Method and system for dynamically curating autonomous vehicle policies
A system for dynamic policy curation includes a computing system and interfaces with an autonomous agent. A method for dynamic policy curation includes collecting a set of inputs; processing the set of inputs; and determining a set of available policies based on processing the set of inputs. Additionally or alternatively, the method can include any or all of: selecting a policy; implementing a policy; and/or any other suitable processes.
1. A method for dynamically and continuously refining policies available for selection by a vehicle, the method comprising:
collecting a set of sensor inputs;
processing the set of sensor inputs to determine a position of the vehicle;
determining a first set of policies for the vehicle based on the position;
refining the first set of policies based on supplementary information to determine a second set of policies;
selecting a policy from the second set of policies, wherein selecting the policy comprises calculating a score associated with each of the second set of policies to produce a set of scores and selecting the policy based on the set of scores; and
controlling a movement of the vehicle according to the selected policy.
2. The method of claim 1 , wherein refining the first set of policies based on supplementary information comprises at least one of: eliminating a subset of the first set of policies from consideration in selecting the policy and adding a third set of policies to the first set of policies for consideration in selecting the policy.
3. The method of claim 1 , wherein the first set of policies is further determined based on referencing a map based on the position.
4. The method of claim 3 , wherein the map comprises a set of policy assignments for each of a set of multiple positions in the map.
5. The method of claim 1 , wherein the supplementary information comprises at least one of:
an input from a remote operator;
a set of parameters associated with the set of scenarios; and
a classified scenario associated with an environment of the vehicle.
6. The method of claim 1 , further comprising performing a set of simulations, wherein the set of scores is produced based on a set of outcomes of the set of simulations.
7. The method of claim 6 , wherein the set of simulations comprises a simulation for each of the second set of policies, wherein in each simulation, a predicted effect of the vehicle implementing a particular policy of the second set of policies is determined, wherein the associated score is determined based on this predicted effect.
8. The method of claim 7 , wherein each of the set of simulations further comprises simulating a motion of a set of monitored vehicles in an environment of the vehicle, wherein the predicted effect is determined, at least in part, based on the motion of the set of monitored objects.
9. The method of claim 1 , wherein refining the first set of policies comprises determining a set of scaling factors for use in adjusting the set of scores.
10. A system for dynamically and continuously refining policies available for selection by a vehicle, the system comprising:
a set of sensors;
a computer in communication with the set of sensors, wherein the computer:
receives sensor data from the set of sensors;
processes the sensor data to determine a position of the vehicle;
determines a first set of policies for the vehicle based on the position;
refines the first set of policies based on supplementary information to determine a second set of policies;
calculates a score associated with each of the second set of policies to produce a set of scores; and
selects a policy from the second set of policies, wherein the policy is selected based on the set of scores;
a controller in communication with the computer, wherein the controller operates the vehicle according to the selected policy.
11. The system of claim 10 , wherein refining the first set of policies based on supplementary information comprises at least one of: eliminating a subset of the first set of policies from consideration in selecting the policy and adding a third set of policies to the first set of policies for consideration in selecting the policy.
12. The system of claim 10 , further comprising a map, wherein the first set of policies is further determined based on referencing the map based on the position.
13. The system of claim 12 , wherein the map comprises a set of policy assignments for each of a set of multiple positions in the map.
14. The system of claim 10 , wherein the supplementary information comprises at least one of:
an input from a remote operator;
a set of parameters associated with the set of scenarios; and
a classified scenario associated with an environment of the vehicle.
15. The system of claim 10 , wherein the computer further performs a set of simulations, wherein the set of scores is produced based on a set of outcomes of the set of simulations.
16. The system of claim 15 , wherein the set of simulations comprises a simulation for each of the second set of policies, wherein in each simulation, the computer determines a predicted effect of the vehicle implementing a particular policy of the second set of policies, wherein the associated score is determined based on this predicted effect.
17. The system of claim 16 , wherein, in each of the set of simulations, the computer further simulates a motion of a set of monitored vehicles in an environment of the vehicle, wherein the predicted effect is determined, at least in part, based on the motion of the set of monitored objects.
18. The system of claim 10 , wherein refining the first set of policies comprises determining a set of scaling factors for use in adjusting the set of scores.