Visualization of planned autonomous vehicle behavior

1. A computer-implemented method for visualizing planned behavior of an autonomous vehicle (AV), the method comprising:

receiving data describing a planned pathway of the AV along a roadway, the data describing the planned pathway comprising a plurality of velocities of the AV at a corresponding plurality of points along the planned pathway;

receiving object data describing a pedestrian in an environment of the AV, the object data comprising a location of the pedestrian;

determining a predicted pathway of the pedestrian based on the location, the predicted pathway entering the roadway and crossing the planned pathway of the AV at a cross point along the roadway;

based on the data describing the planned pathway of the AV and the predicted pathway of the pedestrian, determining that the pedestrian is predicted to reach the cross point before the AV;

classifying the pedestrian as an asserting object in response to determining that the pedestrian is predicted to reach the cross point before the AV; and

generating an image illustrating planned behavior of the AV and planned behavior of the pedestrian, the image comprising the planned pathway of the AV and the pedestrian in the environment of the AV, wherein the pedestrian in the image has a visual characteristic that indicates that the pedestrian is classified as an asserting object and that the pedestrian is predicted to reach the cross point before the AV.

2. The method of claim 1 , wherein the object in the image has a first visual characteristic, the image further comprises a second object classified as a yielding object, and the second object has a second visual characteristic different from the first visual characteristic.

3. The method of claim 2 , wherein the first visual characteristic is a first color, and the second visual characteristic is a second color.

4. The method of claim 1 , further comprising:

identifying a rule for regulating flow of traffic that governs the predicted pathway and the planned pathway; and

predicting whether the object reaches the cross point before the AV or after the AV by applying the identified rule to the predicted pathway and the planned pathway.

5. The method of claim 1 , further comprising:

receiving updated object data comprising an updated predicted pathway of the object;

based on the updated object data, revising the classification of the object a yielding object; and

updating the visual characteristic of the object in the image to reflect the revised classification of the object.

6. The method of claim 1 , further comprising:

in response to the AV approaching an intersection having a traffic light, receiving traffic light status data describing a current state of the traffic light and a predicted remaining duration of the current state of the traffic light,

wherein generating the image further comprises generating a traffic light status indicator, the traffic light status indicator comprising a light timer indicating the predicted remaining duration.

7. The method of claim 6 , wherein the traffic light status data comprises a predicted remaining duration of a yellow light, and the traffic light status indicator comprises a light timer indicating the predicted remaining duration of the yellow light.

8. The method of claim 1 , wherein the predicted pathway is a first predicted pathway, the method further comprising:

determining a second predicted pathway of the pedestrian, the second predicted pathway not crossing the planned pathway of the AV at a cross point;

wherein the pedestrian is classified as an asserting object based on one of the first predicted pathway or the second predicted pathway crossing the planned pathway of the AV.

9. The method of claim 1 , further comprising:

determining that the pedestrian is at a crosswalk; and

determining that the pedestrian is predicted to reach the cross point before the AV further based on the planned pathway intersecting the crosswalk, wherein the cross point is along the crosswalk.

10. The method of claim 9 , further comprising:

receiving crosswalk signal data describing a crosswalk signal controlling movement of the pedestrian across the crosswalk; and

determining that the pedestrian is predicted to reach the cross point before the AV further based on based on the crosswalk signal data.

11. A non-transitory computer-readable medium storing instructions for visualizing planned behavior of an autonomous vehicle (AV), the instructions, when executed by a processor, cause the processor to:

receive data describing a planned pathway of the AV along a roadway, the data describing the planned pathway comprising a plurality of velocities of the AV at a corresponding plurality of points along the planned pathway;

receive object data describing a pedestrian in an environment of the AV, the object data comprising a location of the pedestrian;

determine a predicted pathway of the pedestrian based on the location, the predicted pathway entering the roadway and crossing the planned pathway of the AV at a cross point along the roadway;

based on the data describing the planned pathway of the AV and the predicted pathway of the pedestrian, determine that the pedestrian is predicted to reach the cross point before the AV;

classify the pedestrian as an asserting object in response to determining that the pedestrian is predicted to reach the cross point before the AV; and

generate an image illustrating planned behavior of the AV and planned behavior of the pedestrian, the image comprising the planned pathway of the AV and the pedestrian in the environment of the AV, wherein the pedestrian in the image has a visual characteristic that indicates that the pedestrian is classified as an asserting object and that the pedestrian is predicted to reach the cross point before the AV.

12. The non-transitory computer-readable medium of claim 11 , wherein the object in the image has a first visual characteristic, the image further comprises a second object classified as a yielding object, and the second object has a second visual characteristic different from the first visual characteristic.

13. The non-transitory computer-readable medium of claim 12 , wherein the first visual characteristic is a first color, and the second visual characteristic is a second color.

14. The non-transitory computer-readable medium of claim 11 , wherein the instructions further cause the processor to:

identify a rule for regulating flow of traffic that governs the predicted pathway and the planned pathway; and

predict whether the object reaches the cross point before the AV or after the AV by applying the identified rule to the predicted pathway and the planned pathway.

15. The non-transitory computer-readable medium of claim 11 , wherein the instructions further cause the processor to:

receive traffic light status data describing a current state of a traffic light and a predicted remaining duration of the current state of the traffic light in response to the AV approaching an intersection having a traffic light,

wherein generating the image further comprises generating a traffic light status indicator, the traffic light status indicator comprising a light timer indicating the predicted remaining duration.

16. A computer-implemented system for visualizing planned behavior of an autonomous vehicle (AV), the system comprising:

a path planning system configured to generate a planned pathway of the AV along a roadway, the planned pathway comprising a plurality of velocities of the AV at a corresponding plurality of points along the planned pathway;

an object prediction engine configured to:

receiving object data describing a pedestrian in an environment of the AV, the object data comprising a location of the pedestrian;

generate a predicted pathway of the pedestrian based on the location, the predicted pathway entering the roadway and crossing the planned pathway of the AV at a cross point along the roadway; and

based on the data describing the planned pathway of the AV and the predicted pathway of the pedestrian, determine that the pedestrian is predicted to reach the cross point before the AV; and

a user interface engine configured to:

classify the pedestrian as an asserting object in response to determining that the pedestrian is predicted to reach the cross point before the AV; and

generate an image illustrating planned behavior of the AV and planned behavior of the pedestrian, the image comprising the planned pathway of the AV and the pedestrian in the environment of the AV, wherein the pedestrian in the image has a visual characteristic that indicates that the pedestrian is classified as an asserting object and that the pedestrian is predicted to reach the cross point before the AV.

17. The system of claim 16 , wherein the user interface engine generates the object in the image with a first visual characteristic, the image further comprises a second object classified as a yielding object, and the second object has a second visual characteristic different from the first visual characteristic.

18. The system of claim 16 , wherein the user interface engine is further configured to:

receive traffic light status data describing a current state of a traffic light and a predicted remaining duration of the current state of the traffic light, and

generate a traffic light status indicator, the traffic light status indicator comprising a light timer indicating the predicted remaining duration.

19. The system of claim 16 , wherein the object prediction engine is further configured to:

identify a rule for regulating flow of traffic that governs the predicted pathway and the planned pathway; and

predict whether the object reaches the cross point before the AV or after the AV by applying the identified rule to the predicted pathway and the planned pathway.

20. The system of claim 16 , wherein the user interface engine is further configured to:

revise the classification of the object a yielding object based on an updated prediction from the object prediction engine; and

update the visual characteristic of the object in the image to reflect the revised classification of the object.