Gm Global Technolgoy Operations, Inc.

Keun Jae Kim

Padma Sundaram

Orhan Bulan

Andrew H Leutheuser

Stephen R Pastor

Michael A Losh

Rami I Debouk

Brent N Bacchus

Phanu Amatyakul

Rakesh Kumar

Dean C Degazio

Brian N Siskoy

Collin N Kappauf

A computer-implemented method for detecting one or more three-dimensional structures in a proximity of a vehicle at runtime includes generating, by a processor, a birds-eye-view (BEV) camera image of the proximity of the vehicle, the BEV camera image comprising two-dimensional coordinates of one or more structures in the proximity. The method further includes generating, by the processor, a BEV height image of the proximity of the vehicle, the BEV height image providing height of the one or more structures in the proximity. The method further includes detecting one or more edges of the three-dimensional structures based on the BEV camera image and the BEV height image. The method further includes generating models of the three-dimensional structures by plane-fitting based on the edges of the one or more three-dimensional structures. The method further includes reconfiguring a navigation system receiver based on the models of the three-dimensional structures.

Detecting three-dimensional structure models at runtime in vehicles

Systems and methods are provided for generating a map for use in controlling a vehicle. In one embodiment, a method includes: receiving, by a processor, aerial image data depicting an environment; processing, by the processor, the aerial image data with a plurality of trained deep learning models to produce a predicted radar map; and controlling the vehicle based on the predicted radar map.

Methods and systems for generating radar maps

Systems and methods are provided for localizing a vehicle. In one embodiment, a method includes: receiving, by a processor, radar data from a radar of the vehicle; creating, by the processor, a local map based on the radar data; retrieving, by the processor, a map of an environment from a datastore based on a predicted vehicle pose; correlating, by the processor, the local map and the retrieved map; determining, by the processor, a localized vehicle pose based on the correlating; and controlling, by the processor, the vehicle based on the localized vehicle pose.

Methods and systems for mapping and localization for a vehicle

An automotive vehicle includes a vehicle steering system, an actuator configured to control the steering system, a first controller, and a second controller. The first controller is in communication with the actuator. The first controller is configured to communicate an actuator control signal based on a primary automated driving system control algorithm. The second controller is in communication with the actuator and with the first controller. The second controller is configured to, in response to a first predicted vehicle path based on the actuator control signal deviating from a desired route by a threshold distance, control the actuator to maintain a current actuator setting. The second controller is also configured to, in response to the first predicted vehicle path not deviating from the desired route by the threshold distance, control the actuator according to the actuator control signal.

Trajectory integrity co-pilot for autonomous vehicle

An automotive vehicle includes a vehicle steering system, an actuator configured to control the steering system, a first controller, and a second controller. The first controller is in communication with the actuator. The first controller is programmed with a primary automated driving system control algorithm and is configured to communicate an actuator control signal based on the primary automated driving system control algorithm. The second controller is in communication with the actuator and with the first controller. The second controller is configured to, in response to a first predicted vehicle path based on the actuator control signal deviating from a current lane, control the actuator to maintain a current actuator setting. The second controller is also configured to, in response to the first predicted vehicle path being within the current lane, control the actuator according to the actuator control signal.

Automated lane keeping co-pilot for autonomous vehicles

Methods and systems are provided for aligning a steering system of a vehicle. In one embodiment, a method includes: determining when the vehicle is driving a straight-line path; determining a steering wheel position error when the vehicle is driving the straight-line path; computing a rear wheel steering offset based on the steering wheel position error and a closed loop control method; and generating a control signal to a rear wheel steering system based on the rear wheel steering offset.

Growing community of inventors

Novi, MI, United States of America

Keun Jae Kim