Other

Kah Seng Tay

James Patrick Marion

Qing Sun

Brody Huval

Joel Pazhayampallil

Vishisht Gupta

Tim Cheng

Xiaotian Chen

Xin Sun

One variation of a method includes: accessing a 2D color image recorded by a 2D color camera and a 3D point cloud recorded by a 3D depth sensor at approximately a first time, the 2D color camera and the 3D depth sensor defining intersecting fields of view and facing outwardly from an autonomous vehicle; detecting a cluster of points in the 3D point cloud representing a continuous surface approximating a plane; isolating a cluster of color pixels in the 2D color image depicting the continuous surface; projecting the cluster of color pixels onto the plane to define a set of synthetic 3D color points in the 3D point cloud, the cluster of points and the set of synthetic 3D color points representing the continuous surface; and rendering points in the 3D point cloud and the set of synthetic 3D color points on a display.

Method for rendering 2D and 3D data within a 3D virtual environment

One variation of a method for updating a localization map for a fleet of autonomous vehicles includes: selecting a set of roads including a first subset of road segments associated with existing incomplete scan data and a second subset of road segments associated with a monitoring request from an external entity; during a passenger period at the autonomous vehicle, autonomously transporting passengers according to a series of ride requests; during a mapping period succeeding the passenger period at the autonomous vehicle, autonomously navigating along the set of road segments, recording a first series of scan data representing surfaces proximal the first subset of road segments, and recording a second series of scan data representing surfaces proximal the second subset of road segments; updating the localization map based on the first series of scan data; and serving the second series of scan data to the external entity.

Method for updating a localization map for a fleet of autonomous vehicles

One variation of a method for autonomous navigation includes, at an autonomous vehicle: recording a first image via a first sensor and a second image via a second sensor during a scan cycle; calculating a first field of view of the first sensor and a second field of view of the second sensor during the scan cycle based on surfaces represented in the first and second images; characterizing a spatial redundancy between the first sensor and the second sensor based on an overlap of the first and second fields of view; in response to the spatial redundancy remaining below a threshold redundancy, disabling execution of a first navigational action&#x2014;action informed by presence of external objects within a first region of a scene around the autonomous vehicle spanning the overlap&#x2014;by the autonomous vehicle; and autonomously executing navigational actions, excluding the first navigational action, following the scan cycle.

Method for autonomous navigation

One variation of a method for dynamically querying a remote operator for assistance includes, at an autonomous vehicle: autonomously navigating along a roadway; at locations along the roadway, testing performance of a set of wireless networks; in response to degradation of the set of wireless networks, decreasing a sensitivity threshold for triggering remote operator events at the autonomous vehicle; in response to a condition at the autonomous vehicle exceeding the sensitivity threshold at a particular location along the roadway, triggering a remote operator event; during the remote operator event, transmitting sensor data to a remote computer system via a subset of wireless networks, in the set of wireless networks, based on performance of the set of wireless networks proximal the particular location; and executing a navigational command received from a remote operator via a wireless network in the set of wireless networks.

Growing community of inventors

Mountain View, CA, United States of America

Kah Seng Tay