Driver adaptive collision warning system
View Patent ↗The present invention involves methods and systems for issuing a collision warning to a driver at timing based on the driver's attitude level. The methods involve the steps of determining the driver attitude level based on the driver's actions in a plurality of driving conditions, determining timing for issuing warning based on the driver attitude level, and issuing warning based on the determined timing. The systems include a first device for collecting data associated with driver attitude, and a second device connected to the first device for determining a driver attitude level. The second device includes a processor for processing the data by applying pre-determined algorithm to determine the driver attitude level. The systems may further include a third device connected to the second device for determining timing for issuing warning corresponding to the driver attitude level, and issuing a warning according to the determined timing. Alternatively or additionally, the third device may include a software for enabling a determination of the rate of change in terms of acceleration and deceleration, and a control for affecting the rate of change. Further, the third device may include a software for enabling a determination of a safety distance and a distance control.
1. A method for forewarning a driver of a potential collision comprising the steps of:
(a) determining a driving style of a driver;
(b) determining timing for issuing a warning based on the determined driving style; and
(c) issuing the warning based on the determined timing, wherein the step of (a) comprises the steps of:
(i) collecting data corresponding to a plurality of parametric models (PMs) relating to driver attitude level of an individual driver;
(j) determining parametric model (PM) value distribution for each PM from the collected data;
(k) determining an average value for each PM value distribution; and
(l) determining a parametric-specific attitude level for each PM by mapping the average value for each PM value distribution to a pre-determined scale specific to each PM.
2. The method of claim 1 , wherein the step of (a) comprises the steps of:
(g) collecting data relating to a plurality of drivers' actions in a plurality of driving conditions; and
(h) determining the driving style of a driver based on the collected data.
3. The method of claim 1 , wherein the step of (c) involves issuing at least one perceptible signal.
4. The method of claim 3 , wherein the perceptible signal includes at least one of an audio signal, a visual signal, and a tactile signal.
5. The method of claim 1 , wherein the step of (i) is performed by at least one sensor of an automobile.
6. The method of claim 1 , wherein the step of (j) is performed by applying pre-determined algorithm corresponding to each of the PMs.
7. The method of claim 1 , wherein the step of (j) includes selecting the collected data based on pre-determined criteria specific to each of the PMs.
8. The method of claim 1 , wherein the step of (j) includes calculating percentile values representing the PM value distribution.
9. The method of claim 1 further comprising the step of:
(m) averaging weighted values of parametric-specific attitude level for all PMs, wherein each of the weighted values of parametric-specific attitude level is derived by multiplying the parametric-specific attitude level with a pre-determined weight corresponding to relevance of the PM to the driver attitude level.
10. The method of claim 1 , wherein the parametric models (PMs) in the step of (i) includes at least one of: host vehicle velocity (hostv) PM for capturing how fast a driver drives in an absence of a lead vehicle, acceleration (host-acc) PM for capturing acceleration profile as a function of hostv, and driver's allowable response time (T ares ) PM for capturing how much time the driver is allowed to react to an extreme stop condition without colliding with a lead vehicle.
11. The method of claim 10 , wherein the step of (i) includes collecting data corresponding to hostv parametric model based on criteria including: no lead vehicle, hostv being greater than a pre-determined speed, and no lane change.
12. The method of claim 11 wherein the step of (j) includes calculating from the collected data, hostv PM value distribution for percentiles from 0 to 100, wherein the step of (k) includes obtaining an average value of hostv PM values.
13. The method of claim 10 , wherein the step of (i) includes collecting data corresponding to host-acc based on criteria including: host-acc being greater than a pre-determined rate, and no lane change.
14. The method of claim 13 , wherein the step of (j) includes calculating from the collected data, host-acc PM value distribution for percentiles from 0 to 100, and wherein the step of (k) includes obtaining an average value of host-acc PM values.
15. The method of claim 10 , wherein step (j) includes calculating T ares values using equation I:
0
=
[
a
max
2
(
1
+
a
max
d
max
)
]
T
ares
2
++
[
host_v
(
1
+
a
max
d
max
)
+
(
a
max
sum
ad
J
max
)
+
a
max
2
d
max
J
max
(
2
a
max
sum
ad
-
sum
ad
2
)
]
T
ares
+
[
(
hostv
2
-
leadv
2
2
d
max
)
+
sum
ad
hostv
J
max
(
1
+
1
d
max
(
a
max
-
sum
ad
2
)
)
+
sum
ad
2
2
J
max
2
(
a
max
-
sum
ad
3
)
+
1
2
d
max
(
2
a
max
sum
ad
-
sum
d
2
2
J
max
)
2
-
range
]
where a max is a maximum acceleration, d max is a maximum deceleration, J max is a maximum allowed jerk, and sum ad is a summation of a max and d max .
16. The method of claim 15 , wherein a max , d max , and J max are constants.
17. The method of claim 15 , wherein the step of (j) further includes selecting T ares values based on criteria comprising: no lead vehicle, no lane change, lead vehicle velocity(leadv)-hostv being less than 5 mph for 1 second, and T are < T ares +(2* std(T ares )) , where T ares is an approximate average and std(T ares ) is an approximate standard deviation, of all selected T ares data.
18. The method of claim 17 , wherein the step of (k) further includes calculating T ares PM value distribution for percentiles from 0 to 100, and wherein the step of (1) includes obtaining an average value of T ares PM values.
19. The method of claim 1 , wherein the parametric models (PMs) in the step of (i) include host vehicle velocity (hostv) PM for capturing how fast a driver drives in an absence of a lead vehicle, acceleration (host–acc) PM for capturing acceleration profile as a function of hostv, and drive's allowable response time (T ares ) PM for capturing how much time the driver is allowed to react to an extreme stop condition without colliding with a lead vehicle;
wherein the step of (j) includes determining PM value distribution for hostv PM, host–acc PM, and T ares PM;
wherein the step of (k) includes determining an average value for hostv PM, host–acc PM, and T ares PM; and wherein the step of (1) includes determining a parametric-specific attitude level for hostv PM, host–acc PM, and T ares PM.
20. The method of claim 10 further comprising the steps of:
calculating a weighted value of hostv PM by multiplying the corresponding parameter-specific driver attitude level with a pre-determined weight corresponding to hostv PM;
calculating a weighted value of host–acc PM by multiplying the corresponding parameter-specific driver attitude level with a pre-determined weight corresponding to host–acc PM;
calculating a weighted value of T ares PM by multiplying the corresponding parameter-specific driver attitude level with a pre-determined weight corresponding to T ares PM; and
averaging the weighted values of hostv PM, host–acc PM, and T ares PM to create an overall driver attitude level.
21. The method of claim 20 further comprising the step of:
(n) determining timing for issuing warning based on the overall driver attitude level.
22. A method for forewarning a driver of a potential collision comprising the steps of:
(a) collecting data corresponding to a plurality of parametric models (PMs) relating to a driver attitude level of an individual driver including at least one parametric model associated with said driver's aggression level;
(b) applying a pre-determined algorithm to the data to determine the driver's aggression level;
(c) determining timing for issuing a warning corresponding to the driver's aggression level; and
(d) issuing a warning based on the determined timing, wherein the step (a) comprises the steps of:
determining parametric model (PM) value distribution for each PM from the collected data;
determining an average value for each PM value distribution; and
determining a parametric-specific attitude level for each PM by mapping the average value for each PM value distribution to a pre-determined scale specific to each PM.