Vehicle selection for on-demand transportation services
An on-demand transportation management service can perform a selection process between a set of safety-driven autonomous vehicles (SDAVs), fully autonomous vehicles (FAVs), and human-driven vehicles (HDVs) to service transport requests based on a variety of selection parameters.
1. An on-demand transport system comprising:
one or more processors; and one or more memory resources storing instructions that, when executed by the one or more processors, cause the on-demand transport system to:
for a given region, receive transport requests from requesting users in connection with an on-demand transportation service, each transport request indicating a pick-up location and a destination;
for each transport request, (i) determine a risk for an optimal route between the pick-up location and the destination, and (ii) based at least in part on the risk, perform a selection process between a set of safety-driven autonomous vehicles (SDAVs), fully autonomous vehicles (FAVs), and human-driven vehicles (HDVs) to service the transport request;
select a SDAV or a FAV as an optimal vehicle from the candidate set of vehicles to service the transport request; and
in response to selecting the SDAV from the candidate set of vehicles to service the transport request:
select a specified software version from a plurality of software versions stored on the SDAV for execution while servicing the transport request; and
transmit a set of transport instructions to the SDAV indicating the specified software version for execution while servicing the transport request; and
in response to selecting the FAV from the candidate set of vehicles to service the transport request:
select a verified software version stored on the FAV for execution while servicing the transport request based on the risk; and
transmit a set of transport instructions to the FAV indicating the verified software version for execution while servicing the transport request.
2. The on-demand transport system of claim 1 , wherein the executed instructions further cause the on-demand transport system to further perform the selection process by determining a set of candidate vehicles within a certain proximity to the pick-up location, the set of candidate vehicles including at least one HDV, at least one SDAV, and at least one FAV.
3. The on-demand transport system of claim 2 , wherein the executed instructions further cause the on-demand transport system to: for each vehicle in the set of candidate vehicles, determine an estimated time of arrival (ETA) to the pick-up location indicated in the transport request; wherein the executed instructions further cause the on-demand transport system to perform the selection process based on an estimated time of arrival (ETA) of each of the set of candidate vehicles to the pick-up location.
4. The on-demand transport system of claim 3 , wherein the executed instructions further cause the on-demand transport system to perform the selection process based on a desired time range for pick-up.
5. The on-demand transport system of claim 2 , wherein the executed instructions further cause the on-demand transport system to: for each vehicle in the set of candidate vehicles, determine, based on historical data, an estimated revenue for the vehicle by servicing the transport request; wherein the executed instructions cause the on-demand transport system to further perform the selection process based on the estimated revenue.
6. The on-demand transport system of claim 1 , wherein the executed instructions further cause the on-demand transport system to: determine whether the optimal route is within an autonomy grid on which the SDAVs and FAVs operate in an autonomous mode.
7. The on-demand transport system of claim 6 , wherein the autonomy grid comprises ground-truth mapped lanes, corresponding to localization maps stored in the SDAVs and FAVs, for the given region.
8. The on-demand transport system of claim 7 , wherein the executed instructions further cause the on-demand transport system to: exclude the FAVs from the selection process when the optimal route extends beyond the autonomy grid.
9. The on-demand transport system of claim 6 , wherein the executed instructions further cause the on-demand transport system to: based on one or more triggering conditions, remotely shut down or activate respective portions of the autonomy grid for autonomous operation.
10. The on-demand transport system of claim 9 , wherein the one or more triggering conditions correspond to traffic conditions.
11. The on-demand transport system of claim 9 , wherein the one or more triggering conditions correspond to weather conditions.
12. The on-demand transport system of claim 9 , wherein the one or more triggering conditions correspond to at least one of a time of day, an event, or road conditions.
13. The on-demand transport system of claim 6 , wherein the executed instructions further cause the on-demand transport system to: collect vehicle fleet utilization data corresponding to an on-demand transportation service; and establish a set of selection priorities for respective areas of the autonomy grid based on the vehicle fleet utilization data, each selection priority indicating whether a respective area of the autonomy grid is to favor HDVs, SDAVs, or FAVs for servicing transport requests; wherein the executed instructions cause the on-demand transport system to further perform the selection process based on the set of selection priorities.
14. A non-transitory computer readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to:
for a given region, receive transport requests from requesting users in connection with an on-demand transportation service, each transport request indicating a pick-up location and a destination;
for each transport request, (i) determine a risk for an optimal route between the pick-up location and the destination, and (ii) based at least in part on the risk, perform a selection process between a set of safety-driven autonomous vehicles (SDAVs), fully autonomous vehicles (FAVs), and human-driven vehicles (HDVs) to service the transport request;
select a SDAV or a FAV as an optimal vehicle from the candidate set of vehicles to service the transport request; and
in response to selecting the SDAV from the candidate set of vehicles to service the transport request:
select a specified software version from a plurality of software versions stored on the SDAV for execution while servicing the transport request; and
transmit a set of transport instructions to the SDAV indicating the specified software version for execution while servicing the transport request; and
in response to selecting the FAV from the candidate set of vehicles to service the transport request:
select a verified software version stored on the FAV for execution while servicing the transport request based on the risk; and
transmit a set of transport instructions to the FAV indicating the verified software version for execution while servicing the transport request.
15. The non-transitory computer readable medium of claim 14 , wherein the executed instructions further cause the one or more processors to further perform the selection process by determining a set of candidate vehicles within a certain proximity to the pick-up location, the set of candidate vehicles including at least one HDV, at least one SDAV, and at least one FAV.
16. The non-transitory computer readable medium of claim 15 , wherein the executed instructions further cause the one or more processors to: for each vehicle in the set of candidate vehicles, determine an estimated time of arrival (ETA) to the pick-up location indicated in the transport request; wherein the executed instructions further cause the on-demand transport system to perform the selection process based on an estimated time of arrival (ETA) of each of the set of candidate vehicles to the pick-up location.
17. The non-transitory computer readable medium of claim 16 , wherein the executed instructions further cause the one or more processors to perform the selection process based on a desired time range for pick-up.
18. A computer-implemented method of facilitating an on-demand transportation service, the method being performed by one or more processors and comprising:
for a given region, receiving transport requests from requesting users in connection with an on-demand transportation service, each transport request indicating a pick-up location and a destination;
for each transport request, (i) determining a risk for an optimal route between the pick-up location and the destination, and (ii) based at least in part on the risk, performing a selection process between a set of safety-driven autonomous vehicles (SDAVs), fully autonomous vehicles (FAVs), and human-driven vehicles (HDVs) to service the transport request;
selecting a SDAV or a FAV as an optimal vehicle from the candidate set of vehicles to service the transport request; and
in response to selecting the SDAV from the candidate set of vehicles to service the transport request:
selecting a specified software version from a plurality of software versions stored on the SDAV for execution while servicing the transport request; and
transmitting a set of transport instructions to the SDAV indicating the specified software version for execution while servicing the transport request; and
in response to selecting the FAV from the candidate set of vehicles to service the transport request:
selecting a verified software version stored on the FAV for execution while servicing the transport request based on the risk; and
transmitting a set of transport instructions to the FAV indicating the verified software version for execution while servicing the transport request.