maneuver prediction based speed profile adaptation for behavior-based automatic cruise control

Abstract

An operating action prediction based speed profile adaptation system for behavior-based automatic cruise control in a host vehicle includes sensors and actuators in communication with a controller. The controller executes an operating action prediction speed profile adaptation (MPSA) application that captures static and dynamic information about the host vehicle and about the environment of the host vehicle. The MPSA application processes sensor information and performs: feature extraction, synchronization, normalization, generates a probabilistic path prediction and detects the intent of the vehicle operator. When the host vehicle is traveling in a fully automatic and / or semi-automatic mode, the MPSA application generates control commands to the actuators to selectively at least change the longitudinal speed of the host vehicle to approximate the historical preferences of the vehicle operator for lateral and longitudinal speed, yaw, and forces observed by the vehicle operator.

Description

Technical Field

[0001] This disclosure relates to a cruise control system in a vehicle, and more specifically, to an advanced driver assistance system in a vehicle that performs cruise control functions. Background Technology

[0002] Vehicles equipped with Advanced Driver Assistance Systems (ADAS) use multiple sensors to detect the vehicle's surroundings and use actuators to adjust vehicle performance in relation to the surrounding environment by using Adaptive Cruise Control (ACC) functionality.

[0003] While current ADAS-based ACC systems and methods have achieved their intended purpose, there is still a need for a new and improved ADAS-based ACC system and method that combines action prediction and speed adaptation to ensure that vehicle performance using ADAS and ACC adaptively adjusts to the input preferences of the vehicle operator or driver, thereby improving vehicle operator comfort while maintaining the efficiency of ADAS and ACC, improving the accuracy and responsiveness of ACC, and delivering a smoother, smarter driving experience, while maintaining or reducing system complexity and increasing redundancy. Summary of the Invention

[0004] According to several aspects, a speed curve adaptive system based on predictive operation of behavior for adaptive cruise control in a host vehicle includes: a host vehicle, one or more sensors, and one or more actuators. The one or more sensors detect static and dynamic state information about the host vehicle, and the one or more actuators modify the static and dynamic performance of the host vehicle. The system also includes a controller having a processor, memory, and input / output (I / O) ports. The I / O ports communicate with the one or more sensors and the one or more actuators. The processor executes a portion of program code stored in memory. The program code portion includes a Predictive Operation of Speed ​​Curve Adaptive (MPSPA) application having at least first, second, third, and fourth control logic. The first control logic captures static and dynamic information about the host vehicle and its surrounding environment. The second control logic processes the static and dynamic information about the host vehicle and its surrounding environment. The processing includes at least: feature extraction, synchronization, normalization, and generation of probabilistic path predictions. The third control logic detects the intent of the vehicle operator. The fourth control logic generates control commands to the one or more actuators. When the master vehicle is operating in fully automatic and / or semi-automatic mode, the control command selectively changes at least the longitudinal speed of the master vehicle to approximate the historical preferences of the vehicle operator for lateral and longitudinal speeds, yaw, and forces observed by the vehicle operator.

[0005] In another aspect of this disclosure, the first control logic further includes control logic for obtaining the position, motion, and acceleration information of the main vehicle in at least three degrees of freedom from the inertial measurement unit (IMU). The first control logic also includes control logic for obtaining the steering angle of the main vehicle from one or more steering angle sensors (SAS), control logic for obtaining the accelerator pedal position from an accelerator pedal position sensor, and control logic for obtaining the brake pedal position from a brake pedal position sensor.

[0006] In another aspect of this disclosure, the second control logic also includes control logic for filtering data from one or more sensors and performing feature extraction. By selecting only certain portions of the data from one or more sensors that can indicate changes in the primary vehicle's state, the filtering sequence causes the system and the MPSPA application to reduce computational resource utilization from a first level to a second level below the first level. Feature extraction further determines that the changes in the primary vehicle's state are large enough to require additional processing.

[0007] In another aspect of this disclosure, the second control logic also includes control logic for synchronizing data from one or more sensors. Synchronizing the data ensures that the data from the one or more sensors are aligned and correlated in a timely and accurate manner, thereby enabling data from each of the one or more sensors to be correlated in a timely manner with data from the other sensors.

[0008] In another aspect of this disclosure, the second control logic further includes control logic for generating probabilistic path predictions using a probability function of each state variable that follows a sigmoid function, wherein: the master vehicle state variables include: master vehicle speed v x Main vehicle steering angle δ, torsion bar torque τ, and torsion bar torque ratio and

[0009]

[0010] Where x is a variable, v x ,δ,τ, M represents the type of operation, including turning, changing lanes, making a U-turn, evasive maneuvers, and / or lane following. The probability of the operation is obtained by multiplying the state probability functions, as shown below:

[0011]

[0012] Among them, P M,x Let be the probability function of variable x for operation action M; k is the sampling time, β M,x For turning parameters, the S-curve steepness of variable x in the operation action M; α M,xFor turning parameters, the S-shaped midpoint of variable x in operation action M; n is the number or quantity of variables; and g d Let g be the direction of the operation action M, such that for the operation action on the right, g d =-1, for the left-hand operation, g d =1.

[0013] In another aspect of this disclosure, the second control logic further includes control logic for integrating environmental knowledge into environmental knowledge categories, including road information, road type, number or quantity of lanes, presence / absence of construction areas, surrounding object information, nearest vehicle path (CIPV) turn signals, and navigation information. The effect of each environmental knowledge category is represented by a gain, wherein the product of the gains of all environmental knowledge categories defines the final environmental gain, expressed as:

[0014] g M,e =g M,r ×g M,s ×g M,n Equation 3

[0015] Among them, g M,r The probability gain of road information for operation action M; g M,s The probability gain of objects surrounding the action M; g M,n The probability gain of navigation information for operation action M; and g M,e Let M be the probability gain of the environmental knowledge surrounding the action. The following equation expresses the probability function of the action considering environmental knowledge:

[0016]

[0017] In another aspect of this disclosure, the second control logic further includes control logic for normalizing the sensor data. Normalizing the sensor data reduces redundant information in the sensor data from a first redundancy level to a second redundancy level, which is lower than the first redundancy level. The second control logic also includes control logic for generating a confidence score calculation. The output of the confidence score calculation defines the probability that the master vehicle is performing a specific type of operation action M.

[0018] In another aspect of this disclosure, the third control logic also includes control logic for leveraging a data-driven classification machine learning (ML) model in conjunction with a shadow planner to enhance adaptive cruise control (ACC) by predicting the driver's intended path and adjusting the driver's speed profile to match the driver's intent based on analysis of the driver's inputs to the primary vehicle control system and the behavior of surrounding vehicles. The third control logic also includes control logic for performing real-time path classification to assess the type of maneuver being actively pursued by the driver and the ADAS, and control logic for performing an enumeration calculation that limits the classified real-time path to at least one of left turns and right turns.

