Uber Technologies, Inc.

University of California

Andrew T Irish

Upamanyu Madhow

Daniel Iland

Srinivasan Rajagopal

Jason T Isaacs

Venkata Sathya Praveen Gorthy

Francois Quitin

Joao P Hespanha

Mohammed Waleed Kadous

Theodore Russell Sumers

Dinesh Ramasamy

Vivek Sankaravadivel

Andrew Joseph Beinstein

Zelin Liu

Sejal Kiran Shinde

Kartheeka Srinivas Varma Konduru

Hemabh Shekar

Shu Yang Li

Various examples are directed to systems and methods for generating improved autonomous vehicle (AV) navigation route data and estimated time of arrival (ETA) data. A system may receive delivery data from an AV delivering a passenger or product, where the delivery data includes a delivery origin, a delivery destination, and an ETA and navigation route generated by the AV. The AV may have accurate short-term information, such as the next few navigation maneuvers planned by the AV. The system may be able to provide more accurate longer-term information, such as generating an overall calculated route and calculated completion time based a current location and the delivery destination. The system may receive updated short-term information and combine this information with the calculated route and calculated completion to generate a more accurate set of route data and ETA data.

Autonomous vehicle delivery ingestion and routing

A travel coordination system determines a route for a rider using a public transit system and a provider. The travel coordination system may determine route that describes a public transit stop at which the rider exits the public transit system, and the travel coordination system can route a provider so that the provider transports the rider when the rider arrives at the public transit stop. The travel coordination system may update the rider's route after transmitting the route to the rider. The travel coordination system may determine a public transit vehicle on which the rider is traveling or predict the rider's destination. If multiple riders are traveling on the public transit vehicle and if those riders exit the public transit station using the same public transit stop, the travel coordination system may match those riders together for transport by a provider.

Coordinating travel on a public transit system and a travel coordination system

A network system, such as a transport management system, infers movement and a location of a vehicle associated with a transportation service using sensor data from a provider client device and a wireless device mounted in a fixed position in the vehicle. Before or during a transportation service, the provider client device transmits sensor data to the network system for use in detecting the occurrence of one or more specified events, such as a sudden deceleration or a harsh turn. The network system fuses the received sensor data to infer the movement of the vehicle along forward, lateral, and vertical axes and implements an event detector by analyzing movement of the vehicle in the forward direction. Fused sensor data received from the wireless device is used to validate the detected movement and to determine a position of the vehicle.

Inferring vehicle location and movement using sensor data fusion

A method of determining location of a user device includes receiving global navigation satellite system (GNSS) fix data that represents GNSS calculated position of the user device, receiving signal strength data associated with each satellite communicating with the user device, and receiving satellite data regarding locations of satellites. The method further includes retrieving satellite blocking values from a cache that describe a likelihood of a satellite signal being blocked at a plurality of possible locations. A non-linear filter, implemented by one or more processors, is applied to the GNSS fix data, signal strength data, and satellite blocking values to generate an updated position estimate of the user device.

System and method for localization and tracking

Various embodiments each include at least one of systems, methods, devices, and software for GNSS simultaneous localization and mapping (SLAM). The disclosed techniques demonstrate that simultaneous localization and mapping (SLAM) can be performed using only GNSS SNR and geo-location data, collectively termed GNSS data henceforth. A principled Bayesian approach for doing so is disclosed. A 3-D environment map is decomposed into a grid of binary-state cells (occupancy grid) and the receiver locations are approximated by sets of particles. Using a large number of sparsely sampled GNSS SNR measurements and receiver/satellite coordinates (all available from off-the-shelf GNSS receivers), likelihoods of blockage are associated with every receiver-to-satellite beam. Loopy Belief Propagation is used to estimate the probabilities of each cell being occupied or empty, along with the probability of the particles for each receiver location.

Systems and methods for GNSS SNR probabilistic localization and 3-D mapping

A travel coordination system determines a route for a rider using a public transit system and a provider. The travel coordination system may determine a route that describes a public transit stop at which the rider exits the public transit system, and the travel coordination system can route a provider so that the provider transports the rider when the rider arrives at the public transit stop. The travel coordination system may update the rider's route after transmitting the route to the rider. The travel coordination system may determine a public transit vehicle on which the rider is traveling or predict the rider's destination. If multiple riders are traveling on the public transit vehicle and if those riders exit the public transit station using the same public transit stop, the travel coordination system may match those riders together for transport by a provider.

An intermediary node receives broadcasts from servers each indicating three-dimensional map tiles loaded into the server's memory using geohashes. The intermediary node updates a set of ordered lists of server identifiers based on the geohashes. The intermediary node receives a request a client device including location information and a client identifier. The intermediary node generates a client geohash based on the location information. The intermediary node identifies an ordered list using the client geohash. The intermediary node selects a server identifier from the identified ordered list based on the client identifier. The intermediary node initiates a network connection between the client and the selected server.

Load balancing sticky session routing

A localization server improves position estimates of global navigation satellite systems (GNSS) using probabilistic shadow matching and pseudorange matching is disclosed herein. The localization server may utilize one or more of the following information: the locations of the satellites, the GNSS receiver's location estimate and associated estimated uncertainty, the reported pseudoranges of the satellites, the GNSS estimated clock bias, the SNRs of the satellites, and 3D environment information regarding the location of the receiver. The localization server utilizes a Bayesian framework to calculate an improved location estimate using the GNSS location fixes, pseudorange information, and satellite SNRs thereby improving localization and tracking for a user device.

Localization and tracking using location, signal strength, and pseudorange data

A method of determining location of a user device includes receiving global navigation satellite system (GNSS) fix data that represents GNSS calculated position of the user device. The method further includes receiving signal strength data associated with each satellite communicating with the user device, and receiving map information regarding environment surrounding the user device. The received GNSS fix data and signal strength data is provided to a non-linear filter, wherein the non-linear filter fuses the GNSS fix data and signal strength data to generate an updated position estimate of the user device. In addition, the non-linear filter utilizes probabilistic shadow matching estimates that represent a likelihood of received signal strength data as a function of hypothesized user device locations within the environment described by the received map information.

System and method for localization and tracking using GNSS location estimates, satellite SNR data and 3D maps

A client device maintains location state data including a first location estimate of a geographic location of the client device. The first location estimate is based on a first motion measurement obtained over a first time period. The client device retrieves a second motion measurement obtained over a second time period subsequent to the first time period and uses it and the first location estimate to generate a second location estimate. The client device sends the second location estimate to a server. The server further processes the second location estimate to generate an updated second location estimate. The client device retrieves a third motion measurement and generates a third location estimate. The client device receives the updated second location estimate and uses it to adjust the third location estimate. A fourth location estimate is generated using the adjusted third location estimate.

Growing community of inventors

San Diego, CA, United States of America

Andrew T Irish