Zoox, Inc.

Zheyuan Xie

Steven Cheng Qian

Joseph Funke

Genie Kim

A prediction component of an autonomous vehicle may detect and use a combination of object attributes and semantic features to determine potential destinations and corresponding potential trajectories for the dynamic objects in an environment. In some examples, the prediction component may analyze sensor data using trained models to determine a number of object attributes for a dynamic object, as well as semantic features associated with the intent or purpose of the dynamic object. Using a combination of the object attributes and the semantic features, for the dynamic object and any number of potential destination(s), the prediction component may predict a likely potential destination and determine a predicted trajectory for the dynamic object. Based on predicted trajectory of the dynamic object, the autonomous vehicle may select and execute driving maneuvers to control the autonomous vehicle safely and efficiently within the environment.

Object destination and trajectory prediction based on semantic features

Techniques for determining parking spaces and/or parking trajectories for a vehicle based on multistage filtering are discussed herein. In some examples, a vehicle may navigate to a destination within an environment. The vehicle may receive a set of parking locations and perform filtering operations to determine a subset of the parking locations. For example, the vehicle may use a list of parking characteristics (e.g., conditions) to filter the set of parking spaces. Based on determining the subset of parking spaces, the vehicle may generate a set of candidate trajectories to the subset of parking spaces. The vehicle may determine a subset of the candidate trajectories based on filtering the set of candidate trajectories according to various constraints. Further, the vehicle may determine a cost for the subset of candidate trajectories. Based on the costs, the vehicle may determine a candidate trajectory for the vehicle to follow.

Parking location filter

Techniques for determining a parking trajectory for a vehicle are discussed herein. A parking management component may determine or receive a three-dimensional grid ('grid'), discretized based at least in part on a heading offset, a lateral offset, and/or a longitudinal offset between a first and second pose. A cell of the grid may include a cost associated indicating a minimum difference to the second pose when moving from the first as may be limited based on kinematic and/or dynamic constraints. When driving, a vehicle may determine a relative state of the vehicle to a desired location, use the relative state to access a cost from the grid, and determine whether to follow a trajectory based on the cost.

Trajectory determination based on heuristic grid data

A passenger may be rather vulnerable to safety risks during pickup and/or drop-off of a passenger by a vehicle. To mitigate or eliminate such risk, the vehicle may determine an endpoint for a vehicle route to pickup or drop-off a passenger at a location. The vehicle may determine an estimated path between the endpoint and the location and may determine a safety confidence score by a machine-learned model for the estimated path and/or may predict a trajectory of a detected object to ascertain whether the estimated path is safe. The vehicle may execute any of a number of different mitigation actions to reduce or eliminate a safety risk if one is detected.

Autonomous vehicle operations related to detection of an unsafe passenger pickup/delivery condition

Determining parking locations based on multistage filtering

Techniques for determining a parking trajectory for an autonomous vehicle. In some examples, a vehicle may navigate to a destination within an environment. The vehicle may receive parking locations proximate the destination, and determine a cost for navigating to each parking location, or multiple locations within a parking space. Each of multiple trajectories may include desired state(s) that represent the desired state information of the vehicle at a specific locations along the trajectory. The vehicle may determine a cost for the trajectory by comparing the predicted pose of the vehicle at the parking state to the target pose of the parking location. The vehicle may follow the trajectory based on the cost.

Trajectory determination based on pose data

Techniques for controlling a vehicle to select parking spaces are discussed herein. The vehicle may receive a request to travel to a destination and obtain information about a first set and a second set of parking spaces within a threshold distance of the destination. The vehicle may determine a first likelihood of availability associated with the first set of parking spaces and a second likelihood of availability associated with the second set of parking spaces. The vehicle may determine a first cost associated with a first route to the first set of parking spaces and a second cost associated with a second route to the second set of parking spaces by the cost function. Upon determining that the first cost is equal to or less than the second cost, the vehicle may navigate based on the first route and park within one of the first set of parking spaces.

Fleet control for parking space selection

Techniques and methods for identifying parking zones. For instance, a vehicle may identify a parking zone located near a destination location for the vehicle. The vehicle may then generate one or more lines representing the parking zone. Additionally, the vehicle may generate polygons representing objects located proximate to the parking zone. The vehicle may then determine whether the one or more lines intersect with one or more of the polygons. If the vehicle determines that the one or more lines intersect with one or more of the polygons, then the vehicle may identify one or more first portions of the parking zone that are occupied by one or more objects and as such, unavailable. Using the one or more first portions, the vehicle may identify one or more second portions of the parking zone that are not occupied by objects and as such, available.

Growing community of inventors

Foster City, CA, United States of America

Zheyuan Xie