Granted Patent B2

US 12,215,976 · App. 17/663,810 · Granted Feb 4, 2025

Inventor: Yuuichi Minamiguchi (Kariya, JP)

Assignee: DENSO CORPORATION

G01C21/188B60W40/068B60W40/103G01C21/32B60W2520/20B60W2552/40B60W2556/40

Abstract

Claims (119)

1. An estimation device configured to estimate a state quantity including a position of a vehicle, the estimation device comprising:

a dead reckoning unit configured to estimate the state quantity by dead reckoning based on (i) a dynamics model including a parameter that causes a change in a behavior of the vehicle and (ii) internal information acquired from an inside of the vehicle;

a map matching unit configured to observe the state quantity by map matching based on (i) map information indicative of a traveling environment of the vehicle and (ii) external information acquired from an outside of the vehicle; and

a parameter correction unit configured to correct the parameter, which is to be fed back for the dead reckoning, based on an offset amount that is a difference between the state quantity observed by the map matching and the state quantity estimated by the dead reckoning; and

an error compensation unit configured to compensate for an error in the state quantity estimated by the dead reckoning by filtering based on the state quantity observed by the map matching,

wherein

the dead reckoning unit is further configured to estimate the state quantity based on the parameter fed back after correction by the parameter correction unit, and

the map matching unit is further configured to perform map matching based on the parameter fed back after the correction,

wherein the map matching unit is further configured to acquire the map information, the external information, and the state quantity which has been estimated by the dead reckoning unit, and observe the state quantity by map matching,

wherein the dead reckoning unit is further configured to acquire the parameter which has been corrected by the parameter correction unit, cause the dynamics model to contain the parameter corrected as a parameter that causes the change in the behavior of the vehicle, acquire the dynamics model containing the corrected parameter and the internal information, and estimate the state quantity by the dead reckoning based on the internal information,

wherein the parameter correction unit is configured to:

predict a deviation that is contained in the offset amount and is a difference, caused by a change amount of the parameter, between the state quantity observed by the map matching and the state quantity estimated by the dead reckoning; and

correct the parameter using the change amount corresponding to the predicted deviation,

wherein the parameter correction unit predicts, as a change amount function δz p , the deviation which is the difference caused by the change amount of the parameter based on at least one of the following formulas:

δ z p =ΔZ ( p t )−Δ Z ( p t-1 );

δ z p =ΔZ ( p t-1 +δp )−Δ Z ( p t-1 ); and

δ z p =ΔZ (δ p ), and

wherein:

pt represents a dynamic parameter at a time t;

pt−1 represent a dynamic parameter at a time t−1, which is previous time of the time t; and

δp represents the change amount from the dynamic parameter pt−1 at the time t−1 to the dynamic parameter pt at the time t, and

wherein:

the state quantity of which the error has been compensated by the error compensation unit is output to an electronic control device configured to execute an advanced driver assistance or an automated driving of the vehicle; and

the electronic control device controls the vehicle to execute the advanced driver assistance or the automated driving.

2. The estimation device according to claim 1 , wherein

the parameter correction unit is configured to predict the deviation that minimizes a difference between the deviation and the offset amount.

3. The estimation device according to claim 1 , wherein

the parameter correction unit is configured to correct the parameter that varies depending on a factor other than the behavior of the vehicle.

4. The estimation device according to claim 3 , wherein

the parameter correction unit is configured to correct the parameter that contains a kinetic friction coefficient of a road surface on which the vehicle is traveling.

5. The estimation device according to claim 1 , wherein:

the parameter correction unit receives the state quantity estimated by the dead reckoning unit and the state quantity observed by the map matching unit to correct the parameter causing the change in the behavior of the vehicle and transmits the corrected parameter to the dynamics model in the dead reckoning unit.

