Host vehicle operation using remote vehicle intention prediction

• US 10,152,882 B2
• Filed: 11/30/2015
• Issued: 12/11/2018
• Est. Priority Date: 11/30/2015
• Status: Active Grant

• Current Assignee: Nissan Motor Co., Ltd.
• Original Assignee: Nissan North America Incorporated (Nissan Motor Co., Ltd.)
• Inventors: Zhao, Yue, Mortazavi, Ali
• Primary Examiner(s): Smith, Jelani A
• Assistant Examiner(s): Alsomiri, Majdi
• Application Number:
  US14/954,083
• Publication Number:
  US 20170154529A1
• Time in Patent Office: 1,107 Days
• CPC Class Codes: B60W 2556/00 Input parameters relating to data
• CPC Class Codes: B60W 30/12 Lane keeping
• CPC Class Codes: G08G 1/0112 from the vehicle, e.g. floating car data [FCD]
• CPC Class Codes: G08G 1/0133 for classifying traffic situation
• CPC Class Codes: G08G 1/0141 for traffic information dissemination
• CPC Class Codes: G08G 1/0145 for active traffic flow control
• CPC Class Codes: G08G 1/07 Controlling traffic signals
• CPC Class Codes: G08G 1/096725 where the received information generates an automatic action on the vehicle control
• CPC Class Codes: G08G 1/096775 where the origin of the information is a central station
• CPC Class Codes: G08G 1/096783 where the origin of the information is a roadside individual element
• CPC Class Codes: G08G 1/096791 where the origin of the information is another vehicle
• CPC Class Codes: G08G 1/161 Decentralised systems, e.g. inter-vehicle communication
• CPC Class Codes: G08G 1/163 involving continuous checking
• CPC Class Codes: G08G 1/164 Centralised systems, e.g. external to vehicles
• CPC Class Codes: G08G 1/167 Driving aids for lane monitoring, lane changing, e.g. blind spot detection

Abstract

A host vehicle receives remote vehicle spatial state information for a remote vehicle and identifies vehicle transportation network information representing a portion of a transportation network based on that information. At least one initial probability value is generated based on comparing the spatial state information and the transportation network information at an initial time point, each initial probability value indicating a likelihood that the remote vehicle is following a lane within the transportation network. A deviation between adjacent values for the spatial state information relative to the transportation network information is generated for a plurality of time points. For each single lane and deviation, the likelihood that the remote vehicle is following the lane using a new probability value based on the deviation is updated, a trajectory using the updated likelihood is generated, and the host vehicle traverses the transportation network using the transportation network information and trajectory.

Claims

1. A method of generating projected vehicle information for use in traversing a vehicle transportation network, the method comprising:

receiving, by at least one of a sensor or an electronic communication interface of a host vehicle, remote vehicle spatial state information for a remote vehicle;

identifying vehicle transportation network information representing a portion of the vehicle transportation network based on the remote vehicle spatial state information;

generating, by a processor in response to instructions stored on a non-transitory computer readable medium, at least one initial probability value based on a comparison of the remote vehicle spatial state information and the vehicle transportation network information at an initial time point, each initial probability value indicating a likelihood that the remote vehicle is following a single lane within the vehicle transportation network information;

generating, for a plurality of time points including the initial time point forming a period of time, a deviation of the remote vehicle between adjacent values for the remote vehicle spatial state information relative to the vehicle transportation network information;

generating, for each single lane and for each deviation, a new probability value that corresponds to the deviation;

updating, for each single lane and for each deviation, the likelihood that the remote vehicle is following the single lane by modifying a previous probability value at a previous time point of the plurality of time points, starting with the initial probability value, using the new probability value of a current time point;

generating, for the remote vehicle, a trajectory using the likelihood as updated; and

traversing, by the host vehicle, the vehicle transportation network using the vehicle transportation network information and the trajectory for the remote vehicle.

2. The method of claim 1, wherein identifying vehicle transportation network information representing a portion of the vehicle transportation network based on the remote vehicle spatial state information comprises:

identifying at least one lane that the remote vehicle may traverse by comparing the remote vehicle spatial state information with the vehicle transportation network.

3. The method of claim 2, further comprising:

receiving, at the host vehicle, remote vehicle kinematic state information for the remote vehicle over the period of time; and

wherein

generating the trajectory using the likelihood as updated comprises generating the trajectory using the remote vehicle kinematic state information.

4. The method of claim 3, wherein the remote vehicle spatial state information includes spatial coordinates for the remote vehicle, and the remote vehicle kinematic state information includes at least one of a remote vehicle velocity for the remote vehicle or a remote vehicle acceleration for the remote vehicle.

