Path segment risk regression system for on-demand transportation services

• US 10,489,721 B2
• Filed: 05/23/2017
• Issued: 11/26/2019
• Est. Priority Date: 05/23/2017
• Status: Active Grant

• Current Assignee: Aurora Operations Incorporated (Aurora Innovation, Inc.)
• Original Assignee: Uatc, LLC (Uber Technologies, Inc.)
• Inventors: Kislovskiy, Dima, Bradley, David McAllister
• Primary Examiner(s): Zanelli, Michael J
• Application Number:
  US15/602,197
• Publication Number:
  US 20180341261A1
• Time in Patent Office: 917 Days
• CPC Class Codes: G01C 21/3438 Rendez-vous, i.e. searching a destination where several users can meet, and the routes to this destination for these users; Ride sharing, i.e. searching a route such that at least two users can share a vehicle for at least part of the route
• CPC Class Codes: G01C 21/3461 Preferred or disfavoured areas, e.g. dangerous zones, toll or emission zones, intersections, manoeuvre types, segments such as motorways, toll roads, ferries
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 10/047 Optimisation of routes or paths, e.g. travelling salesman problem
• CPC Class Codes: G06Q 50/40 Business processes related to the transportation industry shipping G06Q10/083

Abstract

A risk regression system can computationally determine aggregate risk values for each of a plurality of routes for each transport request of an on-demand transportation service. The aggregate risk values can be based on fractional risk quantities determined through, for example, autonomous vehicle logs form autonomous vehicles operating throughout a given region. Based on the aggregate risk values, the risk regression system can facilitate vehicle matching between autonomous vehicles and human-driven vehicles.

Claims

1. A risk regression system for an on-demand transportation service, comprising:

one or more processors; and

one or more memory resources storing instructions that, when executed by the one or more processors, cause the risk regression system to;

receive, over one or more networks, transport requests from requesting users within a given region, each of the transport requests indicating a pick-up location and a destination;

for each transport request, obtain a set of possible routes between the pick-up location and the destination;

for each transport request, based on the set of possible routes between the pick-up location and the destination, compute an aggregate risk quantity for each possible route in the set, each aggregate risk quantity comprising a set of fractional risk quantities of respective path segments of a route; and

determine whether to enable autonomous vehicles (AVs) to service the transport request based on the aggregate risk quantities.

2. The risk regression system of claim 1, wherein the executed instructions further cause the risk regression system to:

collect log data from AVs operating throughout the given region; and

based, at least in part on the log data, determine a fractional risk quantity for each path segment on which the AVs operate.

3. The risk regression system of claim 2, wherein the executed instructions further cause the risk regression system to:

for each determination of a fractional risk quantity, temporally correlate the fractional risk quantity to a current set of risk conditions at a time of the determination;

wherein the executed instructions cause the risk regression system to compute the aggregate risk quantity for each of the set of possible routes based on a current set of risk conditions at a time of the transport request.

4. The risk regression system of claim 3, wherein the current set of risk conditions comprise at least one of current environmental conditions, current traffic conditions, current path surface conditions, predicted environmental conditions, predicted traffic conditions, or predicted path surface conditions.

5. The risk regression system of claim 3, wherein the executed instructions further cause the risk regression system to:

correlate the fractional risk quantity to a static set of risk parameters corresponding to nominal environmental conditions and nominal path conditions;

wherein the executed instructions cause the risk regression system to compute the aggregate risk quantity for each of the possible set of routes based on the static set of risk parameters when the nominal environmental conditions and the nominal path conditions exist at a time of the transport request.

6. The risk regression system of claim 5, wherein the static set of risk parameters comprise at least one of path geometry and path complexity for AVs.

7. The risk regression system of claim 5, wherein the nominal environmental conditions comprise a nominal risk range corresponding to typical external entity behavior at a particular location.

8. The risk regression system of claim 2, wherein each path segment comprises a lane segment of a set of capability-in-scope lanes on which the AVs are operable within the given region.

9. The risk regression system of claim 8, wherein the capability-in-scope lanes correspond to a set of localization maps accessed by the AVs to navigate throughout the given region along the capability-in-scope lanes.

10. The risk regression system of claim 2, wherein the executed instructions cause the risk regression system to further determine the fractional risk quantity for each path segment based on received data from human-driven vehicles.

11. The risk regression system of claim 2, wherein the executed instructions cause the risk regression system to further determine the fractional risk quantity for each path segment based on off-vehicle replay of AV-logged data through new software, test track evaluation of the current system-under-test, actuarial statistics, and driving research publications.

12. The risk regression system of claim 1, wherein the computed aggregate risk quantity by the risk regression system is applicable to a set of AVs having common hardware resources, and wherein the executed instructions cause the risk regression system to determine whether to instruct an AV to service the transport request from the set of AVs having common hardware resources.

13. The risk regression system of claim 12, wherein the common hardware resources comprise a common set of sensors.

14. The risk regression system of claim 1, wherein the executed instructions further cause the risk regression system to:

when the aggregate risk quantity is below a predetermined risk threshold, enable the AVs to service the transport request.

15. The risk regression system of claim 1, wherein the executed instructions further cause the risk regression system to:

collect historical data identifying failed rides for the on-demand transportation service, each failed ride corresponding to a determined route;

wherein the executed instructions further cause the risk regression system to compute the aggregate risk quantity for each possible route by determining a failure probability for the possible route based on the historical data across the possible route.

16. The risk regression system of claim 15, wherein the aggregate risk quantity for the possible route comprises a weighted sum between the failure probability and the set of fractional risk quantities of respective path segments for the possible route.

17. A non-transitory computer readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to:

receive, over one or more networks, transport requests from requesting users within a given region, each of the transport requests indicating a pick-up location and a destination;

for each transport request, obtain a set of possible routes between the pick-up location and the destination;

for each transport request, based on the set of possible routes between the pick-up location and the destination, compute an aggregate risk quantity comprising a set of fractional risk quantities of respective path segments for each possible route in the set; and

determine whether to enable autonomous vehicles (AVs) to service the transport request based on the aggregate risk quantities.

18. The non-transitory computer readable medium of claim 17, wherein the executed instructions further cause the one or more processors to:

collect log data from AVs operating throughout the given region; and

based, at least in part on the log data, determine a fractional risk quantity for each path segment on which the AVs operate.

19. The non-transitory computer readable medium of claim 18, wherein the executed instructions further cause the one or more processors to:

for each determination of a fractional risk quantity, temporally correlate the fractional risk quantity to a current set of risk conditions at a time of the determination;

wherein the executed instructions cause the one or more processors to compute the aggregate risk quantity for each of the set of possible routes based on a current set of risk conditions at a time of the transport request.

20. A computer-implemented method of facilitating on-demand transportation, the method being performed by one or more processors and comprising:

receiving, over one or more networks, transport requests from requesting users within a given region, each of the transport requests indicating a pick-up location and a destination;

obtaining, for each transport request, a set of possible routes between the pick-up location and the destination;

for each transport request, based on the set of possible routes between the pick-up location and the destination, computing an aggregate risk quantity comprising a set of fractional risk quantities of respective path segments for each possible route in the set; and

determining whether to enable autonomous vehicles (AVs) to service the transport request based on the aggregate risk quantities.