Uber Technologies, Inc.

Aditya Sarawgi

Brent Evans Hamby

Matthew Sweeney

Sophia Cui

Jennie Nguyen

Srihari Reddy

Govind Kabra

Manik Gupta

Mohammed Waleed Kadous

Vijay Mahadevan

Vasudev Parameswaran

Juthika Kiran Dabholkar

Brandon Iles

Timo Pekka Pylvaenaeinen

William J Salisbury

Paul Mahon

Will Salisbury

Zach Song

A system and method for managing a network service is described. A system can provide a map interface for a user that includes a location pin. The system can detect a user input that sets the location pin at a selected pickup location on the map interface. In response to detecting the user input, the system can (i) determine an alternative pickup location based at least in part on historical information corresponding to clustered trip entries of the network service, and (ii) cause the map interface to dynamically and visually relocate the location pin to the alternative pickup location. The system may then receive a transport request from the mobile computing device of the user, and in response to receiving the transport request, select a transport provider to rendezvous with the user at the alternative pickup location.

Determining and providing predetermined location data points to service providers

A system and method for arranging a transport service is described. A system can receive a request for transport service from a user device. The request can include a requested pickup location data point. The system determines a predetermined location data point associated with the requested pickup location data point. The first predetermined location data point can be different from the requested pickup location data point. The system selects a driver to perform a transport service for the user based, at least in part, on the requested pickup location data point, and transmits the predetermined location data point to a driver device of the selected driver.

A system trains and applies a machine learning model to label maps of a region. Various data modalities are combined as inputs for multiple data tiles used to characterize a region for a geographical map. Each data modality reflects sensor data captured in different ways. Some data modalities include aerial imagery, point cloud data, and location trace data. The different data modalities are captured independently and then aggregated using machine learning models to determine map labeling information about tiles in the region. Data is ingested by the system and corresponding tiles are identified. A tile is represented by a feature vector of different data types related to the various data modalities, and values from the ingested data are added to the feature vector for the tile. Models can be trained to predict characteristics of a region using these various types of input.

Determining geographical map features with multi-sensor input

A service arrangement system coordinates service providers who provide geographical location-based services to users. The location-based services may include transportation of users. For example, a service provider operates a vehicle to transport a user from an origin location to a destination location. To provide an engaging user experience and encourage users to continue using the services, the service arrangement system accurately and expediently determines the origin location and the destination location based on querying available data sources. The service arrangement system can use selection rules for multiplexing, fallback logic, or load-balancing to select a high quality data source from the available data sources.

Geographical location search using multiple data sources

A travel coordination system identifies geographic locations responsive to a location search request. The geographic locations are organized into a plurality of hierarchical geohash levels, each geohash level mapping the geographic location into a geohash of a different area. The locations search request designates a search context for the search, such as whether the user is searching for an origin of a trip or a destination. The context of the search can be used to determine a geohash level for selecting candidate search results. The candidate search results may be scored based on the context using prior user selections and trips scheduled with the trip coordination system. A geohash and context associated with the user's request may be used to modify a scoring function of the candidate search results based on prior trips.

Contextual search for location services

Growing community of inventors

San Francisco, CA, United States of America

Aditya Sarawgi