5. The method of claim 1, further comprising:

receiving, at the host vehicle, remote vehicle kinematic state information for the remote vehicle over the period of time; and

wherein

generating the trajectory using the likelihood as updated comprises;

generating, for each specific lane, a respective trajectory to form a plurality of trajectories projected out from a current position of the remote vehicle using the remote vehicle kinematic state information, the plurality of trajectories extending into a same future point in time.

6. The method of claim 1, wherein generating the at least one initial probability value comprises:

comparing the remote vehicle spatial state information and the vehicle transportation network information to define a current number of lanes available to the remote vehicle within the vehicle transportation network information at the initial time point; and

wherein

each initial probability value is generated using the current number of lanes.

7. The method of claim 1, wherein generating the trajectory using the likelihood as updated comprises:

generating a set of first trajectory values using a current velocity and heading of the remote vehicle;

generating a set of second traj ectory values using the likelihood as updated; and

generating the trajectory by combining the set of first trajectory values and the set of second trajectory values.

8. The method of claim 7, wherein combining the set of first trajectory values and the set of second trajectory values comprises:

applying a cosine function to weight each first trajectory value of the set of first trajectory values with a respective second trajectory value of the set of second trajectory values at an end of each of a plurality of time intervals for an interpolation time.

9. The method of claim 1, wherein the single lane is one of a plurality of lanes, the method further comprising, for each lane of the plurality of lanes:

defining a centerline for the lane; and

for each current time point of the plurality of time points;

receiving a current value for the remote vehicle spatial state information, the current value associated with a current position of the remote vehicle at the current time point; and

one of;

projecting a centerline of the remote vehicle in parallel with the centerline for the lane from a previous value for the remote vehicle spatial state information to a Euclidean line extending between the current value for the remote vehicle spatial state information and the centerline; and

generating the deviation as a relative deviation of the remote vehicle from the centerline for the lane based on a Euclidean distance between the current value for the remote vehicle and the centerline of the remote vehicle on the Euclidean line;

or

projecting a centerline of the remote vehicle in parallel with the centerline for the lane from a Euclidean line extending between a previous value for the remote vehicle spatial state information and the centerline to the current value for the remote vehicle spatial state information; and

generating the deviation as a relative deviation of the remote vehicle from the centerline for the lane based on a Euclidean distance between the previous value for the remote vehicle and the centerline of the remote vehicle on the Euclidean line;

wherein the new probability value is based on the relative deviation.

10. The method of claim 9, wherein updating the initial probability value using the new probability value based on the relative deviation further comprises:

relating the relative deviation to the new probability value using a cumulative distribution function of a Gaussian distribution.

11. A method of generating projected vehicle information for use in traversing a vehicle transportation network, the method comprising:

receiving, at a host vehicle, remote vehicle spatial state information for a remote vehicle;

identifying vehicle transportation network information representing a portion of the vehicle transportation network based on the remote vehicle spatial state information;

generating, by a processor in response to instructions stored on a non-transitory computer readable medium, at least one probability value based on a comparison of the remote vehicle spatial state information and the vehicle transportation network information at an initial time point, each probability value associated with a respective lane within the vehicle transportation network information indicating a likelihood that the remote vehicle is following the lane within the vehicle transportation network information;

for each of a plurality of time points after the initial time point and for each lane within the vehicle transportation network information;

generating a deviation between a current position of the remote vehicle relative to a centerline of the lane at a current time point and a previous position of the remote vehicle at a preceding time point relative to the centerline of the lane;

determining a probability of deviation from the previous position based on the deviation;

updating the probability value for the previous time point using the probability of deviation to obtain the probability value for the current time point;

generating, for the remote vehicle, a trajectory using the probability value for the current time point; and

traversing, by the host vehicle, the vehicle transportation network using the vehicle transportation network information and the trajectory for the remote vehicle.

12. The method of claim 11, wherein updating the probability value comprises:

generating the probability value for the current time point by dividing the probability value for the previous time point by the probability of deviation when the remote vehicle is closer to the centerline at the current time point than at the previous time point but did not cross the centerline between the previous time point and the current time point;

generating the probability value for the current time point by multiplying the probability value for the previous time point by the probability of deviation when the remote vehicle is further away from the centerline at the current time point than at the previous time point but did not cross the centerline between the previous time point and the current time point; and

when the remote vehicle crossed the centerline between the previous time point and the current time point, generating the probability value for the current time point by;

dividing, to generate a product, the previous probability value for the previous time point by a first probability value based on a portion of the deviation attributable to the remote vehicle traveling from the previous position at the previous time point to the centerline; and

multiplying, to generate the probability value for the current time point, the product by a second probability value based on a portion of the deviation attributable to the remote vehicle traveling from the centerline to a current position at the current time point.