[0019] In another aspect of this disclosure, the third control logic also includes control logic for generating trajectory predictions based on real-time paths, Global Positioning System (GPS) data, and navigation information (including selected navigation routes).

[0020] In another aspect of this disclosure, the fourth control logic further includes control logic for generating control commands to one or more actuators. The control commands include predicted lateral commands and predicted longitudinal commands. The fourth control logic also includes control logic for activating and controlling one or more actuators of the main vehicle drivetrain control system, engine control system, and braking control system to at least change the longitudinal speed of the main vehicle while performing a currently classified real-time path operation action M according to the predicted lateral commands and predicted longitudinal commands.

[0021] In another aspect of this disclosure, a speed curve adaptation method based on predictive maneuvers for behavior-based automatic cruise control in a host vehicle includes: detecting static and dynamic state information about the host vehicle using one or more sensors of the host vehicle, and modifying the static and dynamic performance of the host vehicle using one or more actuators of the host vehicle. The method also includes a portion of program code stored in the memory of the host vehicle's controller, executed by a processor of the host vehicle's controller. The controller also includes input / output (I / O) ports communicating with one or more sensors and one or more actuators. The program code portion includes a Predictive Maneuver Speed ​​Curve Adaptation (MPSPA) application. The MPSPA application includes control logic for capturing static and dynamic information about the host vehicle and its surrounding environment, and control logic for processing the static and dynamic information about the host vehicle and its surrounding environment. The processing includes at least: feature extraction, synchronization, normalization, and generating probabilistic path predictions. The processing also includes detecting the intent of the vehicle operator and generating control commands to one or more actuators. When the master vehicle is operating in fully automatic and / or semi-automatic mode, the control command selectively changes at least the longitudinal speed of the master vehicle to approximate the historical preferences of the vehicle operator for lateral and longitudinal speeds, yaw, and forces observed by the vehicle operator.

[0022] In another aspect of this disclosure, the method further includes obtaining the position, motion, and acceleration information of the main vehicle in at least three degrees of freedom from an inertial measurement unit (IMU), obtaining the steering angle of the main vehicle from one or more steering angle sensors (SAS), obtaining the accelerator pedal position from an accelerator pedal position sensor, and obtaining the brake pedal position from a brake pedal position sensor.

[0023] In another aspect of this disclosure, the method further includes filtering and performing feature extraction on data from one or more sensors. By selecting only certain portions of the data from one or more sensors that can indicate changes in the primary vehicle's state, filtering allows the method and the MPSPA application to reduce computational resource utilization from a first level to a second level below the first level. Feature extraction further determines that changes in the primary vehicle's state are large enough to require additional processing.

[0024] In another aspect of this disclosure, the method further includes synchronizing data from one or more sensors. Synchronizing the data enables the data from the one or more sensors to be aligned and correlated in a timely and accurate manner, thereby ensuring that data from each of the one or more sensors is correlated in a timely manner with data from each of the other sensors.

[0025] In another aspect of this disclosure, the method further includes generating a probabilistic path prediction using a probability function of each state variable that follows a sigmoid function, wherein: the master vehicle state variables include: master vehicle speed v x Main vehicle steering angle δ, torsion bar torque τ, and torsion bar torque ratio

[0026]

[0027] Where x is a variable, v x ,δ,τ, M represents the type of operation, including turning, changing lanes, making a U-turn, evasive maneuvers, and / or lane following. The probability of the operation is obtained by multiplying the state probability functions, as shown below:

[0028]

[0029] Among them, P M,x Let be the probability function of variable x for operation action M; k is the sampling time, β M,x For turning parameters, the S-curve steepness of variable x in the operation action M; α M,x For turning parameters, the S-shaped midpoint of variable x in operation action M; n is the number or quantity of variables; and g d Let g be the direction of the operation action M, such that for the operation action on the right, g d =-1, for the left-hand operation, g d =1.

[0030] In another aspect of this disclosure, the method further includes integrating environmental knowledge into environmental knowledge categories that include road information, road type, number or quantity of lanes, presence / absence of construction zones, surrounding object information, nearest vehicle path (CIPV) turn signals, and navigation information. The effect of each environmental knowledge category is represented by a gain, where the product of the gains of all environmental knowledge categories defines the final environmental gain, expressed as:

[0031] g M,e =g M,r ×g M,s ×g M,n Equation 3

[0032] Among them, g M,r The probability gain of road information for operation action M; g M,s The probability gain of objects surrounding the action M; g M,n The probability gain of navigation information for operation action M; and g M,e Let M be the probability gain of the environmental knowledge surrounding the action. The following equation expresses the probability function of the action considering environmental knowledge:

[0033]

[0034] In another aspect of this disclosure, the method further includes normalizing the sensor data. Normalizing the sensor data reduces redundancy in the sensor data from a first redundancy level to a second redundancy level, which is lower than the first redundancy level. The method also includes generating a confidence score calculation. The output of the confidence score calculation defines the probability that the master vehicle is performing a specific type of operation action M.

[0035] In another aspect of this disclosure, the method also includes leveraging a data-driven classification machine learning (ML) model in conjunction with a shadow planner to enhance adaptive cruise control (ACC) by predicting the vehicle operator's intended path and adjusting the speed profile of the primary vehicle to match the vehicle operator's intent based on analysis of inputs from the vehicle operator to the primary vehicle control method and the behavior of surrounding vehicles. The method also includes performing real-time path classification to assess the type of maneuver being actively pursued by the primary vehicle operator and the ADAS; and performing an enumeration calculation that limits the classified real-time path to at least one of left turns and right turns.

[0036] In another aspect of this disclosure, the method further includes generating trajectory predictions based on real-time path, Global Positioning System (GPS) data, and navigation information (including selected navigation routes). The method also includes generating control commands to one or more actuators. The control commands include predicted lateral commands and predicted longitudinal commands. The method further includes activating and controlling one or more actuators of the main vehicle's drivetrain control system, engine control system, and braking control system to at least change the longitudinal speed of the main vehicle while performing a currently classified real-time path maneuver M according to the predicted lateral and predicted longitudinal commands.

