Stranding and scoping analysis for autonomous vehicle services
Aspects of the disclosure provide for identifying problematic areas within a service area for an autonomous vehicle transportation service. For instance, a starting location within the service area corresponding to a potential pickup location for passengers or cargo for the service may be identified. A destination within the service area may be identified. A simulation may be run in order to determine a route for a simulated vehicle to travel between the starting location and the destination. That the route includes a particular type of maneuver may be determined. A new simulation without allowing the simulated vehicle to complete the particular type of maneuver may be run. Whether the simulated vehicle reaches the destination in the new simulation may be determined. Based on the determination of whether the simulated vehicle reaches the destination in the new simulation, the starting location and destination location may be flagged as potentially problematic areas.
1. A method of identifying problematic areas within a service area for an autonomous vehicle transportation service, the method comprising:
identifying, by one or more processors, a starting location within the service area corresponding to a potential pickup location for passengers or cargo for the service;
identifying, by the one or more processors, a destination location within the service area;
running, by the one or more processors, a simulation to determine a route for a simulated vehicle to travel between the starting location and the destination location;
determining, by the one or more processors, that the determined route includes a particular type of maneuver;
running, by the one or more processors, a new simulation to determine a new route for the simulated vehicle to travel between the starting location and the destination location without allowing the simulated vehicle to complete the particular type of maneuver;
determining, by the one or more processors, whether the simulated vehicle reaches the destination location in the new simulation; and
based on a determination by the one or more processors that the simulated vehicle does not reaches the destination location in the new simulation, flagging, by the one or more processors, at least one of the starting location or the destination location as potentially problematic areas.
2. The method of claim 1 , further comprising:
running a set of simulations for the starting location by identifying a plurality of destination locations for the starting location and determining routes between the starting location and each of the plurality of destination locations;
identifying ones of the determined routes that include the particular type of maneuver;
running a set of new simulations without allowing the simulated vehicle to complete the particular type of maneuver; and
determining whether the simulated vehicle reaches the destination location in each of the set of new simulations, and wherein flagging is further based on the determination of whether the simulated vehicle reaches the destination location in each of the set of new simulations.
3. The method of claim 2 , wherein flagging the starting location is further based on a comparison of a number of new simulations where the simulated vehicle does not reach the destination location to a threshold value.
4. The method of claim 1 , wherein the service area includes a boundary across which the simulated vehicle is unable to cross, and wherein determining whether the simulated vehicle reaches the destination location in the new simulation is based on whether the simulated vehicle becomes stranded by the boundary.
5. The method of claim 1 , wherein the starting location is identified by randomly selecting a location within the service area.
6. The method of claim 1 , wherein the starting location is identified from a plurality of predetermined pickup and drop off locations within the service area.
7. The method of claim 6 , wherein the starting location is identified from the plurality of predetermined pickup and drop off locations randomly.
8. The method of claim 6 , wherein the starting location is identified from the plurality of predetermined pickup and drop off locations in order to test a specific area within the service area.
9. The method of claim 1 , wherein the destination location is identified from a plurality of predetermined pickup and drop off locations within the service area.
10. The method of claim 1 , wherein the simulation is run using a routing system software stack for an autonomous vehicle to determine the determined route.
11. The method of claim 1 , wherein the particular type of maneuver is changing lanes.
12. The method of claim 1 , further comprising:
flagging a plurality of problematic areas; and
drawing a polygon around the plurality of problematic areas and the starting location.
13. A system for identifying problematic areas within a service area for an autonomous vehicle transportation service, the system comprising:
one or more server computing devices having one or more processors configured to:
identify a starting location within the service area corresponding to a potential pickup location for passengers or cargo for the service;
identify a destination location within the service area;
run a simulation to determine a route for a simulated vehicle to travel between the starting location and the destination location;
determine that the determined route includes a particular type of maneuver;
run a new simulation to determine a new route for the simulated vehicle to travel between the starting location and the destination location without allowing the simulated vehicle to complete the particular type of maneuver;
determine whether the simulated vehicle reaches the destination location in the new simulation; and
based on a determination that the simulated vehicle does not reach the destination location in the new simulation, flag at least one of the starting location or destination location as potentially problematic areas.
14. The system of claim 13 , wherein the one or more processors are further configured to:
run a set of simulations for the starting location by identifying a plurality of destination locations for the starting location and determining routes between the starting location and each of the plurality of destination locations;
identify ones of the determined routes that include the particular type of maneuver;
run a set of new simulations without allowing the simulated vehicle to complete the particular type of maneuver; and
determine whether the simulated vehicle reaches the destination location in each of the set of new simulations, and wherein flagging is further based on the determination of whether the simulated vehicle reaches the destination location in each of the set of new simulations.
15. The system of claim 13 , wherein the service area includes a boundary across which the simulated vehicle is unable to cross, and wherein determining whether the simulated vehicle reaches the destination location in the new simulation is based on whether the simulated vehicle becomes stranded by the boundary.
16. The system of claim 13 , wherein the starting location is identified by randomly selecting a location within the service area.
17. The system of claim 13 , wherein the starting location is identified from a plurality of predetermined pickup and drop off locations within the service area.
18. The system of claim 13 , wherein the destination location is identified from a plurality of predetermined pickup and drop off locations within the service area.
19. The system of claim 13 , wherein the simulation is run using a routing system software stack for an autonomous vehicle to determine the determined route.
20. The system of claim 13 , wherein the one or more processors are further configured to:
flag a plurality of problematic areas; and
draw a polygon around the plurality of problematic areas and the starting location.