13. The method of claim 11, wherein:

the deviation is a positive value when the remote vehicle is moving in a first direction relative to a centerline of the lane; and

the deviation is a negative value when the remote vehicle is moving in a second direction relative to the centerline of the single lane, the second direction being opposite to the first direction.

14. An apparatus for generating projected vehicle information for use in traversing a vehicle transportation network, the apparatus comprising:

a non-transitory memory; and

a processor configured to execute instructions stored in the non-transitory memory to;

receive remote vehicle spatial state information for a remote vehicle;

identify vehicle transportation network information representing a portion of the vehicle transportation network based on the remote vehicle spatial state information;

generate at least one initial probability value based on a comparison of the remote vehicle spatial state information and the vehicle transportation network information at an initial time point, each initial probability value indicating a likelihood that the remote vehicle is following a single lane within the vehicle transportation network information;

generate, for a plurality of time points including the initial time point that forms a period of time, a deviation of the remote vehicle between adjacent values for the remote vehicle spatial state information relative to the vehicle transportation network information;

generate, for each single lane and for each deviation, a new probability value that corresponds to the deviation;

update, for each single lane and for each deviation, the likelihood that the remote vehicle is following the single lane by modifying a previous probability value at a previous time point of the plurality of time points, starting with the initial probability value, using the new probability value of a current time point;

generate, for the remote vehicle, a trajectory using the likelihood as updated; and

traverse, using a host vehicle, the vehicle transportation network using the vehicle transportation network information and the trajectory for the remote vehicle.

15. The apparatus of claim 14, wherein the processor is configured to identify the vehicle transportation network information representing a portion of the vehicle transportation network based on the remote vehicle spatial state information by:

identifying a plurality of lanes, each of the plurality of lanes being one of;

a current lane on which the remote vehicle is traveling at the initial time point;

an adjacent lane adjacent to the current lane;

a previous lane occurring before the current lane;

or

a future lane occurring after the current lane.

16. The apparatus of claim 14, wherein the processor is configured to:

receive remote vehicle kinematic state information for the remote vehicle over the period of time; and

generate the trajectory using the likelihood as updated by generating the trajectory using the remote vehicle kinematic state information.

17. The apparatus of claim 16, wherein the processor is configured to receive the remote vehicle kinematic state information within a wireless signal transmitted from the remote vehicle.

18. The apparatus of claim 16, wherein the processor is configured to receive the remote vehicle kinematic state information using at least one of a sensor or an electronic communication interface of the host vehicle.

19. The apparatus of claim 14, wherein the processor is configured to generate the at least one initial probability value by:

comparing the remote vehicle spatial state information and the vehicle transportation network information to define a current number of lanes available to the remote vehicle within the vehicle transportation network information at the initial time point;

wherein

each initial probability value is equal to one divided by the current number of lanes; and

wherein

the current number of lanes is greater than one.

20. The apparatus of claim 19, wherein the processor is configured to generate respective values for the deviation for each lane and is configured to update the likelihood using the new probability value, for each lane, by:

generating a current probability value for a current time point of the plurality of time points by dividing a previous probability value for a previous time point within the plurality of time points by the new probability value when the remote vehicle is closer to a centerline of the lane at the current time point than at the previous time point but did not cross the centerline between the previous time point and the current time point;

generating the current probability value for the current time point of the plurality of time points by multiplying the previous probability value for the previous time point within the plurality of time points by the new probability value when the remote vehicle is further away from the centerline at the current time point than at the previous time point but did not cross the centerline between the previous time point and the current time point; and

when the remote vehicle crossed the centerline between the previous time point and the current time point, generating the current probability value for the current time point of the plurality of time points by;

dividing, to generate a product, the previous probability value for the previous time point within the plurality of time points by a first probability value based on a portion of the deviation attributable to the remote vehicle traveling from a starting position at the previous time point to the centerline; and

multiplying, to generate the current probability value, the product by a second probability value based on a portion of the deviation attributable to the remote vehicle traveling from the centerline to a current position at the current time point.

21. The apparatus of claim 20, wherein the processor is configured to normalize the current probability value at each of the plurality of time points based on the current number of lanes before generating the trajectory for each lane.