[0037] In another aspect of this disclosure, a speed curve adaptation method based on predictive maneuvers for behavior-based adaptive cruise control in a host vehicle includes: detecting static and dynamic state information about the host vehicle using one or more sensors of the host vehicle; and modifying the static and dynamic performance of the host vehicle using one or more actuators of the host vehicle. The method also includes a portion of program code stored in the memory of the controller, executed by a processor of the host vehicle's controller. The controller also includes input / output (I / O) ports communicating with one or more sensors and one or more actuators. The program code portion includes a predictive maneuver speed curve adaptation (MPSPA) application, including: capturing static and dynamic information about the host vehicle and its surrounding environment, including: obtaining host vehicle position, motion, and acceleration information in at least three degrees of freedom from an inertial measurement unit (IMU); obtaining the host vehicle steering angle from one or more steering angle sensors (SAS); obtaining the accelerator pedal position from an accelerator pedal position sensor; and obtaining the brake pedal position from a brake pedal position sensor. The method also includes processing the static and dynamic information about the host vehicle and its surrounding environment. The processing includes at least feature extraction, synchronization, normalization, and generating probabilistic path predictions, including filtering and feature extraction of data from one or more sensors. Filtering, by selecting only portions of data from one or more sensors that can indicate changes in the master vehicle's state, reduces computational resource utilization from a first-level to a second-level below-first-level for this method and the MPSPA application. Feature extraction further determines that changes in the master vehicle's state are large enough to require additional processing. Data synchronization ensures timely and accurate alignment and correlation of data from one or more sensors, thereby ensuring that data from each of the one or more sensors is timely correlated with data from each of the other sensors. The method also includes generating probabilistic path predictions using a probability function of each state variable following a sigmoid function, where the master vehicle state variables include: master vehicle speed v. x Main vehicle steering angle δ, torsion bar torque τ, and torsion bar torque ratio

[0038]

[0039] Where x is a variable, v x ,δ,τ, M represents the type of operation, including turning, changing lanes, making a U-turn, evasive maneuvers, and / or lane following. The probability of the operation is obtained by multiplying the state probability functions, as shown below:

[0040]

[0041] Among them, P M,xLet be the probability function of variable x for operation action M; k is the sampling time, β M,x For turning parameters, the S-curve steepness of variable x in the operation action M; α M,x For turning parameters, the S-shaped midpoint of variable x in operation action M; n is the number or quantity of variables; and g d Let g be the direction of the operation action M, such that for the operation action on the right, g d =-1, for the left-hand operation, g d =1. The method also includes integrating environmental knowledge into environmental knowledge categories that include road information, road type, number or quantity of lanes, presence / absence of construction zones, surrounding object information, nearest vehicle path (CIPV) turn signals, and navigation information. The effect of each environmental knowledge category is represented by a gain, where the product of the gains of all environmental knowledge categories defines the final environmental gain, expressed as:

[0042] g M,e =g M,r ×g M,s ×g M,n Equation 3

[0043] Among them, g M,r The probability gain of road information for operation action M; g M,s The probability gain of objects surrounding the action M; g M,s The probability gain of navigation information for operation action M; and g M,e Let M be the probability gain of environmental knowledge related to the action. The following equation represents the probability function of the action considering environmental knowledge:

[0044]

[0045] The method also includes normalizing sensor data, wherein normalization reduces redundancy in the sensor data from a first redundancy level to a second redundancy level below the first redundancy level, and generating a confidence calculation. The output of the confidence calculation limits the probability that the master vehicle is performing a specific type of maneuver M. The method also includes detecting the intent of the vehicle operator, including: leveraging a data-driven classification machine learning (ML) model to integrate a shadow planner to predict the driver's expected path based on analysis of the driver's input to the master vehicle control method and the behavior of surrounding vehicles, thereby enhancing adaptive cruise control (ACC). The method also includes performing real-time path classification to assess the type of maneuver pursued by the master vehicle operator and the ADAS; and performing an enumeration calculation that limits the classified real-time path to at least one of left and right turns. The method also includes generating trajectory predictions based on the real-time path, GPS data, and navigation information (including the selected navigation route); and generating control commands to one or more actuators. The control commands include predicted lateral commands and predicted longitudinal commands. The method also includes activating and controlling one or more actuators of the master vehicle's drivetrain control system, engine control system, and braking control system to change at least the longitudinal speed of the master vehicle while performing a currently classified real-time path maneuver M based on predicted lateral and longitudinal commands. When the master vehicle is operating in fully automatic and / or semi-automatic mode, the control commands selectively change at least the longitudinal speed of the master vehicle to approximate the vehicle operator's historical preferences for lateral and longitudinal speeds, yaw, and forces observed by the vehicle operator.

[0046] Further applicability will become apparent from the description provided herein. It should be understood that the specification and specific examples are for illustrative purposes only and are not intended to limit the scope of this disclosure. Attached Figure Description

[0047] The accompanying drawings described herein are for illustrative purposes only and are not intended to limit the scope of this disclosure in any way.

[0048] Figure 1 This is a schematic diagram of a speed curve adaptive system based on predictive operation action for behavior-based automatic cruise control, according to one aspect of this disclosure.

[0049] Figure 2 It describes one aspect of the use according to this disclosure. Figure 1 A perspective top view of the road intersection where the main vehicle is turning in the system navigation;

[0050] Figure 3 It describes one aspect according to this disclosure. Figure 1The flowchart of the logical flow of the system's operation action prediction speed curve adaptive (MPSPA) application;

[0051] Figure 4A Utilization of one aspect of this disclosure Figure 3 The MPSPA application provides a first graphical representation of the lateral and longitudinal commands issued by the actuators of the master vehicle during turning; and

[0052] Figure 4B This is an application of another aspect of this disclosure. Figure 3 The MPSPA application is a second graphical representation of the lateral and longitudinal commands issued by the actuators of the main vehicle during turning. Detailed Implementation

[0053] The following description is merely exemplary in nature and is not intended to limit this disclosure, its application, or its uses.

[0054] refer to Figure 1 This illustration shows a speed profile adaptive system 10 based on predicted maneuvers for behavior-based automatic cruise control (ACC) 11. System 10 includes a host vehicle 12. The host vehicle 12 is illustrated as a passenger car; however, it should be understood that the host vehicle 12 can be any type of vehicle / vehicle, including but not limited to: cars, trucks, SUVs, vans, motorhomes, semi-trailers, tractor-trailers, transport vehicles (including those used in warehouses), tricycles, motorcycles, aircraft, amphibious vehicles, or any other such vehicle 12. Furthermore, without departing from the scope or intent of this disclosure, the host vehicle 12 can be an aircraft, vessel, etc.

