Gm Global Technolgoy Operations, Inc.

The University of Waterloo

Brent Navin Roger Bacchus

Bo Yu

Sayyed Rouhollah Jafari Tafti

Michael A Losh

Wenyuan Qi

Rakesh Kumar

Jiabin Shen

Tarek Abdel Raouf Abdel Rahman

Donald K Grimm

Curtis L Hay

Amir Khajepour

Joon Hwang

Shu Chen

Mansoor Alghooneh

Guangyu J Zou

Gui Chen

Indranil Palit

Peter Edward LeVasseur

Ryan Jay Yergin

Neel Pratik Bhatt

Ahmad R Alghooneh

Ruihe Zhang

A perception and map edge association system for an autonomous vehicle includes one or more controllers executing instructions to receive a plurality of perception lane edge candidates that are each representative of a perception lane edge and a plurality of map lane edge candidates that are each representative of a map lane edge. The one or more controllers identify a discrepancy between a selected perception lane edge candidate and a best fit map edge candidate for a lane characteristic based on a posterior association probability matrix and a discrepancy flag. The one or more controllers segment the pair of inconsistent perception and map lane edges into one or more matching segments. The one or more matching segments indicate where the perception lane edge and the map lane edge match one another based on the lane characteristics. The perception and map edge association system improves localization and scene creation performance.

Perception and map edge association system for accommodating inconsistent perception and map data

A system for estimating cloud cover from moving vehicles includes: vehicle mounted sensors communicating with one or more control modules. The control modules have processors, memory, and input/output (I/O) ports. The I/O ports of the control modules of the vehicles communicate with the sensors. The control modules execute program code stored in the memory including first and second algorithm portions. The first algorithm portion, collects sensor data from the sensors, estimates atmospheric parameters from the sensor data; and computes measurement coordinates from the sensor data and atmospheric parameters. The second algorithm portion generates and selectively updates a cloud map based on from data received from the first algorithm portion, and from high-definition map (HD Map) data.

System and method for estimating cloud cover from a moving vehicle

A lane edge fusion system for an autonomous vehicle includes one or more controllers executing instructions to receive perception data and map data of a roadway the autonomous vehicle is traveling along. The one or more controllers derive a plurality of map lane edge points from the map data and a plurality of perception lane edge points from the perception data and select an evaluation point based on the plurality of map lane edge points and the plurality of perception lane edge points. The one or more controllers fit an implicit function for the evaluation point based on an implicit moving least squares approach. The one or more controllers build a fused lane edge by setting a point on the implicit curve as one of a plurality fused lane edge points.

Lane edge fusion system for an autonomous vehicle

A method for quantifying map errors includes receiving first map data and second map data. The method includes receiving a road topographic map. The method further includes dividing the road into road segments. The method further includes creating a plurality of bounding boxes for each of the plurality of road segments. The method includes creating a first map tile and a second map tile by filtering out the bounding boxes. The method includes executing point cloud registration to align the plurality of first data points in the first map tile with the plurality of second data points in the second map tile to determine a plurality of absolute offsets between the plurality of first data points and the plurality of second data points. The method includes determining a relative map error between the first map and the second map based on the absolute offsets.

Method and system for quantifying map errors

A method can be used to provide smart notifications to avoid collisions while the vehicle maneuvers in a tight structural environment, such as a home garage or an underground parking lot. The method includes receiving historical vehicle-trajectory data. The historical vehicle-trajectory data includes the location and the heading of the vehicle for each of the plurality of historical trajectories along the structure. The method further includes clustering the plurality of historical trajectories of the vehicle along the structure by types of maneuvers to generate a plurality of trajectory clusters. The method also includes creating a probability distribution bitmap using the plurality of trajectory clusters and creating a topographic map based on the probability distribution bitmap.

Smart notifications for vehicle parking garage maneuvers

An automobile vehicle visual wireless-based positioning system, incudes an automobile vehicle having a radio receiver. A map contains candidate locations of access-points (APs) and access-point corresponding media-access-control (MAC) IDs. A wireless range sensor determines different ranges of various detected APs visible to the automobile vehicle. An image collection feature identifies image data visible to the automobile vehicle. A real-time feature matching element matching features identified by image collection feature with data from the map. A filter receives an output from the real-time feature matching element to generate an automobile vehicle pose.

Visual wireless based positioning exploiting multipath for outdoor positioning

A method generating planning-based attention signals includes receiving driving-scene data. The driving-scene data is indicative of a driving scene around a host vehicle. The driving-scene data includes map data and localization data. The driving scene includes a plurality of actors. The method further includes converting the driving-scene data into a scene-graph. The method further includes inputting the scene-graph into a deep neural network (DNN). Further, the method includes determining an attention score for each of the plurality of actors using the DNN and the scene-graph. The attention score of each of the plurality of actors represents a priority given to each of the plurality of actors in the driving scene. The method further includes commanding the host vehicle to autonomously drive according to a trajectory determined by taking into account the attention score of each of the plurality of actors.

