Vehicle behavior prediction device and vehicle behavior prediction method, and driving assistance device

• US 9,727,820 B2
• Filed: 01/20/2012
• Issued: 08/08/2017
• Est. Priority Date: 01/20/2012
• Status: Active Grant

• Current Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Original Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Inventors: Ono, Sayaka
• Primary Examiner(s): Wong, Lut
• Application Number:
  US14/372,359
• Publication Number:
  US 20140358841A1
• Time in Patent Office: 2,027 Days
• Field of Search: 706 52
• CPC Class Codes: B60W 2554/4041 Position
• CPC Class Codes: B60W 2554/803 Relative lateral speed
• CPC Class Codes: B60W 2554/804 Relative longitudinal speed
• CPC Class Codes: B60W 2556/50 of positioning data, e.g. GPS [Global Positioning System] data
• CPC Class Codes: B60W 30/0956 the prediction being responsive to traffic or environmental parameters
• CPC Class Codes: G06N 7/00 Computing arrangements based on specific mathematical models
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G08G 1/0112 from the vehicle, e.g. floating car data [FCD]
• CPC Class Codes: G08G 1/0129 for creating historical data or processing based on historical data
• CPC Class Codes: G08G 1/09626 where the origin of the information is within the own vehicle, e.g. a local storage device, digital map
• CPC Class Codes: G08G 1/096716 where the received information does not generate an automatic action on the vehicle control
• CPC Class Codes: G08G 1/096725 where the received information generates an automatic action on the vehicle control
• CPC Class Codes: G08G 1/096741 where the source of the transmitted information selects which information to transmit to each vehicle
• CPC Class Codes: G08G 1/096775 where the origin of the information is a central station
• CPC Class Codes: G08G 1/096791 where the origin of the information is another vehicle
• CPC Class Codes: G08G 1/163 involving continuous checking
• CPC Class Codes: G08G 1/166 for active traffic, e.g. moving vehicles, pedestrians, bikes

Abstract

In order to provide a vehicle behavior prediction device and a vehicle behavior prediction method, and a driving assistance device that uses said vehicle behavior prediction device, which enables vehicle behavior to be predicted in a highly accurate manner, a vehicle control device is installed in a vehicle. This vehicle control device is provided with a distribution calculation unit and a vehicle behavior prediction unit. The distribution calculation unit determines, on the basis of driving condition information that has been collected, probability distributions for when deceleration behavior occurs and does not occur. The vehicle behavior prediction unit determines the relative positional relationships between the driving speed and acceleration of vehicles ahead with respect to the probability distributions, and, on the basis of the relative positional relationships that have been determined, predicts the behavior of a vehicle that is subject to prediction.

Claims

1. A vehicle behavior prediction device for predicting vehicle behavior based on traveling state information representing a traveling state of a vehicle, wherein

the traveling state information includes at least one value of traveling speed of the vehicle and acceleration of the vehicle,

the vehicle behavior prediction device includes a vehicle behavior prediction unit configured to:

collect the traveling state information,

input a plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value obtained based on the collected traveling state information, and

predict vehicle behavior of a prediction-subject vehicle based on whether the at least one value concerning the prediction-subject vehicle belongs to one of the plurality of kinds of probability distributions,

it is determined whether the at least one value belongs to the probability distribution depending upon whether the at least one value is included in one continuous region in which occurrence probability in the probability distribution is greater than zero,

the traveling state information further includes a traveling location in which the at least one value was detected,

the traveling location includes one of traffic elements including an intersection, a curve, and a temporary stop position,

the vehicle behavior prediction unit is further configured to;

input the plurality of kinds of probability distributions as probability distributions obtained for each traveling location, and

predict the vehicle behavior of the prediction-subject vehicle in a case where a traveling location of the prediction-subject vehicle has characteristics that are common with one of traveling locations in which the plurality of probability distributions were detected,

the plurality of kinds of probability distributions includes a probability distribution at occurrence of specific vehicle behavior and a probability distribution at non-occurrence of the specific vehicle behavior,

the vehicle behavior is deceleration behavior of the vehicle, and

the vehicle behavior prediction unit is further configured to;

collect information representing traveling speed and acceleration of the vehicle as the traveling state information,

input a plurality of kinds of probability distributions obtained for each of traveling speed and acceleration of the vehicle,

predict non-occurrence of the vehicle behavior using the probability distribution based on the traveling speed, and

predict occurrence of the vehicle behavior using the probability distribution based on the acceleration.

