Gm Global Technolgoy Operations, Inc.

The University of Waterloo

SeyedAlireza Kasaiezadeh Mahabadi

Bakhtiar Brian Litkouhi

Seyedeh Asal Nahidi

Hualin Tan

James Heath Holbrook

Ehsan Asadi

Naser Mehrabi

Reza Hajiloo

Saurabh Kapoor

Yubiao Zhang

Shamim Mashrouteh

Shih-Ken Chen

Amir Khajepour

Hassan Askari

Ehsan Hashemi

Nikolai K Moshchuk

Qingrong Zhao

John Randall Yost

Mansour Ataei

Jinsong Wang

Arun Adiththan

Kausalya Singuru

Ruixing Long

Jin-Jae Chen

Bharath Pattipati

Arash Hashemi

Xueying Kang

Gianmarc Coppola

Sresht Gurumoorthi Annadevara

Michael W Neal

Isaac Tabatschnic

Lapo Frascati

Amir Abolhassani

Bo Yu

Wen-Chiao Lin

Jason D Fahland

Shiming Duan

Patrick J Monsere

Edward T Heil

Praveen Palanisamy

Ramesh Sethu

Michael A Ryba

Amin Habibnejad Korayem

Yechen Qin

Mustafa Hakan Turhan

Ramesh S

Asal Nahidi

Gill Lipton

Nauman Sohani

Chao Yu

Cody Wayne Bulkley

A data health monitoring and recording system for a vehicle includes a data recording infrastructure and one or more controllers. The one or more controllers include a multi-layer control software architecture including two or more software layers. The one or more controllers execute instructions to monitor, by a health monitoring structure that is part of the multi-layer control software architecture, one or more parameters calculated by each of the two or more software layers. The one or more controllers compare a respective parameter of a respective software layer with a predetermined corresponding performance metric for a required time horizon. In response to determining the respective parameter of the respective software layer does not meet the predetermined corresponding performance metric for the required time horizon, a trigger that instructs the data recording infrastructure to record the parameters calculated by each of the two or more software layers is generated.

Vehicle data health monitoring and recording system for multi-layer control software architecture

A system for vehicle motion control (VMC) performance enhancement utilizing real-time data reliability and criticality assessments includes a vehicle having sensors and actuators. The sensors collect real-time dynamic state information about the vehicle. The actuators actively and continuously adjust the dynamic state of the vehicle. The system executes a VMC application that obtains, from the sensors and actuators, dynamic state information of the vehicle, and estimates a real-time vehicle dynamic state from the dynamic state information. The VMC application determines a signal criticality and a signal reliability for vehicle dynamic state information, and executes a VMC strategy based on the real-time vehicle dynamic state, the signal criticality and reliability. The VMC application detects and mitigates signal degradation by enacting a VMC strategy, and generates a VMC output command to the actuators based on the VMC strategy. The VMC strategy actively, continuously, and automatically adapts to seamlessly overcome signal degradation.

Vehicle motion control performance enhancement using real-time data reliability and criticality assessments

A system for learning-model predictive control (LMPC) with multi-step prediction for motion control of a vehicle includes sensors and actuators. One or more control modules each having a processor, a memory, and input/output (I/O) ports are in communication with the sensors and actuators, the processor executing program code portions stored in the memory. The program code portions cause the sensors and actuators to obtain vehicle state information, receive a driver input, and generate a desired dynamic output based on the driver input and the vehicle state information. A program code portion estimates actions of the actuators based on the vehicle state information and the driver input, and utilizes the vehicle state information, the driver input, and the estimated actions of the actuators to select one or more models of a physics-based vehicle model and a machine-learning model of the vehicle to selectively adjust commands to the actuators.

