Zoox, Inc.

Toyota Research Institute, Inc.

Marcus Hammond

Timothy David Kentley-Klay

Marc Wimmershoff

Ryan Marcotte

Randall J St Romain, Ii

Tyler William Packard

A method of controlling availability of autonomy functions of a vehicle includes determining one or more uncertainty levels, each corresponding to a driving data stream including driving data, wherein the uncertainty levels correspond to a probability of a collision or driving off the roadway. The method further includes generating a total uncertainty level based on the one or more uncertainty levels and disabling a first autonomy function based on availability criteria of the first autonomy function. The availability criteria define a first threshold of the total uncertainty level. The method further includes providing a second autonomy function based on the second autonomy function's availability criteria that include a total uncertainty threshold higher than the first threshold.

Layered architecture for availability of advanced driver assistance features

A method for identifying high-risk driving situations in driving data may include receiving driving data, identifying trigger events in the driving data, and using the trigger events in combination with first portions of the driving data corresponding to periods of time preceding the trigger events, training a machine learning model to identify features, in the driving data, associated with the trigger events. The method may further include using second portions of the driving data where no trigger events are identified, further training the machine learning model to ignore features common to both the first portions and the second portions. The method further includes detecting a high risk driving situation in real time driving data by using the machine learning model to identify the features associated with the trigger events, and performing a defensive driving maneuver in response to the detection of the high risk driving situation.

Systems and methods for identifying high-risk driving situations from driving data

Techniques for performing multiple simultaneous pose generation for an autonomous vehicle. For instance, a system that navigates the autonomous vehicle can include at least a first component that determines first poses for the autonomous vehicle using at least a first portion of sensor data captured by one or more sensors and a second component that determines second poses for the autonomous vehicle using at least a second portion of the sensor data. The first component may have more computational resources than the second component and determine poses at a different frequency than the second component. The system may generate trajectories for the autonomous vehicle using the first poses when the first component is operating correctly. Additionally, the system may generate trajectories for the autonomous vehicle using the second poses when the first component is not operating correctly.

Redundant pose generation system

A method for controlling a vehicle based on a prediction from a semantic map is presented. The method includes receiving a snapshot of an environment from one or more sensors. The method also includes generating the semantic map based on the snapshot and predicting an action of a dynamic object in the snapshot based on one or more surrounding objects. The method still further includes controlling an action of the vehicle based on the predicted action.

Online agent predictions using semantic maps

Remote controlling of a vehicle, such as an autonomous vehicle, may sometimes be more efficient and/or reliable. Such control, however, may require processes for ensuring safety of surrounding persons and objects. Aspects of this disclosure include using onboard sensors to detect objects in an environment and alter remote commands according to such objects, e.g. by reducing a maximum permitted velocity of the vehicle as a function of distance to detected objects. In some examples described herein, such remote controlling may be performed by using objects in the environment as control objects, with movements of the control objects resulting in movement of the vehicle.

Growing community of inventors

Mountain View, CA, United States of America

Marcus Hammond