6. A method of estimating a state quantity including a position of a vehicle, the method comprising:

estimating the state quantity by dead reckoning based on (i) a dynamics model including a parameter that causes a change in a behavior of the vehicle and (ii) internal information acquired from an inside of the vehicle;

observing the state quantity by map matching based on (i) map information indicative of a traveling environment of the vehicle and (ii) external information acquired from an outside of the vehicle;

correcting the parameter, which is to be fed back for the dead reckoning, based on an offset amount that is a difference between the state quantity observed by the map matching and the state quantity estimated by the dead reckoning;

compensating for an error in the state quantity by filtering based on the state quantity observed by the map matching;

estimating the state quantity based on the parameter fed back after correction; and

performing map matching based on the parameter fed back after the correction,

wherein

the map information, the external information, and the state quantity, which has been estimated, are acquired, and the state quantity is observed by map matching,

the parameter which has been corrected is acquired, the dynamics model is caused to contain the parameter corrected as a parameter that causes the change in the behavior of the vehicle, the dynamics model containing the corrected parameter and the internal information are acquired, and the state quantity is estimated by the dead reckoning based on the internal information,

wherein the method further comprises:

predicting a deviation that is contained in the offset amount and is a difference, caused by a change amount of the parameter, between the state quantity observed by the map matching and the state quantity estimated by the dead reckoning; and

correcting the parameter using the change amount corresponding to the predicted deviation,

wherein the deviation is predicted, as a change amount function δz p , which is the difference caused by the change amount of the parameter based on at least one of the following formulas:

δ z p =ΔZ ( p t )−Δ Z ( p t-1 );

δ z p =ΔZ ( p t-1 +δp )−Δ Z ( p t-1 ); and

δ z p =ΔZ (δ p ), and

wherein:

pt represents a dynamic parameter at a time t;

pt−1 represent a dynamic parameter at a time t−1, which is previous time of the time t; and

δp represents the change amount from the dynamic parameter pt−1 at the time t−1 to the dynamic parameter p t at the time t, and

wherein:

the state quantity of which the error has been compensated is output to an electronic control device configured to execute an advanced driver assistance or an automated driving of the vehicle; and

the electronic control device controls the vehicle to execute the advanced driver assistance or the automated driving.

7. The method according to claim 6 , wherein

correcting the parameter includes predicting the deviation that minimizes a difference between the deviation and the offset amount.

8. The method according to claim 6 , wherein

correcting the parameter includes correcting the parameter that varies depending on a factor other than the behavior of the vehicle.

9. The method according to claim 8 , wherein

correcting the parameter includes correcting the parameter that contains a kinetic friction coefficient of a road surface on which the vehicle is traveling.

10. The method according to claim 6 , further comprising:

compensating for an error in the state quantity estimated by the dead reckoning by filtering based on the state quantity observed by the map matching.

11. A non-transitory computer readable storage medium storing a computer program for estimating a state quantity including a position of a vehicle, the computer program comprising instructions configured to, when executed by at least one processor, cause the at least one processor to:

estimate the state quantity by dead reckoning based on (i) a dynamics model including a parameter that causes a change in a behavior of the vehicle and (ii) internal information acquired from an inside of the vehicle;

observe the state quantity by map matching based on (i) map information indicative of a traveling environment of the vehicle and (ii) external information acquired from an outside of the vehicle;

correct the parameter, which is to be fed back for the dead reckoning, based on an offset amount that is a difference between the state quantity observed by the map matching and the state quantity estimated by the dead reckoning;

estimate the state quantity based on the parameter fed back after correction; and

perform map matching based on the parameter fed back after the correction,

wherein

the map information, the external information, and the state quantity, which has been estimated, are acquired, and the state quantity is observed by map matching,

the parameter which has been corrected is acquired, the dynamics model is caused to contain the parameter corrected as a parameter that causes the change in the behavior of the vehicle, the dynamics model containing the corrected parameter and the internal information are acquired, and the state quantity is estimated by the dead reckoning based on the internal information,

wherein the instructions are configured to, when executed by the at least one processor, further cause the at least one processor to:

predict a deviation that is contained in the offset amount and is a difference, caused by a change amount of the parameter, between the state quantity observed by the map matching and the state quantity estimated by the dead reckoning; and

correct the parameter using the change amount corresponding to the predicted deviation,

wherein the deviation is predicted, as a change amount function δz p , which is the difference caused by the change amount of the parameter based on at least one of the following formulas:

δ z p =ΔZ ( p t )−Δ Z ( p t-1 );

δ z p =ΔZ ( p t-1 +δp )−Δ Z ( p t-1 ); and

δ z p =ΔZ ( Sp ), and

wherein:

pt represents a dynamic parameter at a time t;

pt−1 represent a dynamic parameter at a time t−1, which is previous time of the time t; and

δp represents the change amount from the dynamic parameter pt−1 at the time t−1 to the dynamic parameter pt at the time t,

wherein:

the state quantity of which the error has been compensated is output to an electronic control device configured to execute an advanced driver assistance or an automated driving of the vehicle; and

the electronic control device controls the vehicle to execute the advanced driver assistance or the automated driving.