[0055] System 10 also includes one or more sensors 14 disposed, attached to, or otherwise integrated into the host vehicle 12. Other sensors 14 may be located remotely from the host vehicle 12 and transmit information to the host vehicle 12, as will be described in further detail below. The sensors 14 of the host vehicle 12 may include any of a variety of types of sensors, including but not limited to: electromagnetic (EM) sensors 14, such as cameras, infrared cameras, video cameras, light detection and ranging (LiDAR) sensors, radio detection and ranging (RADAR) sensors, sound navigation and ranging (SONAR) sensors, etc. In some examples, the cameras and / or other sensors 14 of the host vehicle 12 are equipped with an external field of view (FOV), and the data collected by these cameras includes optical information about the environment in which the host vehicle 12 is operating. In other non-limiting examples, the cameras and / or other sensors 14 are oriented towards the interior of the host vehicle 12 or the passenger compartment, thereby providing information about the occupants and operators of the host vehicle 12. Other sensors 14 may include, but are not limited to: an inertial measurement unit (IMU) 16, a suspension control unit (e.g., a semi-active damping suspension (SADS) sensor 18), a global positioning system (GPS) sensor 22, a wheel speed sensor 24 capable of measuring the rotational speed of one or more wheels 26 of the main vehicle 12, an accelerator and / or throttle pedal position sensor 28, a brake pedal position sensor 30, a steering position sensor 32 capable of measuring the position, steering rate, and steering speed of the steering system 34, and a tire pressure monitoring system 36, etc.

[0056] IMU 16 can measure the motion, acceleration, etc. of the host vehicle 12 in several degrees of freedom. In a specific example, IMU 16 can measure position, motion, acceleration, etc. in at least three degrees of freedom. Similarly, SADS sensor 18 can be an IMU 16 capable of measurement in three or more degrees of freedom. In some examples, SADS 18 can be a suspension hub accelerometer, etc. Therefore, sensor 14 of the host vehicle 12 can detect and record wheel speed data, the position and location of the host vehicle 12, and static and dynamic state information (e.g., speed, acceleration, etc.) of the host vehicle 12.

[0057] As used herein, “forward,” “rearward,” “inward,” “outward,” “outward,” “above,” and “below” are terms used relative to the orientation of the main vehicle 12 shown in the accompanying drawings of this application. Thus, “forward” refers to the direction toward the front of the vehicle 12, “rearward” refers to the direction toward the rear of the vehicle 12, “inward” and “outward” refer to the direction toward the interior of the vehicle 12 or the passenger compartment 38, “outward” and “outward” refer to the direction toward the exterior of the vehicle 12, “below” refers to the direction toward the bottom of the main vehicle 12, and “above” refers to the direction toward the top of the main vehicle 12.

[0058] System 10 also includes one or more controllers 40 that communicate with various sensors 14 of the host vehicle 12, process information received therefrom, and generate output signals to assist the vehicle operator 42 in maintaining attention and avoiding highway drowsiness or white-line heat. The controllers 40 are integrated into the host vehicle 12. More specifically, the controllers 40 are non-general-purpose electronic control devices having a pre-programmed digital computer or processor 44, a non-transitory computer-readable medium or memory 46 for storing data (e.g., control logic, software applications, instructions, computer code, data lookup tables, etc.), and input / output (I / O) ports 48. The computer-readable medium or memory 46 includes any type of computer-accessible medium, such as read-only memory (ROM), random access memory (RAM), hard disk drive, optical disc (CD), digital video disc (DVD), or any other type of memory. The “non-transitory” computer-readable memory 46 does not include wired, wireless, optical, or other communication links for transmitting transient electrical or other signals. Non-transitory computer-readable storage 46 includes media capable of permanently storing data, as well as media capable of storing and later overwriting data, such as rewritable optical discs or erasable memory devices. Computer code includes any type of program code, including source code, object code, and executable code. Processor 44 is configured to execute code or instructions. The main vehicle 12 may have other controllers 40, such as a dedicated Wi-Fi controller, engine control module, transmission control module, body control module, infotainment control module, etc. Without departing from the scope or intent of this disclosure, I / O port 48 may be configured to communicate via wired communication means, wireless communication means such as Wi-Fi protocols under IEEE 802.11x, etc.

[0059] The controller 40 also includes one or more applications 50. An application 50 is a software program configured to perform a specific function or set of functions. An application 50 may include one or more computer programs, software components, instruction sets, procedures, functions, objects, classes, instances, associated data, or portions thereof suitable for implementation in suitable computer-readable program code. Applications 50 may be stored within memory 46 or in other or separate memory 46. Examples of applications 50 include audio or video streaming services, games, browsers, social media, etc. In other examples, applications 50 are used to manage the functions of the main vehicle 12 body control system, suspension control system 54 functions, steering control system 56 functions, powertrain control system 58 functions, including the control system functions of the transmission 59 and / or engine 60, braking system 62 control functions, etc. More specifically, the main vehicle 12 is equipped with multiple control systems that manage the static and dynamic performance characteristics of the main vehicle 12 through multiple on-board actuators 64 equipped on the vehicle 12.

[0060] Without departing from the scope or intent of this disclosure, the actuator 64 may take various forms and manage many different and / or interconnected main vehicle 12 control systems. It should be understood that the actuator 64 may be any combination of the above-described types of actuable devices for altering one or more static and / or dynamic performance attributes of the main vehicle 12, including electric, hydraulic, pneumatic, mechanical, electromechanical, electrohydraulic, electropneumatic, magnetorheological, hydraulic-pneumatic, electromagnetic, and / or other types used to change one or more static and / or dynamic performance attributes of the main vehicle 12.

[0061] In some non-limiting examples, the suspension control system 54 includes one or more suspension system actuators 64, such as active or semi-active shock absorbers 66, capable of altering the damping forces transmitted from the wheels 26 of the main vehicle 12 to the body 68 of the main vehicle 12 as the main vehicle 12 travels on a road surface. The steering control system 56 actuators may include electric motors, electro-hydraulic actuators, electro-pneumatic actuators, or other such motors or steering actuators 70, which apply torque to the steering shaft 72 or steering rack 74 of the main vehicle 12, thereby changing the direction of travel of the main vehicle 12 by altering the position or angular orientation of the steering wheels 26 of the main vehicle 12. Conversely, the on-board actuators 64 of the transmission 59 or engine 60 control system may include the transmission 59 or engine 60 itself and / or actuating components therein, capable of altering the torque output or torque ratio of the engine 60 or transmission 59, etc. In some non-limiting examples, the transmission 59 or engine 60 control system actuator 76 may include a throttle or electronic throttle 78 capable of changing the torque output of the engine 60, a transmission actuator 80 capable of changing the gear ratio and torque output transmitted from the engine 60 through the transmission 59, etc. It should also be understood that, although... Figure 1The main vehicle 12 shown is equipped with an internal combustion engine (ICE) 60, but without departing from the scope or intent of this disclosure, the engine 60 can be any type of engine 60 or prime mover, such as an ICE engine 60, an electric motor, a hybrid engine 60, a combination thereof, or any other known type of engine 60. Similarly, the brake system 62 actuator 64 includes brakes 82 of the main vehicle 12, which are capable of selectively slowing the rotational speed of the wheels 26 of the main vehicle 12, thereby changing the speed of the main vehicle 12 itself.

