Prediction variance estimation
Observed driveline mean and variance data are used for determining the variance of a trajectory of tracked objects for use by a host vehicle. A map of a portion of a vehicle transportation network are determined, wherein the map is comprised of observed driveline mean and variance data for one or more map points. At least one trajectory of a tracked object is predicted, wherein a trajectory includes a series of location each corresponding to a respective predicted position of the tracked object at a future time. A map-based variance is generated for the location of the trajectory using a smoothed curvature of the trajectory within the map. A control system of the vehicle operates the vehicle using the map-based variance as input.
1 . A method, comprising:
determining a map of a portion of a vehicle transportation network, wherein the map comprises observed driveline mean and variance data for one or more map points;
predicting at least one trajectory for a tracked object, wherein a trajectory of the at least one trajectory includes a series of locations, a location of the series of locations corresponding to a respective predicted position of the tracked object at a future time;
generating, for each location of the trajectory, a map-based variance based on a smoothed curvature of the trajectory and the observed driveline mean and variance data for the one or more map points; and
operating a vehicle using the map-based variance as input to a control system of the vehicle.
2 . The method of claim 1 , comprising:
updating the map-based variance of the location in response to a map point of the one or more map points corresponding to the location.
3 . The method of claim 1 , wherein the map-based variance increases to a maximum variance in response to the location corresponding to a future time greater than a maximum time threshold.
4 . The method of claim 1 , comprising:
generating, for the map-based variance, a dynamic variance in response to a change in conditions within the vehicle transportation network as the tracked object follows the trajectory, wherein the dynamic variance shifts the map-based variance laterally.
5 . The method of claim 4 , comprising:
increasing the map-based variance, for a location, to a maximum variance, wherein the location corresponds to a future time greater than a maximum time threshold.
6 . The method of claim 5 , wherein the maximum variance is between 0 and 1 and the maximum time threshold is between 2 and 3 seconds.
7 . The method of claim 1 , wherein the at least one trajectory for the tracked object is based on an in-lane assessment.
8 . The method of claim 1 , wherein the at least one trajectory for the tracked object is based on an inferred relationship with nearby tracked objects.
9 . The method of claim 1 , wherein the map-based variances comprise a left variance and a right variance.
10 . The method of claim 1 , wherein generating the map-based variance comprises:
determining a previous curvature and a current curvature for the trajectory;
calculating the smoothed curvature based on the previous curvature and the current curvature; and
calculating the map-based variance based on the smoothed curvature, a variance factor, and a fixed variance.
11 . An apparatus, comprising:
a memory; and
a processor configured to execute instructions stored in the memory to:
determine a map of a portion of a vehicle transportation network, wherein the map comprises observed driveline mean and variance data for one or more map points;
predict at least one trajectory for a tracked object, wherein a trajectory of the at least one trajectory includes a series of locations, a location of the series of locations corresponding to a respective predicted position of the tracked object at a future time;
generate, for each location of the trajectory, a map-based variance based on a smoothed curvature of the trajectory and the observed driveline mean and variance data for the one or more map points;
increase the map-based variance, for a location, to a maximum variance, wherein the location corresponds to a future time greater than a maximum time threshold; and
operate a vehicle using the map-based variance as input to a control system of the vehicle.
12 . The apparatus of claim 11 , wherein the processor is configured to execute instructions stored in the memory to:
update the map-based variance of the location in response to a map point of the one or more map points corresponding to the location.
13 . The apparatus of claim 11 , wherein the map-based variance increases to a maximum variance in response to the location corresponding to a future time greater than a maximum time threshold.
14 . The apparatus of claim 11 , wherein the processor is configured to execute instructions stored in the memory to:
generate, for the map-based variance, a dynamic variance in response to a change in a condition within the vehicle transportation network as the tracked object follows the trajectory, wherein the dynamic variance shifts the map-based variance laterally.
15 . The apparatus of claim 11 , wherein generating the map-based variance comprises:
determining a previous curvature and a current curvature for the trajectory;
calculating the smoothed curvature based on the previous curvature and the current curvature; and
calculating the map-based variance based on the smoothed curvature, a variance factor, and a fixed variance.
16 . A non-transitory computer-readable medium storing instructions operable to cause one or more processors to perform operations comprising:
determining a map of a portion of a vehicle transportation network, wherein the map comprises observed driveline mean and variance data for one or more map points;
predicting at least one trajectory for a tracked object, wherein a trajectory of the at least one trajectory includes a series of locations, a location of the series of locations corresponding to a respective predicted position of the tracked object at a future time;
determining a previous curvature and a current curvature for the trajectory;
calculating a smoothed curvature based on the previous curvature and the current curvature;
generating, for each location of the trajectory, a map-based variance based on the smoothed curvature, a variance factor, a fixed variance, and the observed driveline mean and variance data for the one or more map points; and
operating a vehicle using the map-based variance as input to a control system of the vehicle.
17 . The non-transitory computer-readable medium storing instructions of claim 16 , the operations comprising:
generating, for the map-based variance, a dynamic variance in response to a change in a condition within the vehicle transportation network as the tracked object follows the trajectory, wherein the dynamic variance shifts the map-based variance laterally.
18 . The non-transitory computer-readable medium storing instructions of claim 17 , the operations comprising:
increasing the map-based variance, for a location, to a maximum variance, wherein the location corresponds to a future time greater than a maximum time threshold.
19 . The apparatus of claim 11 , wherein the maximum variance is between 0 and 1 and the maximum time threshold is between 2 and 3 seconds.
20 . The non-transitory computer-readable medium storing instructions of claim 16 , wherein the map-based variances comprise a left variance and a right variance.