Nissan North America, Inc.

University of Massachusetts

Renault S.a.s.

United States of America As Represented by the Administrator of Nasa

Kyle Hollins Wray

Stefan Witwicki

Shlomo Zilberstein

Liam Pedersen

Christopher Ostafew

Richard Lui

David Ilstrup

Kuniaki Noda

Omar Bentahar

Arec Jamgochian

Atsuhide Kobashi

Connor Basich

Julius S Gyorfi

Melissa Cefkin

Najamuddin Mirza Baig

Erik Vinkhuyzen

Laura Cesafsky

Astha Vagadia

Sparsh Sharma

Luis Lorenzo Bill

Luis Lorenzo Bill Clark

Justin Svegliato

Decision-making for a vehicle uses a data determining interface between perception and decision-making. The interface receives, from at least two data sources, operational environment data representing objects external to the vehicle while the vehicle is traversing a vehicle transportation network. The operational environment data is modified to determine output data for data types needed to determine a candidate vehicle control action responsive to the distinct vehicle operation scenario identified using the operational environment data. The solution is the same candidate vehicle control action for the same conditions regardless of the data types available from the data sources. That is, even where different data sources are available, consistent output is provided to a decision-making system. Consistent decision-making results whether the vehicle has a high-definition map, a standard-definition map, or no map as a data source, for example.

Data determining interface for vehicle decision-making

Trajectory planning include identifying state data based on sensor data from sensors of a vehicle. The state data are input to a machine-learning model to obtain parameters of a trajectory planner. The parameters are input to the trajectory planner to obtain a short term speed plan. The vehicle is autonomously controlled according to at least a portion of the short term speed plan.

Imitation learning for trajectory planning

Engine activation planning in a hybrid electric vehicle (HEV) is disclosed. Historical driving data of the HEV are analyzed to identify recurring driving scenarios and patterns specific to a driver of the HEV. Future driving scenarios are predicted based on the historical driving data. An engine activation policy is generated for the HEV. The engine activation policy optimizes battery charging and usage in response to the future driving scenarios. The engine activation policy is used to control activation of a gasoline engine in the HEV where a control decision is dynamically adjusted based on a comparison of real-time driving conditions with the future driving scenarios.

Navigation map learning for intelligent hybrid-electric vehicle planning

A method for planning an activation action for an engine of a vehicle is disclosed. The method includes planning, according to a model, an activation action of an engine of a vehicle, and activating the engine according to the activation action. The model includes a state space comprising a current charge level of the battery and whether the engine is currently on or off. The activation action is selected from a set comprising a first action to turn on the engine to charge the battery and a second action to turn off the engine.

Intelligent engine activation planner

A first method includes detecting, based on sensor data, an environment state; selecting an action based on the environment state; determining an autonomy level associated with the environment state and the action; and performing the action according to the autonomy level. The autonomy level can be selected based at least on an autonomy model and a feedback model. A second method includes calculating, by solving an extended Stochastic Shortest Path (SSP) problem, a policy for solving a task. The policy can map environment states and autonomy levels to actions and autonomy levels. Calculating the policy can include generating plans that operate across multiple levels of autonomy.

Online state space refinement for competence-aware systems

An occlusion is identified in a vehicle transportation network. A visibility grid is identified on a second side of the occlusion for a vehicle that is on a first side of the occlusion. The visibility grid is identified with respect to a region of interest that is at least a predefined distance above ground. The visibility grid is used to identify first portions of roads sensed by a sensor positioned on the vehicle and second portions of the roads that are not sensed by the sensor. A driving behavior of the vehicle is altered based on the visibility grid.

3D occlusion reasoning for accident avoidance

Vehicle guidance with systemic optimization may include traversing, by a current vehicle, a vehicle transportation network, by obtaining, by the current vehicle, systemic-utility vehicle guidance data for a current portion of the vehicle transportation network and traversing, by the current vehicle, the current portion of the vehicle transportation network in accordance with the systemic-utility vehicle guidance data. Obtaining the systemic-utility vehicle guidance data may include obtaining vehicle operational data for a region of a vehicle transportation network, wherein the vehicle operational data includes current operational data for a plurality of vehicles operating in the region, operating a systemic-utility vehicle guidance model for the region, obtaining systemic-utility vehicle guidance data for the region from the systemic-utility vehicle guidance model in response to the vehicle operational data, and outputting the systemic-utility vehicle guidance data to the current vehicle.

