESHOLD 5 >1,
ENERGY/ENERGY NOISE >THRESHOLD 6 , where THRESHOLD 6 >1 and ENERGY NOISE corresponds to the parameter ENERGY of the signals processed previously and not having resulted in validation.
17. Device specially adapted to implement a method according to claim 1 , this device comprising:
an object provided with at least N SENS acoustic sensors, N SENS being a natural integer at least equal to 2, to pick up respectively N SENS signals s ki (t) from acoustic waves generated in the object by an impact on a surface belonging to said object, i being an index between 1 and N SENS which denotes the corresponding sensor,
a central processing unit linked to said acoustic sensors and able to
calculate, for each reference point of index j, at least one validation parameter representative of at least one intercorrelation of complex phases representative of said picked up signals S ki (t) and of reference signals r ji (t), each reference signal r ji (t) corresponding to the signal that would be received by the sensor i in case of impact at a reference point j out of N REF reference points belonging to said surface, N REF being a natural integer at least equal to 1 and j being an index between 1 and N REF , and
locate the impact by determining at least one reference point as close as possible to the point of impact, by applying at least one validation criterion to the validation parameter, characterized in that said validation parameter is representative of at least one intercorrelation:
PROD kj i 1 i 2 i 2p (ω)=φ S ki1 (ω)φ R ji1 (ω)*φ S ki2 (ω)*φ R ji2 (ω) . . . φ S ki2p (ω)*φ R ji2p (ω)
where:
φS ki (ω) is the complex phase of S ki (ω), for i=i 1 , i 2 , . . . , i 2p ,
φR ji (ω) is the complex phase of R ji (ω), for i=i 1 , i 2 , . . . , i 2p ,
* denotes the conjugate operator, applied to φS ki (ω) for i=i 2m and to φR ji (ω) for i=i 2m-1 , m being an integer between 1 and p;
S ki (ω) is the Fourier transform of S ki (t),
R ji (ω) is the Fourier transform of r ji (t),
p is a non-zero natural integer less than or equal to N SENS /2,
i 1 , i 2 , . . . i 2p are 2p indices denoting 2p sensors, each between 1 and N SENS .