Method and system for conditional operation of an autonomous agent
A method for conditional operation of an autonomous agent includes: collecting a set of inputs; processing the set of inputs; determining a set of policies for the agent; evaluating the set of policies; and operating the ego agent. A system for conditional operation of an autonomous agent includes a set of computing subsystems (equivalently referred to herein as a set of computers) and/or processing subsystems (equivalently referred to herein as a set of processors), which function to implement any or all of the processes of the method.
1 . A method for operating an autonomous vehicle, the method comprising:
determining a set of policies for evaluation by the autonomous vehicle, the set of policies comprising:
a set of multiple action policies, wherein each of the set of multiple action policies prescribes:
a set of multiple actions; and
a set of trigger conditions associated with transitions between consecutive actions of the set of multiple actions;
selecting a policy from the set of policies for implementation by the autonomous vehicle, wherein the selected policy is a multiple action policy of the set of multiple action policies; and
operating the autonomous vehicle according to the selected policy, comprising implementing a first action of the set of multiple actions of the selected policy.
2 . The method of claim 1 , wherein the set of policies further comprises a set of single action policies.
3 . The method of claim 1 , wherein selecting the policy comprises calculating a set of metrics associated with the set of policies, wherein the set of metrics is determined based at least in part on a set of predicted movements associated with a set of tracked agents in an environment of the autonomous vehicle.
4 . The method of claim 3 , wherein the set of predicted movements is produced based on performing a set of simulations associated with the set of policies.
5 . The method of claim 1 , wherein operating the autonomous vehicle according to the selected policy further comprises:
checking for satisfaction of a first trigger condition of the set of trigger conditions; and
in an event that the first trigger condition is satisfied, operating the autonomous vehicle according to a second action of the set of multiple actions of the selected policy.
6 . The method of claim 1 , wherein at least a portion of the set of multiple action policies is determined based on a location of the autonomous vehicle.
7 . The method of claim 6 , wherein the portion of the set of multiple action policies is determined based on referencing a labeled map based on the location.
8 . The method of claim 6 , wherein at least a second portion of the set of multiple action policies is determined absent of the location.
9 . The method of claim 1 , wherein at least a portion of the set of multiple action policies is predetermined.
10 . The method of claim 1 , wherein at least a portion of the set of multiple action policies is determined based on a scenario detected by the autonomous vehicle, the scenario comprising at least one of: a crosswalk, an intersection, or a parking lot.
11 . The method of claim 1 , wherein evaluating each of the first set of policies comprises performing a simulation for the policy, wherein evaluating the first set of policies is performed based on a set of metrics produced based on a set of outcomes occurring in the set of simulations.
12 . A method for operating an autonomous vehicle, the method comprising:
determining a first set of policies for evaluation by the autonomous vehicle, the first set of policies comprising a set of multiple action policies, wherein each of the set of multiple action policies prescribes:
a set of multiple actions; and
a set of trigger conditions associated with the set of multiple actions;
evaluating each of the first set of policies to select a first policy from the first set of policies for implementation by the autonomous vehicle, wherein the first policy is a multiple action policy of the set of multiple action policies;
operating the autonomous vehicle according to a first action of the first policy;
while the autonomous vehicle is being operated according to the first action:
determining a second set of policies for evaluation by the autonomous vehicle;
evaluating the second set of policies to produce a second set of metrics;
based on evaluating the second set of policies, selecting a second policy, the second policy separate and distinct from a second action of the first policy.
13 . The method of claim 12 , wherein the first set of policies for evaluation further comprises a set of single action policies.
14 . The method of claim 12 , wherein the set of trigger conditions is further associated with transitions between consecutive actions of the set of multiple actions.
15 . The method of claim 11 , wherein performing the simulation for each of the set of multiple action policies comprises monitoring for a first trigger condition of the set of triggering conditions in the multiple action policy being simulated.
16 . The method of claim 15 , wherein performing the simulation further comprises transitioning to simulating a second action of the multiple action policy in response to the first trigger condition being satisfied.
17 . The method of claim 11 , wherein the second set of policies is evaluated according to a predetermined election cycle period.
18 . The method of claim 17 , wherein the predetermined election cycle period is shorter in duration than a duration of time simulated in each of the set of simulations.
19 . The method of claim 11 , wherein performing a simulation of the set of simulations for the first policy comprises implementing a trigger condition in response simulating the autonomous vehicle following a right-of-way driving convention relative to a set of tracked agents in the simulation.
20 . The method of claim 12 , wherein operating the autonomous vehicle according to the first action comprises monitoring for a first trigger condition of the set of trigger conditions.