[0062] The main vehicle 12 can operate in any of a variety of different modes, including a fully manual mode in which the vehicle operator 42 has complete control over the static and dynamic performance characteristics of the main vehicle 12. In other non-limiting examples, the main vehicle 12 can operate in a fully automatic or semi-automatic mode that controls some or all of the static and dynamic performance characteristics of the main vehicle 12. More specifically, the system 10 of this disclosure operates on a main vehicle 12 having an advanced driver assistance system (ADAS) 84 capable of controlling the adaptive cruise control (ACC) 11 function, maneuvering the vehicle, braking, and any of a variety of other ways of controlling the onboard main vehicle 12 systems.

[0063] Now for reference Figure 2 and Figure 3 And continue to refer to Figure 1 System 10 utilizes one or more applications 50, particularly the Predictive Speed ​​Curve Adaptive (MPSPA) application 52, which uses sensor 14 data, vehicle operator 42 preference data, position data, etc. to adjust the ACC 11 speed curve of the master vehicle 12, while allowing the vehicle operator 42 to comfortably and accurately perform route or direction changes via steering input without disengaging from ACC.

[0064] System 10 and MPSPA application 52 enhance the control of the master vehicle 12 by combining action prediction and speed curve adaptation. System 10 and MPSPA application 52 operate similarly to a shadow planner, predicting where the vehicle operator 42 wants to go and adjusting the speed curve of the master vehicle 12 to optimally match the operator's intentions. By analyzing the inputs of the vehicle operator 42 and surrounding vehicles 12', and predicting the actions of the surrounding vehicles 12', system 10 and MPSPA application 52 adaptively adjust the ACC 11 speed curve to ensure efficient and comfortable operation of the master vehicle 12. In a non-limiting example, the master vehicle 12 may be decelerating when the vehicle operator 42 uses ACC to make a left turn. System 10 and MPSPA application 52 improve the accuracy and responsiveness of ACC, resulting in a smoother and smarter driving experience than System 10 without such features. System 10 and MPSPA application 52 activate ACC 11 when the master vehicle 12 is actively operating in complex manual driving scenarios (e.g., turning, driving along different paths, etc.).

[0065] In several aspects, System 10 and MPSPA application 52 perform operational action prediction in mathematical processing or derivation. To construct probabilistic operational action predictions, probability functions are used to represent each vehicle state variable by following a sigmoid function. The current master vehicle 12 state variables include: master vehicle 12 speed v. x The main vehicle's 12-degree steering angle δ, torsion bar torque τ, and torsion bar torque rate

[0066]

[0067] Where x is a variable, v x ,δ,τ, M represents the type of operation, including turning, changing lanes, making a U-turn, evasive maneuvers, and / or lane following.

[0068] The probability of an action is obtained by multiplying it by the state probability function, as shown below:

[0069]

[0070] Among them, P M,x Let be the probability function of variable x for operation action M; k is the sampling time, β M,x For turning parameters, the S-curve steepness of variable x in the operation action M; α M,x For turning parameters, the S-shaped midpoint of variable x in operation action M; n is the number or quantity of variables; and g d For the direction of the operation action M, for the right-side operation action, g d =-1, for the left-hand operation, gd =1.

[0071] System 10 and MPSPA application 52 also integrate environmental knowledge to increase the confidence level of the predictions generated by System 10 and MPSPA application 52. Environmental knowledge is used for different categories, including but not limited to: road information, such as road type, number or number of lanes, presence / absence of construction zones; surrounding object information, such as nearest vehicle path (CIPV) turn signals; navigation information, such as selected navigation routes, etc. The effect of each category of environmental knowledge is represented by a gain, depending on the operation being performed by the master vehicle 12. When it is determined that some or all of the environmental knowledge is irrelevant to a particular operation, the gain is set to one (1), while when the environmental knowledge is relevant to a particular operation, the gain is greater than 1 based on the importance level. For example, when the master vehicle 12 is traveling on a highway, the road information gain for turning is 1, but the road information gain for lane changing is greater than 1 because the probability of lane changing is higher than that of turning on a highway. These gains are multiplied to form the final environmental gain, which can be expressed as:

[0072] g M,e =g M,r ×g M,s ×g M,n Equation 3

[0073] Among them, g M,r The probability gain of road information for operation action M; g M,s The probability gain of objects surrounding the action M; g M,n The probability gain of navigation information for operation action M; and g M,e Let M be the probability gain of the environmental knowledge surrounding the action. The following equation expresses the probability function of the action considering environmental knowledge:

[0074]

[0075] To maintain the comfort and trust of the vehicle operator 42 with system 10 and MPSPA application 52, system 10 and MPSPA application 52 actively adjust the ACC 11 function through driving style adaptation. Driving style adaptation can be viewed as a subroutine or algorithm of MPSPA application 52. During driving style adaptation, the probability of each predicted maneuver, along with the associated master vehicle 12 state variables, is stored over time in a buffer of size L. This is achieved by adjusting the initial turning parameters... and The mean and standard deviation of the buffer state variables are taken, and the turning parameter α is used. M,x and β M,x Updated as follows:

[0076]

[0077] in, Limit the level of distribution. and The learning rate is used. Finally, system 10 and MPSPA application 52 are based on road curvature ρ and slope. Adjust the turning parameters using slope θ and road friction.

[0078]

[0079] Then, the ACC 11 update based on the above-mentioned operation action prediction calculation is performed, so that the longitudinal control command (i.e., the command to control the longitudinal acceleration and / or deceleration of the master vehicle 12) can be expressed as:

[0080] [ΔV x，ref ,Δa x，ref ]′=f(E,R,P M Equation 12 (V,τ,…)

[0081] In a more specific but non-limiting embodiment, the longitudinal control command can be expressed as:

[0082] [ΔV x，ref ,Δa x，ref ]′=(k1·V·k2(R)·(1-k3·E)+k4·τ)·k5(P M Equation 13

[0083] Considering the speed V of the main vehicle 12; the current speed of the main vehicle 12 (in m / s), the road type R, such as highway, urban road, rural road, etc.; environmental complexity E; driving environment complexity (e.g., traffic density, weather), the torque (τ) of the vehicle operator 42 is the torque (Nm) applied by the vehicle operator 42 to the steering wheel of the main vehicle 12; and the prediction of the driving action (P) M This is a predicted maneuver type (e.g., turning, lane changing, U-turn). It should be understood that each maneuver type may have different effects on the longitudinal control command. For example: Turning: may require deceleration to ensure safety. Lane changing: may require slight acceleration to safely merge. U-turn: usually requires significant deceleration. Longitudinal control command [ΔV] x,ref ,Δa x,ref ]′: The required acceleration or deceleration command (m / s) 2 ) and its derivative.

