Gm Global Technolgoy Operations, Inc.

Marcus J Huber

Alireza Esna Ashari Esfahani

Hyukseong Kwon

Lei Wang

Miguel Arturo Saez

Praveen Palanisamy

Qinglin Zhang

Sudhakaran Maydiga

Rodolfo Valiente Romero

Michael Cui

Upali Priyantha Mudalige

Sayyed Rouhollah Jafari Tafti

Andrew Howe

Alexei Kopylov

Alexander Waagen

An end-to-end perception perturbation modeling system for a vehicle includes one or more controllers storing a detection model in memory. The detection model includes a plurality of detection model plots that each indicate a probability value that an object in an environment surrounding the vehicle is detected based on a current weather condition and a distance measured between the vehicle and the object detected in the environment surrounding the vehicle. The one or more controllers execute instructions to receive an input state of the vehicle that is observed during non-inclement weather conditions and indicates one or more vehicle states, the current weather condition, and the distance. The controllers determine a perturbed state of the vehicle observed during the current weather condition.

End-to-end perception perturbation modeling system for a vehicle

According to several aspects, a method for path planning for a vehicle includes determining a predicted trajectory of a remote vehicle. The predicted trajectory of the remote vehicle includes a plurality of predicted trajectory nodes. The method further includes determining a plurality of possible trajectories for the vehicle. The plurality of possible trajectories includes a plurality of possible trajectory nodes. The method further includes determining one or more evaluation metrics of each of the plurality of possible trajectory nodes based at least in part on a weather condition in an environment surrounding the vehicle. The method further includes selecting an optimal trajectory for the vehicle from the plurality of possible trajectories based at least in part on the one or more evaluation metrics of each of the plurality of possible trajectory nodes. The method further includes performing a first action based at least in part on the optimal trajectory.

Weather-informed path planning for a vehicle

A system comprises a computer including a processor and a memory. The memory includes instructions such that the processor is programmed to: receive sensor data representing a perceived driving environment, select a reinforcement learning agent from a plurality of reinforcement learning agents based on a challenge score calculated using the sensor data and a desired driving style, and generate, via the selected reinforcement learning agent, a driving action based on the sensor data.

Automated driving system with desired level of driving aggressiveness

A method for identifying distracted pedestrians. The method includes determining operating conditions of a vehicle using a plurality of vehicle controllers. Pedestrian parameters for a pedestrian in a vicinity of the vehicle are acquired using a plurality of vehicle sensors. The pedestrian parameters include at least one of face positions, body positions, gait and hand gestures. Information related to an environment surrounding the vehicle is acquired. Pedestrian awareness level is determined based on the acquired pedestrian parameters and based on the information related to the environment surrounding the vehicle. A determination is made whether the pedestrian awareness level is below a predefined threshold. The pedestrian is classified as distracted, in response to determining that the pedestrian awareness level is below the predefined threshold.

Identification of distracted pedestrians

A method for improving pedestrian safety is provided. The method includes determining operating conditions of a vehicle using a plurality of vehicle sensors. A path of the vehicle is predicted based on the determined operating conditions. Pedestrian parameters for a pedestrian in a vicinity of the vehicle are acquired using the plurality of vehicle sensors. The pedestrian parameters include at least one of a position, a speed of the pedestrian, gait, body posture and a level of distractedness. A determination is made whether a notification of the pedestrian is necessary based on the determined vehicle operating conditions and the acquired pedestrian parameters. A mode of notification of the pedestrian is selected from a plurality of modes of notification, in response to determining that the notification of the pedestrian is necessary. The pedestrian is notified using the selected mode of notification.

Multimodal vehicle-to-pedestrian notification system

Systems, Apparatuses and Methods for implementing a neural network system for controlling an autonomous vehicle (AV) are provided, which includes: a neural network having a plurality of nodes with context to vector (context2vec) contextual embeddings to enable operations of the AV; a plurality of encoded context2vec AV words in a sequence of timing to embed data of context and behavior; a set of inputs which comprise: at least one of a current, a prior, and a subsequent encoded context2vec AV word; a neural network solution applied by the at least one computer to determine a target context2vec AV word of each set of the inputs based on the current context2vec AV word; an output vector computed by the neural network that represents the embedded distributional one-hot scheme of the input encoded context2vec AV word; and a set of behavior control operations for controlling a behavior of the AV.

Systems, apparatus, and methods for embedded encodings of contextual information using a neural network with vector space modeling

Systems and methods are provided for controlling a vehicle. Control signals are generated at a high-level controller based on one or more sources of input data, comprising at least one of: sensors that provide sensor output information, map data and goals. The high-level controller comprises first controller modules comprising: an input processing module, a projection module, a memories module, a world model module, and a decision processing module that comprises a control model executor module. The control signals are processed at a low-level controller to generate commands that control a plurality of vehicle actuators of the vehicle in accordance with the control signals to execute one or more scheduled actions to be performed to automate driving tasks.

Growing community of inventors

Saline, MI, United States of America

Marcus J Huber