Nissan North America, Inc.

Renault S.a.s.

Nissan Motor Company Limited

University of Massachusetts

Christopher Ostafew

Kyle Hollins Wray

Liam Pedersen

Atsuhide Kobashi

Qizhan Tam

Stefan Witwicki

David Ilstrup

Richard Lui

Therese Cypher-Plissart

Sachin Hagaribommanahalli

Yue Zhao

Ali Mortazavi

Kuniaki Noda

Shlomo Zilberstein

Siddharth Thakur

Najamuddin Mirza Baig

Armelle Guerin

Viju James

Astha Vagadia

Thiago Aguiar

Manh Huynh

Huiching Chen

Observed driveline mean and variance data are used for determining the variance of a trajectory of tracked objects for use by a host vehicle. A map of a portion of a vehicle transportation network are determined, wherein the map is comprised of observed driveline mean and variance data for one or more map points. At least one trajectory of a tracked object is predicted, wherein a trajectory includes a series of location each corresponding to a respective predicted position of the tracked object at a future time. A map-based variance is generated for the location of the trajectory using a smoothed curvature of the trajectory within the map. A control system of the vehicle operates the vehicle using the map-based variance as input.

Prediction variance estimation

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 proactive pedal algorithm is used to modify an accelerator pedal map to ensure the deceleration when the accelerator pedal is released matches driver expectation. Modifying the accelerator pedal map provides the driver of a vehicle the sensation that the vehicle resists moving when travelling in dense scenes with potentially high deceleration requirements and coasts easily in scenes with low deceleration requirements. The accelerator pedal map is modified based on a scene determination to classify other remote vehicles as in-lane, neighbor-lane, or on-coming.

System and method for proactive lane assist

A vehicle lane marking detection system includes a 3D sensor, a driver assist component and an electronic controller. The 3D sensor is installed to a vehicle and is configured to scan physical objects around the vehicle outputting a plurality of data points each corresponding to a surface point of a physical feature. Each data point being defined by distance, direction, intensity and vertical location relative to the vehicle. The electronic controller is connected to the 3D sensor and the driver assist component. The electronic controller evaluates a point cloud defined by the data points identifying lane markings based on the intensity of the data points. The data points having intensities greater than a predetermined level are determined to correspond to lane marking and are provided to the driver assist component with the lane markings for use thereby.

Vehicle lane marking detection system

A vehicle includes a vehicle engine, a steering control unit, an on-board sensor network and a virtual hazard inference system. The vehicle engine generates a torque output of the vehicle. The steering control unit controls a steering angle of the vehicle. The on-board sensor network is programmed to detect external objects within a detection zone. The virtual hazard inference system has a processor programmed to infer a presence of an occluded object that is not detected by the on-board sensor network. The processor controls the vehicle engine and the steering control unit to perform at least one of a torque control operation and a steering control operation.

Virtual hazard inference system

Categorizing driving behaviors of other road users includes maintaining a first history of first lateral-offset values of a road user with respect to a center line of a lane of a road; determining a first pattern based on the first history of the first lateral-offset values; determining a driving behavior of the road user based on the first pattern; and autonomously performing, by a host vehicle, a driving maneuver based on the driving behavior of the road user. The first history can be maintained for a predetermined period of time. An apparatus includes a processor that is configured to track a trajectory history of a road user; determine, based on the trajectory history, a driving behavior of the road user; and transmit a notification of the driving behavior.

Road user categorization through monitoring

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

Detecting prediction errors includes detecting a road user; determining respective predicted data for the road user; storing, in a data structure, the respective predicted data; storing, in the data structure, actual data of the road user; obtaining an average prediction displacement error using at least one of the actual data and at least two corresponding respective predicted data; and determining a prediction accuracy based on the average prediction displacement error. Detecting map errors includes detecting a road user; storing, in a data structure, actual data of the road user; storing, in the data structure, map data corresponding to the actual data; obtaining an average map displacement error based on a comparison of at least some of the actual data and corresponding at least some map data; and determining a map accuracy based on the average map displacement error.

Autonomous vehicle control and map accuracy determination based on predicted and actual trajectories of surrounding objects and vehicles

Proactively mitigating risk to a vehicle traversing a vehicle transportation network is described. First and second hazard zones for first and second objects ahead of the vehicle are respectively determined. The first hazard zone includes a first target lateral constraint that extends over a left lane boundary, and the second hazard zone includes a second target lateral constraint that extends over a right lane boundary. The lateral constraints separately allow the vehicle to avoid the objects without a speed constraint. Where the first and second hazard zones overlap in the longitudinal direction, a lateral buffer is allocated between the lateral constraints to generate first and second allocated lateral constraints. Longitudinal constraints are respectively determined based on times of arrival at each hazard zone. Using the constraints, a proactive trajectory is determined that includes a lateral contingency, a longitudinal contingency, or both. The vehicle is controlled according to the trajectory.

Proactive risk mitigation

Object avoidance by an autonomous vehicle (AV) is disclosed. A method includes detecting a first object along a coarse driveline of a drivable area of the AV; receiving a predicted path of the first object; determining, based on the predicted path of the first object, an adjusted drivable area; and determining a trajectory of the AV through the adjusted drivable area. A system includes a trajectory planner configured to detect a first object along a coarse driveline of a drivable area of the AV; receive a predicted path of the first object; determine, based on the predicted path of the first object, an adjusted drivable area; and determine a trajectory of the AV through the adjusted drivable area.

Trajectory planning

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.

A vehicle data display system includes an electronic display installed within a vehicle, a 3-D sensor and an electronic controller. The 3-D sensor configured to scan areas forward of and along lateral sides of the vehicle producing point cloud. Each data point of the cloud data corresponding to a surface portion of a physical feature. Each data point includes distance, direction and vertical location of each surface point. The electronic controller is connected to the electronic display and the 3-D sensor. The electronic controller receives and evaluates the point cloud from the 3-D sensor generating a 3-D model of detected ones of the physical features around the vehicle including ground surfaces, non-drivable features and driving limiting features relative to the vehicle. The non-drivable features are features that have predetermined geometric relationships with adjacent ground surfaces such that caution is to be taken when driving over or on driving limiting features.

Vehicle data display system

A reactive pedal algorithm is used to modify an accelerator pedal output (APO)-to-torque conversion to produce more deceleration for the same accelerator pedal position. Modifying the APO-to-torque conversion provides the driver of a vehicle the sensation that the vehicle is resisting approaching closer to a lead vehicle. The APO-to-torque conversion is modified based on a scene determination to classify vehicles as in-lane, neighbor-lane, or on-coming. Lane change assist methods and systems are used to modify the APO-to-torque conversion range based on a lead vehicle, a neighbor vehicle, or both.

Growing community of inventors

Mountain View, CA, United States of America

Christopher Ostafew