Algorithm to generate planning-based attention signals

A scene creation system for an autonomous vehicle includes one or more controllers executing instructions to receive perception data and map data of a roadway the autonomous vehicle is traveling along, and identify a plurality of lane segments of the roadway that the autonomous vehicle travels along based on the perception data and the map data. The one or more controllers connect the plurality of lane segments together based on a spatial relationship between the plurality of lane segments to create a lane graph. The one or more controllers classify each of the plurality of lane segments of the lane graph to one or more lane attributes and reassemble the plurality of lane segments to create a representation of the roadway. The one or more controllers recreate a scene of an environment surrounding the autonomous vehicle based on the representation of the lanes that are part of the roadway.

Scene creation system for an autonomous vehicle

A method for generating garage parking notifications includes receiving current parked-vehicle data indicating that a vehicle has been parked inside a garage. The current parked-vehicle data includes the current location of the vehicle parked inside the garage. The method further includes determining whether the current location of the vehicle inside the garage is within a warning zone. The warning zone is an area inside the garage where the vehicle has not been frequently parked. The warning zone area is determined using historical parked-vehicle data. The method further includes generating a notification to warn an operator of the vehicle that the vehicle should be moved in response to determining that the current location of the vehicle inside the garage is within the warning zone.

Smart notifications for garage parking collision avoidance

A method for efficient autonomous driving planning includes receiving a current driving-scene data and a predicted driving-scene data. The current driving-scene data is indicative of a current driving scene around a host vehicle. The predicted driving-scene data is indicative of a predicted driving scene around the host vehicle. The predicted driving scene around the host vehicle is different from the current driving scene around the host vehicle. The method further includes converting the current driving-scene data and the predicted driving-scene data into a first scene-graph and a second scene-graph, respectively. The method further includes determining a plurality of scene change metrics using the first scene-graph and the second scene-graph. The method further includes selecting between a first trajectory planning process and a second trajectory planning process based on the plurality of scene change metrics.

System and method for efficient planning under uncertainty for autonomous vehicles

A monocular camera system for a vehicle includes a front mono camera, a side mono camera, and one or more controllers in electronic communication with the front mono camera and the side mono camera. The one or more controllers execute instructions to determine an ideal lagged distance the vehicle travels between a current time step and a previous time step as two asynchronous camera frames are captured by the front mono camera and the side mono camera, determine a number of delayed frames captured by either the front mono camera or the side mono camera between the current time step and the previous time step based on the ideal lagged distance. The controllers determine a direction of travel of the vehicle that indicates which mono camera is selected to provide a previous camera frame captured at the previous time step.

Monocular camera system performing depth estimation of objects surrounding a vehicle

A system of a motor vehicle includes a radar unit for generating a radar signal associated with a target positioned about the vehicle. The system further includes a tracker generating a tracker signal associated with a radar heading and a doppler based on the radar signal. The system further includes a wheel speed sensor for generating a wheel speed signal associated with a velocity of the vehicle. The system further includes a gyroscope for generating a gyro signal associated with a measured yaw rate. The system further includes a computer having a processor and a computer readable medium. The processor is programmed to determine a gyro drift and a corrected yaw rate while the vehicle is in motion, with the gyro drift and the corrected yaw rate being based on at least the radar heading, the doppler effect, and the velocity of the vehicle, and the measured yaw rate.

System and process for correcting gyroscope drift for a motor vehicle

A system for determining relevance of a light source to an automobile includes at least one camera adapted to capture images of light sources in proximity to the automobile, a controller in communication with the at least one camera and adapted to receive captured images from the at least one camera, the controller further adapted to estimate an orientation of at least one light source relative to the automobile, classify the at least one light source as one of relevant and irrelevant, and, when the at least one light source is classified as relevant, send information about the at least one light source to a planning module for the automobile.

Determination of traffic light orientation

A system that determines a nominal path based on interpolated lane edge data can include a processor and a memory. The memory includes instructions such that the processor is configured to receive a sensor data representing a perceived lane edge; receive map data including a lane edge; interpolate the sensor data and interpolate the map data; fuse the interpolated sensor data and the interpolated map data; and generate predicted lane edge lane edge centers based on the fused interpolated sensor data.

Localizing and updating a map using interpolated lane edge data

A system for associating perceived and mapped lane edges can include a processor and a memory. The memory includes instructions such that the processor is configured to receive a sensor data representing a perceived object; receive map data representing a map object; determine a cost matrix a cost matrix indicative of an association cost for associating the map object to the perceived object; compare the association cost with an association cost threshold; and associate the perceived object with the map object based on the association cost.

Growing community of inventors

Sterling Heights, MI, United States of America

Brent Navin Roger Bacchus