Zoox, Inc.

Joseph Funke

Steven Cheng Qian

Johannes Edren

David Evan Zlotnik

Jacob Patrick Thalman

Marin Kobilarov

Ali Javidan

Kazuhide Okamoto

Zheyuan Xie

Sriram Narayanan

Jacob Lee Askeland

Zachary Stuart Churukian

Genie Kim

Timothy David Kentley-Klay

Chuang Wang

Moritz Boecker

Ashutosh Gajanan Rege

Gary Linscott

Timothy Caldwell

Collin MacGregor

Gowtham Garimella

Marc Wimmershoff

Robert Edward Somers

Mark Jonathon McClelland

Joona Markus Petteri Kiiski

Sy Kelly Olson

Olivier Toupet

Vasumathi Raman

Shayan Darayan

Liam Gallagher

Vincent Andreas Laurense

Joshua Dean Egbert

Amirhossein Tamjidi

Jesse Sol Levinson

Ravi Verma Gogna

Robert J Chen

Matthew Austin Terrel

Jefferson Bradfield Packer

Andreas Christian Reschka

David Martins Belo

Jeremy Schwartz

Jack Riley

Rasoul Kabirzadeh

Brian Michael Filarsky

Yunpeng Pan

Xianan Huang

Dhanushka Nirmevan Kularatne

Caili Li

Ruikun Yu

Greg Zanghi

Nivedita Chandrasekaran

Techniques are discussed for generating and optimizing trajectories for controlling autonomous vehicles in performing on-route and off-route actions within a driving environment. A planning component of an autonomous vehicle can receive or generate time-discretized (or temporal) trajectories for the autonomous vehicle to traverse an environment. Trajectories can be optimized, for example, based on the lateral and longitudinal dynamics of the vehicle, using loss functions and/or costs. In some examples, the temporal optimization of a trajectory may include resampling a previous trajectory based on the differences in the time sequences of the temporal trajectories, to ensure temporal consistency of trajectories across planning cycles. Constraints also may be applied during temporal optimization in some examples, to control or restrict driving maneuvers that are not supported by the autonomous vehicle.

Autonomous vehicle trajectory generation and optimization

Techniques are described herein for determining simulated vehicle positional errors and correlating the positional errors with vehicle features. Such techniques may include receiving log data comprising trajectories and position data for a real vehicle, and executing a log-based simulated vehicle based on the log data. A simulated vehicle may be controlled to follow a simulation trajectory in a simulated environment based on the trajectory of the real vehicle. A simulation system may determine a difference between the positions of the simulated vehicle and corresponding positions of the real vehicle in the real environment. The techniques may further include determining a vehicle state features correlated to the lateral and/or longitudinal position errors of the simulated vehicle, and determining, based on the vehicle state features, position error distributions and/or models that can be used to control subsequent driving simulations.

Simulated vehicle position errors in driving simulations

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

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

An over actuated system capable of controlling wheel parameters, such as speed (e.g., by torque and braking), steering angles, caster angles, camber angles, and toe angles, of wheels in an associated vehicle. The system may determine the associated vehicle is in a rollover state and adjust wheel parameters to prevent vehicle rollover. Additionally, the system may determine a driving state and dynamically adjust wheel parameters to optimize driving, including, for example, cornering and parking. Such a system may also dynamically detect wheel misalignment and provide alignment and/or corrective driving solutions. Further, by utilizing degenerate solutions for driving, the system may also estimate tire-surface parameterization data for various road surfaces and make such estimates available for other vehicles via a network.

Independent control of vehicle wheels

Systems and techniques for determining a sideslip vector for a vehicle that may have a direction that is different from that of a heading vector for the vehicle. The sideslip vector in a current vehicle state and sideslip vectors in predicted vehicles states may be used to determine paths for a vehicle through an environment and trajectories for controlling the vehicle through the environment. The sideslip vector may be based on a vehicle position that is the center point of the wheelbase of the vehicle and may include lateral velocity, facilitating the control of four-wheel steered vehicle while maintaining the ability to control two-wheel steered vehicles.

Determining vehicle position using sideslip vector

This disclosure describes techniques for autonomous vehicles to determine driving paths and associated trajectories through unstructured or off-route driving environments. When an object is detected within a driving environment, a vehicle may determine a cost-based side association for the object. Various costs may be used in different examples, including costs based on a cost plot and/or motion primitives that may vary for terminal and non-terminal desired destinations (or ending vehicle states). Using tree searches to determine estimated candidate costs, the autonomous vehicle may compare right-side and left-side driving path costs around the object to determine a side association for the object. Based on the side association, the autonomous vehicle may determine an updated planning corridor and/or a trajectory to control the vehicle from a current state to a desired ending vehicle state.

Autonomous driving object detection and avoidance

Four-wheel steering of a vehicle, e.g., in which leading wheels and trailing wheels are steered independently of each other, can provide improved maneuverability and stability. A first vehicle model may be used to determine trajectories for execution by a vehicle equipped with four-wheel steering. A second vehicle model may be used to control the vehicle relative to the determined trajectories. For instance, the second vehicle model can determine leading wheels steering angles for steering leading wheels of the vehicle and trailing wheels steering angles for steering trailing wheels of the vehicle, independently of the leading wheels.

Trajectory determination for four-wheel steering

Techniques are described for determining to modify a coordinate system in response to the occurrence of a condition. As non-limiting examples, such conditions comprise distance traveled, speed of the vehicle (or system), an amount of computational resources available, a time since the last change, a number of objects present, the presence (or absence) of a particular object proximate the vehicle, a distance to a proximate object, based on a particular frequency, or the like. Such modifications may improve the operation and safety of a computing system used for path planning and trajectory generation by allowing lower precision numerical representations to be used in safety-critical situations while avoiding potential computational errors that would otherwise result from using such a representation.

Adaptable origin for location offsets

Techniques for accurately predicting vehicle state errors for avoiding collisions with objects detected in an environment of a vehicle are discussed herein. A vehicle safety system can implement a model determine a representation of the vehicle usable in a scenario. The model may dynamically determine a size and/or a heading of the vehicle representation based on differences between a candidate trajectory and a current trajectory of the vehicle. The safety system can identify a potential collision between the vehicle and the object based at least in part on an overlap of the vehicle representation and an object representation.

Vehicle representation determination

Four-wheel steering of a vehicle, e.g., in which leading wheels and trailing wheels are steered independently of each other, can provide improved maneuverability and stability. Steering angle constraints may be used to limit or prevent saturation of steering by only one set of wheels as well as to reduce sideslip. The steering angle constraints are vehicle independent and applicable across a fleet of different vehicles based on vehicle speed. For instance, the steering angle constraints establish angle limits that dynamically adjust based on vehicle speed to ensure vehicle velocity and acceleration remain within predefined limits as established by autonomous vehicle system design.

Four wheel steering angle constraints

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

A vehicle can include a primary computing device and a secondary computing device. The primary computing device can receive a trajectory and can generate control data to control the vehicle based on a computed state. Further, the primary computing device can send the internal data to the secondary computing device configured to control the vehicle in the event of a failure of the primary computing device. The secondary computing device can receive the internal data as first internal data and determine a capability associated with the primary computing device. Using the first internal data, the secondary computing device can determine second internal data and, based on the capability (e.g., in event of a failure of the primary computing device), can control the vehicle to follow a trajectory using the second internal data. Transferring state between an active to a standby computing device can ensure algorithmic synchronization and safe operation.

Multi-controller synchronization

Growing community of inventors

Redwood City, CA, United States of America

Joseph Funke