Vehicle guidance with systemic optimization

A first distinct vehicle operational scenario is identified for an autonomous vehicle (AV). A first set of candidate vehicle control actions are received from a model that provides a first solution to the first distinct vehicle operational scenario. An action is selected from the first set of candidate vehicle control actions. The AV is controlled based on the action. The first solution is obtained offline in a first idealized situation that is decoupled from a current context of the AV.

Risk-aware executor with action set recommendations

Lane-level route planning includes obtaining lane-level information of a road, where the road includes a first lane and a second lane and the lane-level information includes first lane information related to the first lane and second lane information related to the second lane; converting the lane-level information to probabilities for a state transition function; receiving a destination; and obtaining a policy as a solution to a model that uses the state transition function.

Learning in lane-level route planner

Providing explanations in route planning includes determining a route based on at least two objectives received from a user, where a second objective of the at least two objectives is constrained to within a slack value of a first objective of the at least two objectives; receiving, from the user, a request for an explanation as to an action along the route; and providing the explanation to the user. The explanation describes an extent of violating the slack value.

Explainability and interface design for lane-level route planner

Activating an engine of a HEV to charge a battery includes obtaining a first GPS trace from a first trip of the HEV along a first route, where the first GPS trace includes first trace metadata; obtaining a second GPS trace from a second trip, where the second GPS trace includes second trace metadata; adding, to a navigation map, an aggregation of the first trace metadata and the second trace metadata for edges of the navigation map; using the navigation map to obtain an activation action of the engine, where the activation action is selected from a set that includes a first activation action of turning the engine on and a second activation action of turning the engine off; and activating the engine according to the activation action, where activating the engine using the first activation action causes the engine to turn on to charge the battery of the HEV.

Real-time decision-making for a vehicle using belief state determination is described. Operational environment data is received while the vehicle is traversing a vehicle transportation network, where the data includes data associated with an external object. An operational environment monitor establishes an observation that relates the object to a distinct vehicle operation scenario. A belief state model of the monitor computes a belief state for the observation directly from the operational environment data. The monitor provides the computed belief state to a decision component implementing a policy that maps a respective belief state for the object within the distinct vehicle operation scenario to a respective candidate vehicle control action. A candidate vehicle control action is received from the policy of the decision component, and a vehicle control action is selected for traversing the vehicle transportation from any available candidate vehicle control actions.

Belief state determination for real-time decision-making

An autonomous vehicle traverses a vehicle transportation network using a multi-objective policy based on a model for specific scenarios. The multi-objective policy includes a topographical map that shows a relationship between at least two objectives. The autonomous vehicle receives a candidate vehicle control action associated with each of the at least two objectives. The autonomous vehicle selects a vehicle control action based on a buffer value that is associated with the at least two objectives. The autonomous vehicle traverses a portion of the vehicle transportation network in accordance with the selected vehicle control action.

Objective-based reasoning in autonomous vehicle decision-making

Traversing, by an autonomous vehicle, a vehicle transportation network, may include identifying a distinct vehicle operational scenario, wherein traversing the vehicle transportation network includes traversing a portion of the vehicle transportation network that includes the distinct vehicle operational scenario, communicating shared scenario-specific operational control management data associated with the distinct vehicle operational scenario with an external shared scenario-specific operational control management system, operating a scenario-specific operational control evaluation module instance including an instance of a scenario-specific operational control evaluation model of the distinct vehicle operational scenario, and wherein operating the scenario-specific operational control evaluation module instance includes identifying a policy for the scenario-specific operational control evaluation model, receiving a candidate vehicle control action from the policy for the scenario-specific operational control evaluation model, and traversing a portion of the vehicle transportation network based on the candidate vehicle control action.

Shared autonomous vehicle operational management

An apparatus for post-processing of a decision-making model of an autonomous vehicle receives a decision-making model including a plurality of states. The model is processed using multivariate data that comprises values for at least three observations of a vehicle operational scenario. A slice of the model decision space is generated by fixing values of all except two observations, and modifying the values of the two observations to obtain multiple alternative solutions for the model. The alternative solutions and the modified values form the slice. Each alternative solution is associated with a respective first value of a first observation and a respective second value of a second observation. The apparatus also generates a solution to a modified decision-making model that is the model modified by, for at least one state and at least one of the two observations, modifying a probabilistic transition matrix, a probabilistic observation matrix, or both.

Growing community of inventors

Fremont, CA, United States of America

Kyle Hollins Wray