Behavior prediction device

1. A behavior prediction device comprising:

a map database that stores map information;

a behavior prediction model database that stores a behavior prediction model for predicting moving object behavior;

a memory storing one or more instructions;

a processor configured to execute the one or more instructions to:

recognize a position of a host vehicle on a map;

detect moving object behavior including at least a position, an advancing direction, and a speed of a moving object around the host vehicle;

calculate a behavior prediction of the moving object at a future time point using the behavior prediction model based on the detected moving object behavior and the map information;

determine whether a prediction deviation occurs based on the behavior prediction at the future time point and the detected moving object behavior at the future time point;

estimate a deviation occurrence reason based on the behavior prediction, the map information, and the detected moving object behavior corresponding to the behavior prediction when the processor determines that the prediction deviation occurs;

determine a necessity of an update of the behavior prediction model database based on the deviation occurrence reason when the processor determines that the prediction deviation occurs;

update the behavior prediction model based on a determination that the update of the behavior prediction model is necessary; and

autonomously control the host vehicle based on the behavior prediction,

wherein the deviation occurrence reason comprises an abnormality in the map information, and

wherein the processor is further configured to determine that the update of the behavior prediction model database is not necessary and prohibits use of the map information for the behavior prediction in an area around the position of the host vehicle on the map, based on an estimation that the abnormality in the map information is the deviation occurrence reason.

2. The behavior prediction device according to claim 1 ,

wherein based on the processor determining that the prediction deviation occurs, the processor is further configured to calculate an update necessity degree of the behavior prediction model database based on the deviation occurrence reason and determine that the update of the behavior prediction model database is necessary when the update necessity degree is equal to or larger than an update threshold value.

3. The behavior prediction device according to claim 2 ,

wherein the processor is further configured to calculate at least a short-term behavior prediction which is the behavior prediction of the moving object at a short-term prediction time point set in advance and a long-term behavior prediction which is the behavior prediction of the moving object at a long-term prediction time point set in advance as a time point after the short-term prediction time point, and

wherein based on the processor determining that the prediction deviation of the short-term behavior prediction occurs in the calculation of the update necessity degree based on the same deviation occurrence reason, the processor is further configured to calculate the update necessity degree as a large value compared with when a determination is made that the prediction deviation of the short-term behavior prediction does not occur and only the prediction deviation of the long-term behavior prediction occurs.

4. The behavior prediction device according to claim 3 , wherein the map information comprises position information for each lane,

wherein the processor is further configured to:

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, the map information, and the position of the host vehicle on the map, and

wherein the behavior prediction model database is configured to store the behavior prediction model in association with the position on the map in the map information.

5. The behavior prediction device according to claim 4 ,

wherein based on the position of the moving object in the behavior prediction not being included in a moving object no entry region on the map and the position of the moving object in the detection result of the moving object behavior corresponding to the behavior prediction being included in the moving object no entry region on the map, the processor is further configured to estimate the abnormality in the map information as the deviation occurrence reason.

6. The behavior prediction device according to claim 5 , wherein the processor is further configured to:

recognize a type of the moving object; and

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, and the type of the moving object,

wherein the behavior prediction model database is configured to store the behavior prediction model in association with the type of the moving object.

7. The behavior prediction device according to claim 4 , wherein the processor is further configured to:

recognize a type of the moving object; and

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, and the type of the moving object,

wherein the behavior prediction model database is configured to store the behavior prediction model in association with the type of the moving object.

8. The behavior prediction device according to claim 3 , wherein the processor is further configured to:

recognize a type of the moving object; and

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, and the type of the moving object,

wherein the behavior prediction model database is configured to store the behavior prediction model in association with the type of the moving object.

9. The behavior prediction device according to claim 2 , wherein the map information comprises position information for each lane,

wherein the processor is further configured to:

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, the map information, and the position of the host vehicle on the map, and

wherein the behavior prediction model database is configured to store the behavior prediction model in association with the position on the map in the map information.

10. The behavior prediction device according to claim 9 ,

wherein based on the position of the moving object in the behavior prediction not being included in a moving object no entry region on the map and the position of the moving object in the detection result of the moving object behavior corresponding to the behavior prediction being included in the moving object no entry region on the map, the processor is further configured to estimate the abnormality in the map information as the deviation occurrence reason.

11. The behavior prediction device according to claim 10 , wherein the processor is further configured to:

recognize a type of the moving object; and

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, and the type of the moving object,

wherein the behavior prediction model database is configured to store the behavior prediction model in association with the type of the moving object.

12. The behavior prediction device according to claim 9 , wherein the processor is further configured to:

recognize a type of the moving object; and

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, and the type of the moving object,

wherein the behavior prediction model database is configured to store the behavior prediction model in association with the type of the moving object.

13. The behavior prediction device according to claim 2 , wherein the processor is further configured to:

recognize a type of the moving object; and

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, and the type of the moving object,

wherein the behavior prediction model database is configured to store the behavior prediction model in association with the type of the moving object.

14. The behavior prediction device according to claim 1 , wherein the map information comprises position information for each lane;

wherein the processor is further configured to:

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, the map information, and the position of the host vehicle on the map, and

wherein the behavior prediction model database is configured to store the behavior prediction model in association with the position on the map in the map information.

15. The behavior prediction device according to claim 14 ,

wherein based on the position of the moving object in the behavior prediction not being included in a moving object no entry region on the map and the position of the moving object in the detection result of the moving object behavior corresponding to the behavior prediction being included in the moving object no entry region on the map, the processor is further configured to estimate the abnormality in the map information as the deviation occurrence reason.

16. The behavior prediction device according to claim 15 , wherein the processor is further configured to:

recognize a type of the moving object and

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, and the type of the moving object,

wherein the behavior prediction model database is configured to store the behavior prediction model in association with the type of the moving object.

17. The behavior prediction device according to claim 14 , wherein the processor is further configured to:

recognize a type of the moving object; and

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, and the type of the moving object,

wherein the behavior prediction model database is configured to store the behavior prediction model in association with the type of the moving object.

18. The behavior prediction device according to claim 1 , wherein the processor is further configured to:

recognize a type of the moving object; and

calculate the behavior prediction of the moving object based on the detection result of the moving object behavior, the behavior prediction model, and the type of the moving object,

wherein the behavior prediction model database stores the behavior prediction model in association with the type of the moving object.