[0084] The design parameters and calibrations for the above factors will be described in more detail, as follows:

[0085] k1 is the speed scaling factor. The value of k1 is a constant that scales the basic command based on the speed of the master vehicle 12. The k1 speed scaling factor determines how much acceleration or deceleration is applied relative to the current speed of the master vehicle 12. Higher values ​​of the k1 speed scaling factor increase the responsiveness of the system 10 and MPSPA application 52 at higher speeds, while lower values ​​make the system 10 and MPSPA application 52 more conservative, especially in urban environments.

[0086] k2 is the road type adjustment factor. The k2 road type adjustment factor is based on the road type adjustment ACC 11 command. Different road types require different driving behaviors. In some non-restrictive examples, when the main vehicle 12 is driving on a highway, the k2 road type adjustment factor is close to one (e.g., 1.0), while during city driving, the k2 road type adjustment factor is less than one (e.g., 0.8), and when the main vehicle 12 is driving on rural roads or in rural conditions, the k2 road type adjustment factor is slightly greater than one (e.g., 1.1).

[0087] k3 is the environmental complexity factor. The k3 environmental complexity factor takes into account the complexity of the driving environment of the master vehicle 12, thus affecting the degree of reduction of ACC 11 commands under challenging conditions. In several non-limiting examples, a higher k3 environmental complexity factor value causes system 10 and MPSPA application 52 to perform more cautious driving in complex environments (e.g., traffic congestion), while a lower k3 environmental complexity factor value allows system 10 and MPSPA application 52 to perform more aggressive driving in simpler environments.

[0088] k4 is the driver torque influence factor. The k4 driver torque influence factor constant determines the degree to which the steering input (torque) of the vehicle operator 42 affects the longitudinal ACC 11 command. In some non-limiting examples, a higher k4 driver torque influence factor value results in a greater impact of the vehicle operator 42 input on the acceleration / deceleration of the master vehicle 12, while a lower k4 driver torque influence factor value results in a higher level of automatic control, making the acceleration / deceleration of the master vehicle 12 less affected by the vehicle operator 42 input.

[0089] k5(P M k5(P) is the adjustment factor for predicting operational actions. M The operation prediction adjustment factor modifies the command based on the predicted operation type (e.g., turning, lane change). In several non-restrictive examples, turning: (k5 < 1) (e.g., 0.7); lane change: (k5 ~ 1) (e.g., 1.0); and U-turn has been associated with or significantly reduced values ​​(e.g., 0.5).

[0090] exist Figure 2The example shown below illustrates system 10 and MPSPA application 52 in more detail, using the scenario where the main vehicle 12 makes a left turn at intersection 100. The main vehicle 12 operates with adaptive cruise control (ACC) 11 enabled, and is designed for comfortable passage. Figure 2 As shown, for a left turn, system 10 and MPSPA application 52 change the longitudinal speed of the main vehicle 12 by applying the control system functions of the main vehicle 12 brakes 82, engine 60, transmission 59, or other such powertrain 58. Figure 3 In Figure 4, within boxes 200, 202, 204, and 206, system 10 and MPSPA application 52 utilize sensors 14 of the host vehicle 12 to determine the current static and / or dynamic operating state of the host vehicle 12. More specifically, MPSPA application 52 and system 10 utilize one or more IMUs 16, one or more steering position or steering angle sensors (SAS) 32, and accelerator pedal position sensor 28 and brake pedal position sensor 30 to generate state information about the host vehicle 12. Subsequently, in box 208, system 10 and MPSPA application 52 process the data from sensors 14. To further refine the sensor 14 data, system 10 and MPSPA application 52 may utilize camera sensors, vehicle operator 42 monitoring information, wheel speed sensors 26, GPS sensor 20 data, and real-time traffic information.

[0091] In box 208, system 10 and MPSPA application 52 filter sensor 14 data in box 210, perform feature extraction in box 212 based on the filtered data from box 210, and then synchronize the sensor 14 data in box 214. By filtering the data in box 210, MPSPA application 52 and system 10 are able to reduce computational resource utilization from a first utilization level to a second utilization level below the first utilization level by selecting only certain portions of sensor 14 data that can indicate a change in the state or command of the master vehicle 12 or vehicle operator 42. To further refine and selectively manipulate the sensor 14 data, the feature extraction in box 212 is applied to the filtered data from box 210 to determine whether the change in the state or command of the master vehicle 12 or vehicle operator 42 is significant enough to require additional processing. In some examples, features found in the feature extraction in box 212 may include lane changes, route-guided turns, highway lane merging, etc. Synchronization of sensor 14 data ensures that data obtained from each sensor 14 of the master vehicle 12 is timely, correct, and accurate, and is aligned and correlated, so that sensor 14 data from each of the master vehicle 12 sensors is timely correlated and aligned with data obtained from the other sensors 14 of the master vehicle 12. System 10 and MPSPA application 52 then generate probabilistic path predictions in block 216. As previously described, in order to generate probabilistic path predictions, a probability function is used for each master vehicle 12 state variable, expressed as a sigmoid function:

[0092]

[0093] Where x is a variable, v x ,δ,τ, M represents the operation type, including turning, lane changing, U-turn, evasive steering, and / or lane following. The current master vehicle's 12 state variables include: master vehicle 12 speed v. x The main vehicle's 12-degree steering angle δ, torsion bar torque τ, and torsion bar torque rate

[0094] In box 218, system 10 and MPSPA application 52 normalize the sensor 14 data to improve the quality, integrity, flexibility, performance, and availability of the sensor 14 data from a first level to a second level above the first level. Furthermore, the normalization of the sensor 14 data reduces redundant information in the sensor 14 data from a first redundancy level to a second redundancy level below the first redundancy level. In box 220, system 10 and MPSPA application 52 perform confidence calculations using various categories of environmental knowledge, as previously described, where the confidence calculation output defines the probability that the master vehicle 12 is performing a specific type of operational action M. Subsequently, from box 220 in box 208, system 10 and MPSPA application 52 proceed to box 222, where system 10 and MPSPA 52 perform intent detection of vehicle operator 42.