Learning-model predictive control with multi-step prediction for vehicle motion control

A motor vehicle motion control health monitoring system includes sensors and actuators disposed on the motor vehicle. The sensors measure real-time static and dynamic telemetry data about the motor vehicle, and the actuators alter static and dynamic behavior of the motor vehicle. A control module has a processor, a memory, and input/output (I/O) ports. The processor executes program code portions stored in the memory, the program code portions include: an offline portion that collects telemetry data from the motor vehicle, performs failure analysis on the telemetry data and allocates tasks based on the failure analysis; and an online portion that analyzes the telemetry data for failures within specific sensors, actuators, or functions that utilize systems of sensors and/or actuators. The online portion mitigates deviations in the telemetry data by sending a correction to the one or more sensors, actuators, and/or functions of a motor vehicle motion control system.

Architecture and methodology for vehicle motion control health monitoring and mitigation

A method for controlling an autonomous vehicle includes receiving road data. The road data includes information about a plurality of potential events along the road ahead of the autonomous vehicle. The method further includes determining, in real time, a probability that the plurality of potential events along the road ahead of the autonomous vehicle will occur while the autonomous vehicle moves along the road and determining, in real time, an adjusted planned path using a probabilistic predictive control that takes into account the probability that the plurality of potential events along the road ahead of the autonomous vehicle will occur. Further, the method includes controlling the autonomous vehicle to cause the autonomous vehicle to autonomously follow the adjusted planned path.

Architecture and methodology for defensive autonomous driving using a probabilistic control strategy

A fault remediation system for a vehicle includes one or more controllers in electronic communication with one or more consumed interfaces and one or more provided interfaces. The one or more controllers execute instructions to receive, from the one or more consumed interfaces, a consumed signal and perform fault detection upon the consumed signal to determine the presence of an active fault within the consumed signal. In response to detecting an active fault with the consumed signal, the one or more controllers select a remediation state from a group of two or more prospective remediation states based on a significance analysis of the consumed signal. The one or more controllers evaluate a relevant subfunction that corresponds to the consumed signal that the remediation state addresses for the presence of remediation tolerance and generates arbitration instructions based on the remediation tolerance.

Fault remediation system for a vehicle

A driver command interpreter system for a vehicle includes one or more controllers that execute instructions to receive a plurality of dynamic variables, vehicle configuration information, and driving environment conditions, and determine a target vehicle state during transient driving conditions based on the plurality of dynamic variables from the one or more sensors, the vehicle configuration information, and the driving environment conditions. The one or more controllers build a transient vehicle dynamic model based on the target vehicle state during transient driving conditions, the plurality of dynamic variables, the vehicle configuration information, and the driving environment conditions, and solve for desired zeros corresponding to the target vehicle state during transient conditions.

Driver command interpreter system determining achievable target vehicle state during steady-state and transient driving conditions

A chassis control system of a vehicle includes: a first torque source providing torque to a first axle of the vehicle; a second torque source providing torque to a second axle of the vehicle independently of the first torque source; and a vehicle control module controlling the first and second torque sources to transition between a torque redistribution and torque limit control modes based on a wheel torque redistribution threshold, a wheel torque limit of the first axle and/or second axle, and a torque limit of one of the first and second torque sources. The torque redistribution mode refers to when torque is being selectively provided to the first axle by the first torque source and to the second axle by the second torque source. The torque limit control mode refers to when torque to at least one of first axle and the second axle is limited.

Coordinated vehicle dynamic handling balance control using torque management

A system for managing chassis and driveline actuators of a motor vehicle includes a control module executing program code portions that: cause sensors to obtain vehicle state information, receive a driver input and generate a desired dynamic output based on the driver input and the vehicle state information, and then estimate actuator actions based on the vehicle state information, generate one or more control action constraints based on the vehicle state information and estimated actuator actions, generate a reference control action based on the vehicle state information, the estimated actions of the one or more actuators and the control action constraints, and integrate the vehicle state information, the estimated actuator actions, desired dynamic output, reference control action and the control action constraints to generate an optimal control action that falls within a range of predefined actuator capacities and ensures driver control of the vehicle.

Architecture and model predictive control-based methodology to manage chassis and driveline actuators

A method for vehicle motion control includes receiving sensor data from a plurality of sensors of a vehicle and monitoring a vehicle response of the vehicle using the sensor data. The vehicle response is represented by a plurality of vehicle-response signals. The method further includes fusing the plurality of vehicle-response signals to obtain at least one fused signal. The method further includes determining whether to activate a vehicle stability control of the vehicle based on the at least one fused signal and commanding the vehicle to activate the vehicle stability control in response to determining to activate the vehicle stability control of the vehicle based on the at least one fused signal.

