Gm Global Technolgoy Operations, Inc.

Mason D Gemar

Fan Bai

Joon Hwang

Bo Yu

Shuqing Zeng

Gui Chen

Donald K Grimm

Paul Edward Krajewski

Orhan Bulan

Vivek Vijaya Kumar

David Hahn Clifford

Scott E Jackson

Donal B McErlean

Jiang-Ling Du

Kaydee Hartmann

Najah Ghalyan

Geoffrey Wakefield

Carl P Darukhanavala

Yehenew G Mengistu

Sheetal Mahesh

Sean Douglas Vermillion

Mohammed H Al Qizwini

Jim McCord

Mubassira Khan

Rajesh Ayyalasomayajula

Damien P Rennick

A method for crowd-sourcing lane line map data for a vehicle includes receiving a plurality of observations. The method also includes classifying the plurality of observations into a plurality of observation categories. Each of the plurality of observation categories includes at least one of the plurality of observations. The method also includes determining a plurality of aligned point clouds based at least in part on the plurality of observations. One of the plurality of aligned point clouds corresponds to each of the plurality of observation categories. The method also includes determining a plurality of lane line maps based at least in part on the plurality of aligned point clouds. One of the plurality of lane line maps corresponds to each of the plurality of aligned point clouds. The method also includes updating a map database based at least in part on the plurality of lane line maps.

Crowd-sourcing lane line maps for a vehicle

A system for fusing two or more versions of map data together includes one or more central computers that receive road network data representing a road network for a predefined geofenced area. The central computers compute a plurality of points that are each positioned at a predetermined distance from one another. The central computers create a plurality of bounding boxes for the road network based on the plurality of points and create a set of closest matched map data points for each bounding box that is part of the road network by executing a map-matching registration algorithm to align the two or more versions of map data with one another. The central computers execute a maximum likelihood estimation algorithm to determine probability distribution parameters of the set of closest matched map data points compared to the ground truth map data.

System for fusing two or more versions of map data

A method for merging lane line maps includes determining a road topology of a road segment. The method also includes identifying a plurality of fused points based at least in part on the road topology and based at least in part on a first lane line map of the road segment and a second lane line map of the road segment. The method also includes forming a fused lane line map based at least in part on the plurality of fused points. The method also includes performing a first action based at least in part on the fused lane line map.

Merging lane line maps for a vehicle

A method and system of map data reconciliation in connected vehicles. The method includes processing information to detect a same object; determining an identification of the same object; determining a confidence level of the identification of the same object; determining a probability mass function (PMF) of the identification of the same object using Dempster-Shafer Theory; determining a most probably identification of the same object based on the PMF; and communicating the most probably identification of the same object to a vehicle to implement a vehicle function. The system includes a conflict resolution module configured to receive confidence levels from multiple vehicles detecting the same object and to determine a reconciled identification of the same object using an algorithm based on Dempster-Shafer Theory, and a weight calculation module configured to apply a weight function based on a scene context weight function and on a historical score of the vehicles.

Method and system of inconsistent map data reconciliation in connected vehicles

A computer is provided for a system for detecting, characterizing, and mitigating road network congestion. The system includes a plurality of motor vehicles. Each motor vehicle includes a telematics control unit (TCU) for generating one or more location signals for a location of the associated motor vehicle and one or more event signals for an event related to the associated motor vehicle. The computer includes one or more processors for receiving the location signal and/or the event signal from the TCU of the associated motor vehicles. The computer further includes a non-transitory computer readable storage medium (CRM) including instructions, such that the processor is programmed to identify a location of the road network congestion at a current time step. The processor is further programmed to track the road network congestion and predict the road network congestion at a next time step.

System and process for mitigating road network congestion

A method for detecting road edges at a predetermined intersection, comprising: receiving, by the controller, aerial imagery data about the predetermined intersection; receiving, by the controller, vehicle telemetry data from at least one vehicle passing through the predetermined intersection; detecting, using the aerial imagery data and generative adversarial networks (GANs) executed on the controller, road edges at the predetermined intersection; classifying, using the vehicle telemetry data and a random forest classifier (RFC) executed on the controller, each vehicle trajectory passing through the predetermined intersection with a label corresponding to a unique maneuver to create a maneuver labeling at the predetermined intersection; constructing, using the maneuver labeling determined by the RFC and the road edges, a probabilistic finite state automata (PFSA) to pair inbound lanes with outbound lanes at the predetermined intersection; and determining lane edges at the predetermined intersection using a homotopy model.

