Pick-up/drop-off zone availability estimation using probabilistic model

1. A system comprising:

at least one processor of a machine; and

a memory storing instructions that cause the at least one processor to perform operations comprising:

receiving, from a vehicular autonomy system of a vehicle, a request for vehicular pick-up/drop off (PDZ) availability at a location, the request specifying an estimated time of arrival at the location;

estimating, using a probabilistic model, the PDZ availability at the location at the estimated time of arrival, the probabilistic model comprising a machine learning model that is trained based on historical training data to extract one or more features from the historical training data and compute a probabilistic estimation of a given PDZ being free or occupied based on the one or more features extracted from the historical training data;

generating a response to the request based on the estimated PDZ availability;

transmitting, to the vehicular autonomy system, the response to the request, the response indicating the estimated PDZ availability; and

causing the vehicle to navigate to the PDZ at the location according a route.

2. The system of claim 1 , wherein the historical training data comprises global position system data from one or more vehicles, wherein the estimating of the PDZ availability at the location comprises:

identifying a set of PDZs associated with the location; and

determining, using the probabilistic model, an estimated availability of the set of PDZs at the estimated time of arrival.

3. The system of claim 2 , wherein the determining the estimated availability of the set of PDZs at the estimated time of arrival comprises:

determining, using the probabilistic model, an individual estimated availability of each PDZ in the set of PDZs; and

determining the estimated availability of the set of PDZs based on the individual estimated availability of each PDZ in the set of PDZs.

4. The system of claim 2 , wherein the generating of the response comprises:

selecting a target PDZ from the set of PDZs based in part on the estimated availability of the target PDZ in the set of PDZs, wherein the response indicates the target PDZ.

5. The system of claim 4 , wherein the selecting of the target PDZ is further based on a proximity of the target PDZ to at least one other PDZ.

6. The system of claim 4 , wherein the generating of the response further comprises:

generating a set of instructions that, when received by the vehicular autonomy system, cause the vehicle autonomy system to control operation of a vehicle such that the vehicle navigates to the target PDZ.

7. The system of claim 4 , wherein the generating of the response further comprises:

generating a route for the vehicle to the target PDZ based on a current location and target location of the vehicle.

8. The system of claim 1 , wherein the operations further comprise:

obtaining data comprising one or more indicia of PDZ availability at the location;

wherein the estimating of the PDZ availability at the location at the estimated time of arrival is further based on the one or more indicia of PDZ availability at the location.

9. The system of claim 8 , wherein:

the vehicular autonomy system is a first vehicular autonomy system of a first vehicle;

the obtaining of the data comprises:

transmitting a request to a second vehicular autonomy system of a second vehicle that causes the second vehicle to travel to the location; and

detecting an indication of the PDZ availability at the location when the second vehicle is near the location.

10. The system of claim 9 , wherein:

the obtaining of the data further comprises:

causing display, on a display device inside the second vehicle, of a prompt for an occupant of the second vehicle to provide an indication of the PDZ availability at the location; and

receiving a response to the prompt comprising the indication of the PDZ availability specified by the occupant of the second vehicle.

11. The system of claim 8 , wherein:

the vehicular autonomy system is a first vehicular autonomy system of a first vehicle;

the obtaining of the data comprises accessing sensor data generated by one or more sensors of a second vehicular autonomy system of a second vehicle.

12. The system of claim 8 , further comprising:

updating the probabilistic model based on the more indicia of PDZ availability at the location.

13. The system of claim 1 , further comprising:

wherein the probabilistic model is trained based on the historical training data that comprises one or more of: user-generated information; vehicle driving logs; vehicular sensor logs; traffic information; public transit schedules; parking restrictions; global position system data from one or more vehicles; and parking spot occupancy data obtained from parking sensors.

14. A computer-implemented method comprising:

receiving, from a vehicular autonomy system of a vehicle, a request for vehicular pick-up/drop off (PDZ) availability at a location, the request specifying an estimated time of arrival at the location;

estimating, using a probabilistic model, the PDZ availability at the location at the estimated time of arrival, the probabilistic model comprising a machine learning model that is trained based on historical training data to extract one or more features from the historical training data and compute a probabilistic estimation of a given PDZ being free or occupied based on the one or more features extracted from the historical training data;

generating, using one or more processors of a machine, a response to the request based on the estimated PDZ availability;

transmitting, to the vehicular autonomy system, the response to the request, the response indicating the estimated PDZ availability; and

causing the vehicle to navigate to the PDZ at the location according a route.

15. The computer-implemented method of claim 14 , wherein the estimating of the PDZ availability at the location comprises:

identifying a set of PDZs associated with the location; and

determining, using the probabilistic model, an individual estimated availability of each PDZ in the set of PDZs; and

determining the estimated availability of the set of PDZs based on the individual estimated availability of each PDZ in the set of PDZs.

16. The computer-implemented method of claim 15 , wherein the generating of the response comprises:

selecting a target PDZ from the set of PDZs based in part on the individual estimated availability of the target PDZ, wherein the response indicates the target PDZ.

17. The computer-implemented method of claim 16 , wherein the generating of the response further comprises:

generating a route for the vehicle to the target PDZ based on a current location and target location of the vehicle; and

generating a set of instructions that, when received by the vehicular autonomy system, cause the vehicle autonomy system to control operation of a vehicle such that the vehicle travels to along the route to the target PDZ.

18. The computer-implemented method of claim 14 , further comprising:

obtaining data comprising one or more indicia of PDZ availability at the location; and

updating the probabilistic model based on the more indicia of PDZ availability at the location;

wherein the estimating of the PDZ availability at the location at the estimated time of arrival is further based on the one or more indicia of PDZ availability at the location.

19. The computer-implemented method of claim 18 , wherein:

the vehicular autonomy system is a first vehicular autonomy system of a first vehicle;

the obtaining of the data comprises:

transmitting a request to a second vehicular autonomy system of a second vehicle that causes the second vehicle to travel to the location; and

detecting an indication of the PDZ availability at the location when the second vehicle is near the location.

20. A tangible computer-readable medium storing instructions that, when executed by one or more processors of a machine, cause the one or more processors of the machine to perform operations comprising:

receiving, from a vehicular autonomy system of a vehicle, a request for vehicular pick-up/drop off (PDZ) availability at a location, the request specifying an estimated time of arrival at the location;

estimating, using a probabilistic model, the PDZ availability at the location at the estimated time of arrival, the probabilistic model comprising a machine learning model that is trained based on historical training data to extract one or more features from the historical training data and compute a probabilistic estimation of a given PDZ being free or occupied based on the one or more features extracted from the historical training data;

generating a response to the request based on the estimated PDZ availability;

transmitting, to the vehicular autonomy system, the response to the request, the response indicating the estimated PDZ availability; and

causing the vehicle to navigate to the PDZ at the location according a route.