Data fusion-centric method and system for vehicle motion control

A universal machine learning based system for estimating a vehicle state of a vehicle includes one or more controllers executing instructions to receive a plurality of dynamic variables and corresponding historical data. The controllers execute a sensitivity analysis algorithm to determine a sensitivity level for each dynamic variable and corresponding historical data and select two or more pertinent dynamic variables based on the sensitivity level of each dynamic variable and the corresponding historical data. The controllers standardize the two or more pertinent dynamic variables into a plurality of generic dynamic variables, wherein the plurality of generic dynamic variables are in a standardized format that is applicable to any configuration of vehicle, and estimate the vehicle state based on the plurality of generic dynamic variables by one or more machine learning algorithms.

Universal machine learning based system for estimating a vehicle state

A system for adaptive tire force prediction in a motor vehicle includes a control module that executes program code portions that receive real-time static and dynamic data from motor vehicle sensors, that model forces at each tire of the motor vehicle at one or more incremental time steps, that estimate actual forces at each tire of the motor vehicle at each of the one or more incremental time steps, that adaptively predict tire forces at each tire of the motor vehicle at each of the one or more incremental time steps, that generate one or more control commands for actuators of the motor vehicle, that capture discrepancies between real-time force estimations and nominal force calculations at each tire of the motor vehicle, and that apply compensation parameters to reduce tracking errors in the one or more control commands to the one or more actuators of the motor vehicle.

Architecture and methodology of adaptive tire force prediction for integrated vehicle motion control

Systems and methods for determining whether a vehicle is in an understeer or oversteer situation. The system includes a controller circuit coupled to an IMU and an EPS, and programmed to: calculate, for a steered first axle, an axle-based pneumatic trail for using IMU measurements and EPS signals and estimate a saturation level as a function of a distance between the axle-based pneumatic trail and zero. The system estimates, for an unsteered second axle, an axle lateral force curve with respect to a slip angle of the second axle, and a saturation level as a function of when the axle lateral force curve with respect to the slip angle transitions from positive values to negative values. The saturation level of the first axle and the second axle are integrated. The system determines that the vehicle is in an understeer or oversteer situation as a function of the integrated saturation levels.

Systems and methods for determining whether a vehicle is in an understeer or oversteer situation

A method for determining, in real-time, an electronic limited-slip differential (eLSD) clutch torque includes receiving vehicle data in real-time, wherein the vehicle data includes a torque request, determining a preliminary eLSD clutch torque using a neural network and the vehicle data, determining clutch torque bounds of the eLSD using a physics-based model, determining whether the preliminary eLSD clutch torque is outside the clutch torque bounds of the eLSD, adjusting the preliminary eLSD clutch torque using clutch torque bounds to determine a final clutch torque of the eLSD in response to determining that the preliminary eLSD clutch torque is outside the clutch torque bounds of the eLSD, and commanding, in real-time, the eLSD to apply the final clutch torque to a clutch of the eLSD.

Real-time machine learning and physics-based hybrid approach to perform eLSD torque estimation

A data analytics system for a vehicle includes one or more controllers having a multi-layered control software architecture including two or more software layers and a health monitoring structure. The one or more controllers execute instructions to receive a fault, and in response to receiving the fault, generate, by the health monitoring structure, a plurality of unique triggers as the fault propagates throughout the multi-layered control software architecture. The controllers build a network including a plurality of propagation paths and identify one or more main propagation paths that are part of the network based on a data analysis of a mock big data set. The one or more controllers build a complete propagation pathway including two or more main propagation paths. In response to determining only one main propagation path, the controllers determine a root cause of the fault based on the only one main propagation path.

Vehicle data analytics system for multi-layer control software architecture

Growing community of inventors

Shelby Township, MI, United States of America

SeyedAlireza Kasaiezadeh Mahabadi