2. The vehicle behavior prediction device according to claim 1, wherein when the at least one value belongs to the probability distribution at non-occurrence of the specific vehicle behavior and does not belong to the probability distribution at occurrence of the specific vehicle behavior, the vehicle behavior prediction unit predicts that the prediction-subject vehicle will not exhibit corresponding vehicle behavior.

3. The vehicle behavior prediction device according to claim 1, wherein when the at least one value does not belong to the probability distribution at non-occurrence of the specific vehicle behavior or belongs to both the probability distribution at non-occurrence of the specific vehicle behavior and the probability distribution at occurrence of the specific vehicle behavior, the vehicle behavior prediction unit determines that it is indeterminable whether the prediction-subject vehicle will exhibit vehicle behavior.

4. The vehicle behavior prediction device according to claim 1, wherein when the at least one value belongs to the probability distribution at occurrence of the specific vehicle behavior and does not belong to the probability distribution at non-occurrence of the specific vehicle behavior, the vehicle behavior prediction unit predicts that the prediction-subject vehicle will exhibit corresponding vehicle behavior.

5. The vehicle behavior prediction device according to claim 1, wherein when the at least one value does not belong to the probability distribution at occurrence of the specific vehicle behavior or belongs to both the probability distribution at occurrence of the specific vehicle behavior and the probability distribution at non-occurrence of the specific vehicle behavior, the vehicle behavior prediction unit determines that it is indeterminable whether the prediction-subject vehicle will exhibit corresponding vehicle behavior.

10. A driving assistance device for assisting a driving operation of a driver based on a prediction result of vehicle behavior, the driving assistance device including a driving assistance unit that predicts vehicle behavior of a vehicle to be assisted in the driving operation or a vehicle in front traveling in front of the vehicle in the traveling direction using a vehicle behavior prediction device according to claim 1, and executes driving assistance based on the predicted vehicle behavior.

6. A vehicle behavior prediction device for predicting vehicle behavior based on traveling state information representing a traveling state of a vehicle, wherein

the traveling state information includes at least one value of traveling speed of the vehicle and acceleration of the vehicle,

the vehicle behavior prediction device includes a vehicle behavior prediction unit configured to:

collect the traveling state information,

input a plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value obtained based on the collected traveling state information, and

predict vehicle behavior of a prediction-subject vehicle based on whether the at least one value concerning the prediction-subject vehicle belongs to one of the plurality of kinds of probability distributions,

it is determined whether the at least one value belongs to the probability distribution depending upon whether the at least one value is included in one continuous region in which occurrence probability in the probability distribution is greater than zero,

the traveling state information further includes a traveling location in which the at least one value was detected,

the traveling location includes one of traffic elements including an intersection, a curve, and a temporary stop position,

the vehicle behavior prediction unit is further configured to;

input the plurality of kinds of probability distributions as probability distributions obtained for each traveling location, and

predict the vehicle behavior of the prediction-subject vehicle in a case where a traveling location of the prediction-subject vehicle has characteristics that are common with one of traveling locations in which the plurality of probability distributions were detected,

the vehicle behavior prediction device further comprises a distribution calculating unit configured to obtain a plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value that is input to the vehicle behavior prediction unit,

the distribution calculating unit further includes a reliability zone defining unit configured to define a predetermined zone from a center of the probability distribution representing an average value of the vehicle behavior as a reliability zone concerning occurrence probability of the vehicle behavior, and

the vehicle behavior prediction unit is further configured to predict vehicle behavior of the prediction-subject vehicle using the probability distribution in the reliability zone.

7. A vehicle behavior prediction device for predicting vehicle behavior based on traveling state information representing a traveling state of a vehicle, wherein

the traveling state information includes at least one value of traveling speed of the vehicle and acceleration of the vehicle,

the vehicle behavior prediction device includes a vehicle behavior prediction unit configured to:

collect the traveling state information,

input a plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value obtained based on the collected traveling state information, and

predict vehicle behavior of a prediction-subject vehicle based on whether the at least one value concerning the prediction-subject vehicle belongs to one of the plurality of kinds of probability distributions,

it is determined whether the at least one value belongs to the probability distribution depending upon whether the at least one value is included in one continuous region in which occurrence probability in the probability distribution is greater than zero,