[0095] In several aspects, the vehicle operator 42 intent detection process utilizes a machine learning (ML) model 224 to classify the vehicle operator 42 intent in a data-driven classification process. More specifically, system 10 and the MPSPA application 52 incorporate a shadow planner to predict the expected path of the vehicle operator 42 and adjust the speed curve of the master vehicle 12 to match the intent of the vehicle operator 42, thereby enhancing ACC 11, based on analysis of the inputs of the vehicle operator 42 to the master vehicle 12 control system and the behavior of surrounding vehicles 12'. In box 226, system 10 and the MPSPA application 52 perform real-time path classification to assess the type of maneuver that the vehicle operator 42 and / or ADAS 84 of the master vehicle 12 are actively pursuing. Different types of master vehicle 12 maneuvers are classified in a data-driven manner and may include real-time path classifications such as "left turn," "right turn," etc. The vehicle operator 42 intent can be expressed as one or more of lateral commands 300 and longitudinal commands 400, such as Figure 4A and Figure 4B This is illustrated in more detail below. System 10 and MPSPA application 52 then perform enumeration calculations in box 228. The enumeration calculations in box 228 limit the classified real-time path to define left turns, right turns, etc., and then in box 230, system 10 and MPSPA application 52 generate a trajectory prediction based on the real-time path, GPS data, and navigation information (e.g., selected navigation route, etc.). The output of the driver intent detection process in box 222 is the change in vehicle speed ΔV. x,ref The change in vehicle acceleration Δa x,ref The speed of the main vehicle 12 is generated and sent to the ACC 11 to change the speed of the main vehicle 12 by activating and controlling one or more actuators 64 of the transmission 59 and / or engine 60 control system and brake control system 62, etc.

[0096] Now, more specifically, turn to Figure 4A and Figure 4B And continue to refer to Figures 1 to 3 It shows in more detail Figure 2 The figure shows a graphical representation of the operation of system 10 and MPSPA application 52 during an exemplary turn. Figure 4A and Figure 4B Lateral command 300 and longitudinal command 400 are depicted respectively. Lateral command 300 is a command issued to actuator 64 of master vehicle 12 to change the lateral trajectory of master vehicle 12. That is, lateral command 300 is a command that applies differential torque to the wheels 26 of master vehicle 12 via actuator 64 of transmission control system 59 and / or engine control system 60, and / or actuator 64 of brake control system 62, thereby changing one or more torque outputs to the wheels 26 of master vehicle 12 on opposite sides (i.e., the left and right sides of master vehicle 12), causing master vehicle 12 to turn or yaw. In other non-limiting examples, lateral command 300 is an input to steering system 34 or a command acting on steering system 34 via steering actuator 70, applying torque to steering shaft 72 and / or steering rack 74 of master vehicle 12, causing master vehicle 12 to turn or yaw in the lateral direction rather than the longitudinal direction.

[0097] In contrast, longitudinal command 400 defines commands from system 10 and MPSPA application 52 to one or more of the transmission 59 and engine control system, and more specifically, to engine control system actuator 76 and / or transmission actuator 80, to change the torque output of engine 60 and transmission 59, thereby changing the speed, rate, and / or acceleration of the main vehicle 12. Similarly, longitudinal command 400 may also include applying torque to the wheels 26 of the main vehicle 12 via transmission 59 and / or engine 60 control system actuator 64, and / or brake control system 62 actuator 64, thereby causing a change in the rotational speed of the wheels 26 and resulting in a change in the speed, rate, and / or acceleration of the main vehicle 12.

[0098] like Figure 4A and Figure 4B As shown, horizontal command 300 and vertical command 400 are illustrated, where the command amplitude is depicted along the Y-axis and the time is depicted along the X-axis. See details. Figure 4A The vertical dashed line 402 defines the ADAS 84, and specifically, the event where the ACC 11 of the master vehicle 12 takes action to adjust the speed of the master vehicle 12. More specifically, the event could be turning, etc., for example... Figure 2The turning is as depicted in the diagram. It should be understood that in a fully manually driven vehicle 12, when vehicle 12 begins to turn, the vehicle operator 42 adjusts the longitudinal speed of the main vehicle 12 from the previous cruising speed to the desired speed, where the cruising speed of the main vehicle 12 is greater than the desired speed. Upon reaching the desired speed, the vehicle operator 42 initiates the turn by engaging or otherwise generating an input command through the steering actuator 70 acting on the steering system 34, which applies torque to the steering shaft 72 and / or steering rack 74 of the main vehicle 12, causing the main vehicle 12 to turn or yaw, thereby guiding the main vehicle 12 to turn. After the turn is completed, or when sufficient steering input has been made, the steering input can be cut off or otherwise reduced so that the wheels 26 of the main vehicle 12 move forward straight again, and the vehicle operator 42 applies throttle input in the direction of travel of the main vehicle 12 to accelerate the main vehicle 12 back to cruising speed. Figure 4A The illustration depicts a master vehicle 12 operating in a highly automated mode, such as Level 2 autonomous driving mode, in which ADAS 84 applies lateral and longitudinal inputs to actuators 64 of the master vehicle 12 to enable the master vehicle 12 to... Figure 2 The turn shown.

[0099] In comparison, such as Figure 4BAs shown, system 10 and MPSPA application 52 monitor the master vehicle 12's sensors 14 and actuators 64 when the master vehicle 12 is manually operated and when the master vehicle 12 is operated automatically or semi-automatically by ADAS and ACC. Over time, MPSPA application 52 and system 10 are thus trained to recognize situations initiating a turn and adaptively adjust the longitudinal and lateral inputs 404 to achieve lateral and longitudinal speeds, yaws, and forces observed by the vehicle operator 42 consistent with each vehicle operator 42's historical preferences. That is, in some examples, the master vehicle 12's sensors 14 detect information about each vehicle operator 42 and store this vehicle operator 42 information as vehicle operator 42 preferences in the memory 46 of the master vehicle 12's controller 40. The vehicle operator 42 information may include information about each vehicle operator 42's typical driving style and method of operating the master vehicle 12. Based on the vehicle operator 42 preferences, system 10 and MPSPA application 52 generate predicted driving actions, including predicted lateral commands 302 and predicted longitudinal commands 400. In several respects, the predicted lateral command 302 defines the output commands of system 10 and MPSPA application 52 to alter the lateral movement of the master vehicle 12 in a manner approximating or mimicking the lateral command 304 stored in memory 46 applied by vehicle operator 42. In some non-limiting examples, vehicle operator 42 preferences may be stored in the master vehicle 12 controller 40 memory 46 based on each vehicle operator 42 or each vehicle operator 42 identifier. That is, each individual vehicle operator 42 may log in or otherwise indicate their presence in the master vehicle 12, at which point system 10 and MPSPA application 52 recall the vehicle operator 42 preferences in memory 46 corresponding to the currently logged-in vehicle operator 42 and adjust the ACC 11 and ADAS functions and operations of the master vehicle 12 according to the stored preferences of the currently logged-in vehicle operator 42.