System and method for detecting road and lane connections at intersections

A system is provided for mitigating a road network congestion. The system includes a plurality of motor vehicles positioned in associated locations in a road network. Each vehicle has one or more sensors generating an input and a Telematics Control Unit (TCU) for generating at least one location signal for a location of the associated motor vehicle and at least one event signal for an event related to the associated motor vehicle, with the location signal and the event signal corresponding to a High Speed Vehicle Telemetry Data (HSVT data) based on the input from the sensors. The system further includes a computer, which communicates with a display device and the TCUs. The computer includes a processor and a computer readable medium including instructions, such that the processor is programmed to identify the congestion and determine that the congestion is a recurring or a non-recurring congestion.

System and process for determining recurring and non-recurring road congestion to mitigate the same

A system, map server and method of navigating a vehicle through an intersection. The map server includes a remote processor and a communication device. The remote processor determines a first road edge for a first road entering an intersection and a second road edge for a second road entering the intersection. The remote processor constructs an intersection edge that connects a first point on the first road edge to a second point on the second road edge. The communication device communicates the intersection edge to the vehicle. A vehicle processor at the vehicle navigates the vehicle through the intersection using the intersection edge.

Geometry-based model for road edge detection at intersections

A system comprises a processor and a memory storing instructions. The processor receives an image for processing using a reinforcement learning based agent comprising a neural network trained using a reward function. The image includes N lane lines of a roadway, where N is a positive integer. The instructions configure the processor to traverse the image using the agent at least N times from a first end of the image to a second end of the image by: incrementally moving the agent from a first side of the image to a second side of the image after each traversal; and maximizing rewards for the agent using the reward function during each traversal of the image using the agent. The instructions configure the processor to identify the N lane lines of the roadway as a single cluster of lane lines after traversing the image using the agent at least N times.

Identification and clustering of lane lines on roadways using reinforcement learning

System and methods for estimating a centerline of a road that separates traffic moving in opposite directions include aggregating a data set from each of a plurality of vehicles traversing the road over a period of time as telemetry data. Each data set of the telemetry data indicates a location and a heading. The method includes clustering the data sets of the telemetry data based on the heading indicated by each data set, and identifying a separator to separate the data sets indicating a first heading from the data sets indicating a second heading, opposite to the first heading. The centerline is estimated based on applying a spatial smoothing to the separator.

Estimation of road centerline based on vehicle telemetry

Systems and method are provided for predicting a vehicle pose. In one embodiment, a system includes a non-transitory computer readable medium. The non-transitory computer readable medium includes: an asynchronous prediction module configured to, by a processor, receive vehicle data from a vehicle, and compute a prediction of vehicle pose from the vehicle data, wherein the computation is performed asynchronously using a vehicle time frame and a common coordinate system; and a synchronous prediction module configured to, by the processor, compute a final prediction of the vehicle pose across the vehicle population based on the prediction of vehicle pose from the asynchronous prediction module, wherein the computation is performed synchronously using a global time frame and the common system.

Systems and methods for vehicle pose prediction

Apparatuses, methods, and systems of an edge server communicating with vehicles by continuously receiving messages which include data of state transitions of Electronic Control Units (ECUs) contained in each vehicle; monitoring in real-time the streamed message data from each vehicle for detection of updates of data about state transitions of ECUs used in the vehicular operations; crowdsourcing in real-time the state transition data of the ECUs during the vehicular operations to group together state transition data having commonality as crowdsource transition data; processing, using a machine learning model, the crowdsource transition data, to classify the crowdsource transition data of an anomaly event; and transmit in advance to an identified vehicle an alert of the anomaly event in order for a driver or ADAS to execute a response.

Growing community of inventors

Cedar Park, TX, United States of America

Mason D Gemar