the traveling state information further includes a traveling location in which the at least one value was detected,

the traveling location includes one of traffic elements including an intersection, a curve, and a temporary stop position,

the vehicle behavior prediction unit is further configured to;

input the plurality of kinds of probability distributions as probability distributions obtained for each traveling location, and

predict the vehicle behavior of the prediction-subject vehicle in a case where a traveling location of the prediction-subject vehicle has characteristics that are common with one of traveling locations in which the plurality of probability distributions were detected,

the plurality of kinds of probability distributions includes a probability distribution at occurrence of specific vehicle behavior and a probability distribution at non-occurrence of the specific vehicle behavior,

the vehicle behavior prediction device further comprises a distribution calculating unit configured to obtain a plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value that is input to the vehicle behavior prediction unit,

the distribution calculating unit further includes a reliability zone defining unit configured to define a predetermined zone from a center of the probability distribution representing an average value of the vehicle behavior as a reliability zone concerning occurrence probability of the vehicle behavior, and

the vehicle behavior prediction unit is further configured to determine whether vehicle behavior of the prediction-subject vehicle belongs to the probability distribution based on whether the vehicle behavior belongs to the reliability zone of the probability distributions.

8. A vehicle behavior prediction device for predicting vehicle behavior based on traveling state information representing a traveling state of a vehicle, wherein

the traveling state information includes at least one value of traveling speed of the vehicle and acceleration of the vehicle,

the vehicle behavior prediction device includes a vehicle behavior prediction unit configured to:

collect the traveling state information,

input a plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value obtained based on the collected traveling state information, and

predict vehicle behavior of a prediction-subject vehicle based on whether the at least one value concerning the prediction-subject vehicle belongs to one of the plurality of kinds of probability distributions,

it is determined whether the at least one value belongs to the probability distribution depending upon whether the at least one value is included in one continuous region in which occurrence probability in the probability distribution is greater than zero,

the traveling state information further includes a traveling location in which the at least one value was detected,

the traveling location includes one of traffic elements including an intersection, a curve, and a temporary stop position,

the vehicle behavior prediction unit is further configured to;

input the plurality of kinds of probability distributions as probability distributions obtained for each traveling location, and

predict the vehicle behavior of the prediction-subject vehicle in a case where a traveling location of the prediction-subject vehicle has characteristics that are common with one of traveling locations in which the plurality of probability distributions were detected,

the vehicle behavior prediction device further comprises a distribution calculating unit configured to obtain a plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value that is input to the vehicle behavior prediction unit,

information concerning a characteristic of the traveling location is included in the traveling state information,

the distribution calculating unit is further configured to obtain the plurality of kinds of probability distributions for each of the characteristic of the traveling location, and

when the vehicle behavior prediction unit predicts the vehicle behavior, the vehicle behavior prediction unit identifies the traveling location of the prediction-subject vehicle and a characteristic of the traveling location, and uses, for predicting the vehicle behavior, a plurality of kinds of probability distributions obtained by the distribution calculating unit concerning a traveling location and a characteristic of the traveling location that is common with the result of the identification.

9. A vehicle behavior prediction device for predicting vehicle behavior based on traveling state information representing a traveling state of a vehicle, wherein

the traveling state information includes at least one value of traveling speed of the vehicle and acceleration of the vehicle,

the vehicle behavior prediction device includes a vehicle behavior prediction unit configured to:

collect the traveling state information,

input a plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value obtained based on the collected traveling state information, and

predict vehicle behavior of a prediction-subject vehicle based on whether the at least one value concerning the prediction-subject vehicle belongs to one of the plurality of kinds of probability distributions,

it is determined whether the at least one value belongs to the probability distribution depending upon whether the at least one value is included in one continuous region in which occurrence probability in the probability distribution is greater than zero,

the traveling state information further includes a traveling location in which the at least one value was detected,

the traveling location includes one of traffic elements including an intersection, a curve, and a temporary stop position,

the vehicle behavior prediction unit is further configured to;

input the plurality of kinds of probability distributions as probability distributions obtained for each traveling location, and

predict the vehicle behavior of the prediction-subject vehicle in a case where a traveling location of the prediction-subject vehicle has characteristics that are common with one of traveling locations in which the plurality of probability distributions were detected,

the vehicle behavior prediction device further comprises a distribution calculating configured to obtain a plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value that is input to the vehicle behavior prediction unit,

the traveling state information is acquired based on information sent from a plurality of vehicles each having a wireless communication function to a management center of a probe information communication system,

at least one of the distribution calculating unit and the vehicle behavior prediction unit is provided in the management center, and

information representing at least one of the probability distribution and a prediction result of vehicle behavior of the prediction-subject vehicle is provided from the management center as requested.