12. The non-transitory computer readable storage medium storing the computer program according to claim 11 , wherein the instructions are configured to, when executed by the at least one processor, further cause the at least one processor to:

predict the deviation that minimizes a difference between the deviation and the offset amount.

13. The non-transitory computer readable storage medium storing the computer program according to claim 11 , wherein the instructions are configured to, when executed by the at least one processor, further cause the at least one processor to:

correct the parameter that varies depending on a factor other than the behavior of the vehicle.

14. The non-transitory computer readable storage medium storing the computer program according to claim 13 , wherein the instructions are configured to, when executed by the at least one processor, further cause the at least one processor to:

correct the parameter that contains a kinetic friction coefficient of a road surface on which the vehicle is traveling.

15. The non-transitory computer readable storage medium storing the computer program according to claim 11 , wherein the instructions are configured to, when executed by the at least one processor, further cause the at least one processor to:

compensate for an error in the state quantity estimated by the dead reckoning by filtering based on the state quantity observed by the map matching.

16. An estimation device configured to estimate a state quantity including a position of a vehicle, the estimation device comprising:

a dead reckoning unit configured to estimate the state quantity by dead reckoning based on (i) a dynamics model including a parameter that causes a change in a behavior of the vehicle and (ii) internal information acquired from an inside of the vehicle;

a map matching unit configured to observe the state quantity by map matching based on (i) map information indicative of a traveling environment of the vehicle and (ii) external information acquired from an outside of the vehicle;

a parameter correction unit configured to correct the parameter, which is to be fed back for the dead reckoning, based on an offset amount that is a difference between the state quantity observed by the map matching and the state quantity estimated by the dead reckoning; and

an error compensation unit configured to compensate for an error in the state quantity estimated by the dead reckoning by filtering based on the state quantity observed by the map matching, wherein

the parameter correction unit is configured to correct the parameter that varies depending on a factor other than the behavior of the vehicle,

the dead reckoning unit is further configured to estimate the state quantity based on the parameter fed back after correction by the parameter correction unit, and

the map matching unit is further configured to perform map matching based on the parameter fed back after the correction,

the map matching unit is further configured to acquire the map information, the external information, and the state quantity which has been estimated by the dead reckoning unit, and observe the state quantity by map matching,

the dead reckoning unit is further configured to acquire the parameter which has been corrected by the parameter correction unit, cause the dynamics model to contain the parameter corrected as a parameter that causes the change in the behavior of the vehicle, acquire the dynamics model containing the corrected parameter and the internal information, and estimate the state quantity by the dead reckoning based on the internal information,

wherein the parameter correction unit is configured to:

predict a deviation that is contained in the offset amount and is a difference, caused by a change amount of the parameter, between the state quantity observed by the map matching and the state quantity estimated by the dead reckoning; and

correct the parameter using the change amount corresponding to the predicted deviation,

wherein the parameter correction unit predicts, as a change amount function δz p , the deviation which is the difference caused by the change amount of the parameter based on at least one of the following formulas:

δ z p =ΔZ ( p t )−Δ Z ( p t-1 );

δ z p =ΔZ ( p t-1 +δp )−Δ Z ( p t-1 ); and

δ z p =ΔZ (δ p ), and

wherein:

pt represents a dynamic parameter at a time t;

pt−1 represent a dynamic parameter at a time t−1, which is previous time of the time t; and

δp represents the change amount from the dynamic parameter pt−1 at the time t−1 to the dynamic parameter pt at the time t, and

wherein:

the state quantity of which the error has been compensated by the error compensation unit is output to an electronic control device configured to execute an advanced driver assistance or an automated driving of the vehicle; and

the electronic control device controls the vehicle to execute the advanced driver assistance or the automated driving.

Assignments (1)

Priority Claims (1)

Continuity (2)

References Cited (18)