11. A vehicle behavior prediction method for predicting vehicle behavior based on traveling state information representing a traveling state of a vehicle using a vehicle behavior prediction device, wherein

the traveling state information includes at least one of traveling speed of the vehicle and acceleration of the vehicle,

the vehicle behavior prediction method comprises:

collecting the traveling state information;

predicting vehicle behavior of a prediction-subject vehicle based on whether the at least one value concerning the prediction-subject vehicle belongs to one of a plurality of kinds of probability distributions using the plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value obtained based on the collected traveling state information, and

determining whether the at least one value belongs to the probability distribution depending upon whether the at least one value is included in one continuous region in which occurrence probability in the probability distribution is greater than zero,

the traveling state information further includes a traveling location in which the at least one value was detected,

the traveling location includes one of traffic elements including an intersection, a curve, and a temporary stop position,

the vehicle behavior prediction method further comprises using the plurality of kinds of probability distributions as probability distributions obtained for each traveling location,

the predicting of the vehicle behavior of the prediction-subject vehicle is performed in a case where a traveling location of the prediction-subject vehicle has characteristics that are common with one of traveling locations in which the plurality of probability distributions were detected,

a probability distribution at occurrence of specific vehicle behavior and a probability distribution at non-occurrence of the specific vehicle behavior are used as the plurality of kinds of probability distributions,

deceleration behavior of a vehicle is selected as the vehicle behavior, and

in the predicting of the vehicle behavior, non-occurrence of the vehicle behavior is predicted using a probability distribution based on traveling speed, and occurrence of the vehicle behavior is predicted using a probability distribution based on acceleration.

12. The vehicle behavior prediction method according to claim 11, wherein in the predicting of the vehicle behavior, when the at least one value belongs to the probability distribution at non-occurrence of the specific vehicle behavior and does not belong to the probability distribution at occurrence of the specific vehicle behavior, it is predicted that the prediction-subject vehicle will not exhibit corresponding vehicle behavior.

13. The vehicle behavior prediction method according to claim 11, wherein in the predicting of the vehicle behavior, when the at least one value does not belong to the probability distribution at non-occurrence of the specific vehicle behavior or belongs to both the probability distribution at non-occurrence of the specific vehicle behavior and the probability distribution at occurrence of the specific vehicle behavior, it is determined that it is indeterminable whether vehicle behavior of the prediction-subject vehicle will not be exhibited.

14. The vehicle behavior prediction method according to claim 11, wherein in the predicting of the vehicle behavior, when the at least one value belongs to the probability distribution at occurrence of the specific vehicle behavior and does not belong to the probability distribution at non-occurrence of the specific vehicle behavior, it is predicted that the prediction-subject vehicle will exhibit corresponding vehicle behavior.

15. The vehicle behavior prediction method according to claim 11, wherein in the predicting of the vehicle behavior, when the at least one value does not belong to the probability distribution at occurrence of the specific vehicle behavior or belongs to both the probability distribution at occurrence of the specific vehicle behavior and the probability distribution at non-occurrence of the specific vehicle behavior, it is determined that it is indeterminable whether the prediction-subject vehicle will exhibit corresponding vehicle behavior.

16. A vehicle behavior prediction method for predicting vehicle behavior based on traveling state information representing a traveling state of a vehicle using a vehicle behavior prediction device, wherein

the traveling state information includes at least one of traveling speed of the vehicle and acceleration of the vehicle,

the vehicle behavior prediction method comprises:

collecting the traveling state information;

predicting vehicle behavior of a prediction-subject vehicle based on whether the at least one value concerning the prediction-subject vehicle belongs to one of a plurality of kinds of probability distributions using the plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value obtained based on the collected traveling state information, and

determining whether the at least one value belongs to the probability distribution depending upon whether the at least one value is included in one continuous region in which occurrence probability in the probability distribution is greater than zero,

the traveling state information further includes a traveling location in which the at least one value was detected,

the traveling location includes one of traffic elements including an intersection, a curve, and a temporary stop position,

the vehicle behavior prediction method further comprises using the plurality of kinds of probability distributions as probability distributions obtained for each traveling location,

the predicting of the vehicle behavior of the prediction-subject vehicle is performed in a case where a traveling location of the prediction-subject vehicle has characteristics that are common with one of traveling locations in which the plurality of probability distributions were detected,

the vehicle behavior prediction method further comprises defining a predetermined zone from a center of the probability distribution representing an average value of the vehicle behavior as a reliability zone concerning occurrence probability of the vehicle behavior, and

in the predicting of the vehicle behavior, the vehicle behavior is predicted using the probability distribution in the reliability zone as the probability distribution.

17. A vehicle behavior prediction method for predicting vehicle behavior based on traveling state information representing a traveling state of a vehicle using a vehicle behavior prediction device, wherein

the traveling state information includes at least one of traveling speed of the vehicle and acceleration of the vehicle,

the vehicle behavior prediction method comprises:

collecting the traveling state information;

predicting vehicle behavior of a prediction-subject vehicle based on whether the at least one value concerning the prediction-subject vehicle belongs to one of a plurality of kinds of probability distributions using the plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value obtained based on the collected traveling state information, and

determining whether the at least one value belongs to the probability distribution depending upon whether the at least one value is included in one continuous region in which occurrence probability in the probability distribution is greater than zero,

the traveling state information further includes a traveling location in which the at least one value was detected,

the traveling location includes one of traffic elements including an intersection, a curve, and a temporary stop position,

the vehicle behavior prediction method further comprises using the plurality of kinds of probability distributions as probability distributions obtained for each traveling location,

the predicting of the vehicle behavior of the prediction-subject vehicle is performed in a case where a traveling location of the prediction-subject vehicle has characteristics that are common with one of traveling locations in which the plurality of probability distributions were detected,

a probability distribution at occurrence of specific vehicle behavior and a probability distribution at non-occurrence of the specific vehicle behavior are used as the plurality of kinds of probability distributions,

the vehicle behavior prediction method further comprises defining a predetermined zone from a center of the probability distribution representing an average value of the vehicle behavior as a reliability zone concerning occurrence probability of the vehicle behavior, and

in the predicting of the vehicle behavior, it is determined whether vehicle behavior of a prediction-subject vehicle belongs to the probability distribution based on whether the vehicle behavior belongs to the reliability zone of the probability distributions.

18. A vehicle behavior prediction method for predicting vehicle behavior based on traveling state information representing a traveling state of a vehicle using a vehicle behavior prediction device, wherein

the traveling state information includes at least one of traveling speed of the vehicle and acceleration of the vehicle,

the vehicle behavior prediction method comprises:

collecting the traveling state information;

predicting vehicle behavior of a prediction-subject vehicle based on whether the at least one value concerning the prediction-subject vehicle belongs to one of a plurality of kinds of probability distributions using the plurality of kinds of probability distributions representing occurrence probability of vehicle behavior with respect to the at least one value obtained based on the collected traveling state information, and

determining whether the at least one value belongs to the probability distribution depending upon whether the at least one value is included in one continuous region in which occurrence probability in the probability distribution is greater than zero,

the traveling state information further includes a traveling location in which the at least one value was detected,

the traveling location includes one of traffic elements including an intersection, a curve, and a temporary stop position,

the vehicle behavior prediction method further comprises using the plurality of kinds of probability distributions as probability distributions obtained for each traveling location,

the predicting of the vehicle behavior of the prediction-subject vehicle is performed in a case where a traveling location of the prediction-subject vehicle has characteristics that are common with one of traveling locations in which the plurality of probability distributions were detected,

the vehicle behavior prediction method further comprising obtaining a plurality of kinds of probability distributions representing occurrence probability of the vehicle behavior used in the predicting,

identifying an information representing a characteristic of the traveling location is selected as the traveling state information,

in the obtaining of the distribution, the plurality of kinds of probability distributions are obtained for the characteristic of the traveling location,

in the predicting of the vehicle behavior, the method further comprises identifying the traveling location of a prediction-subject vehicle and a characteristic of the traveling location, and selecting, from the probability distributions, a plurality of kinds of probability distributions concerning a traveling location and a characteristic of the traveling location that is common with an identifying result of the identifying, and

the probability distributions selected is used for predicting the vehicle behavior.