[0100] In other non-limiting examples, vehicle operator 42 preferences (including vehicle operator 42 lateral commands 304) are stored in memory 46 local to the controller 40 of the main vehicle 12 and in other or auxiliary memory 46 separate from and remote from the main vehicle 12. That is, vehicle operator 42 lateral commands are also stored in other or auxiliary memory 46 remote from the controller 40 of the main vehicle 12. In some non-limiting examples, without departing from the scope or intent of this disclosure, such a remote controller 40 may be located in or on a cloud computing server 406, a satellite-based server, and / or any other location that is electronically wirelessly communicating with the controller 40 of the main vehicle 12.

[0101] In several respects, the system 10 and MPSPA application 52 of this disclosure can operate on a vehicle 12 with Level 1 and / or Level 2 autonomous driving capabilities. That is, the host vehicle 12 can be equipped with a system that provides steering, braking, and / or acceleration assistance to the vehicle operator 42 under predefined conditions. Such Level 1 autonomous driving may include, for example: adaptive cruise control (ACC) 11, which maintains a safe distance from surrounding vehicles 12' in front of the host vehicle 12 by controlling the acceleration and braking of the host vehicle 12; automatic lane keeping, which helps keep the host vehicle 12 within its current lane by detecting lateral movement of the host vehicle 12 within its lane and correcting such movement by changing the steering position; and / or parking assistance, which can assist the host vehicle 12 in reversing and parking without input from the vehicle operator 42. In contrast, Level 2 autonomous driving provides partial assistance to the vehicle operator 42 by using ACC, lane keeping assist (LKA) providing gentle steering input, and steering, acceleration, and / or braking control to help the primary vehicle 12 stay in its current lane; automatic parking assist for parallel or perpendicular parking through steering and / or braking and acceleration control; and traffic jam assist, which assists the primary vehicle 12 in maintaining a set speed and following the primary vehicle 12 ahead in slow-moving traffic. In a non-limiting example, the system 10 and MPSPA application 52 of this disclosure adaptively control the longitudinal command 400 via ACC 11 in the Level 1 autonomous vehicle 12 using task or route planning information. In contrast, in the Level 2 autonomous vehicle 12, the system 10 and MPSPA application 52 of this disclosure adaptively control at least the longitudinal command 400 via ACC, and in some cases also control the lateral command 300.

[0102] The System 10 and MPSPA application 52 disclosed herein offer several advantages. These advantages include leveraging the capabilities of ADAS-based ACC 11, which combines action prediction and speed adaptation to ensure that vehicle performance using ADAS and ACC 11 adaptively adjusts to the input preferences of the vehicle operator or driver, thereby improving vehicle operator comfort while maintaining the efficiency of ADAS and ACC 11, improving the accuracy and responsiveness of ACC, resulting in a smoother, smarter driving experience, while maintaining or reducing the complexity of System 10 and increasing the redundancy of System 10.

[0103] The description in this disclosure is merely exemplary in nature, and variations thereof without departing from the spirit and scope of this disclosure are intended to fall within its scope. Such variations should not be considered as departing from the spirit and scope of this disclosure.

Claims

1. A speed profile adaptive system based on predictive operation for behavior-based automatic cruise control in a primary vehicle, comprising: Main vehicle; One or more sensors, the one or more sensors detecting static state information and dynamic state information about the master vehicle; One or more actuators that alter the static and dynamic properties of the master vehicle; A controller having a processor, memory, and input / output (I / O) ports communicating with the one or more sensors and the one or more actuators, the processor executing a portion of program code stored in the memory, the program code including an adaptive MPSPA application for predicting operational motion velocity curves, comprising: The first control logic is used to capture static and dynamic information about the main vehicle and the surrounding environment of the main vehicle; The second control logic is used to process the static and dynamic information about the master vehicle and the surrounding environment of the master vehicle, wherein the processing includes at least: feature extraction, synchronization, normalization and generation of probabilistic path prediction; The third control logic is used to detect the vehicle operator's intention; and A fourth control logic is used to generate control commands to the one or more actuators, wherein when the master vehicle is traveling in fully automatic and / or semi-automatic mode, the control commands selectively change at least one longitudinal speed of the master vehicle to approximate the historical preferences of the vehicle operator for lateral and longitudinal speeds, yaw, and forces observed by the vehicle operator.

2. The system according to claim 1, wherein, The first control logic further includes: Control logic for obtaining the position, motion, and acceleration information of the main vehicle in at least three degrees of freedom from the inertial measurement unit (IMU); Control logic for obtaining the main vehicle steering angle from one or more steering angle sensors (SAS); Control logic for obtaining the accelerator pedal position from the accelerator pedal position sensor; and Control logic used to obtain the brake pedal position from the brake pedal position sensor.

3. The system according to claim 1, wherein, The second control logic also includes: Control logic for filtering data from the one or more sensors and performing feature extraction. The filtering, by selecting only certain portions of the data from the one or more sensors that indicate changes in the master vehicle's state, causes the system and MPSPA application to reduce computational resource utilization from a first level to a second level below the first level, and wherein the feature extraction further determines that the changes in the master vehicle's state are sufficient to require additional processing.

4. The system according to claim 3, wherein, The second control logic also includes: Control logic for synchronizing the data from the one or more sensors, wherein synchronizing the data enables the data from the one or more sensors to be aligned and correlated in a timely and accurate manner, thereby enabling the data from each of the one or more sensors to be correlated in a timely manner with the data from each of the other sensors in the one or more sensors.

5. The system according to claim 4, wherein, The second control logic also includes: Control logic for generating probabilistic path predictions using a probability function of each state variable that follows a sigmoid function, where: The main vehicle's state variables include: the main vehicle's speed v x Main vehicle steering angle δ, torsion bar torque τ, and torsion bar torque ratio Where x is a variable, v x ,δ,τ, M represents the type of operation action, including turning, changing lanes, making a U-turn, evasive maneuvering, and / or lane following, and the probability of the operation action is obtained by multiplying the state probability functions, as shown below: Among them, P M,x Let be the probability function of variable x for operation action M; k is the sampling time, β M,x For turning parameters, the S-curve steepness of variable x in the operation action M; α M,x For turning parameters, the S-shaped midpoint of variable x in operation action M; n is the number or quantity of variables; and g d Let g be the direction of the operation action M, such that for the operation action on the right, g d =-1, for the left-hand operation, g d =1.

6. The system according to claim 5, further comprising: Control logic is used to integrate environmental knowledge into environmental knowledge categories, including road information, road type, number of lanes, presence / absence of construction areas, surrounding object information, nearest vehicle path CIPV turn signals, and navigation information. The effect of each environmental knowledge category is represented by a gain, and the product of the gains of all environmental knowledge categories defines the final environmental gain, expressed as: g m,e =g M,r ×g M,s ×g M,n Equation 3 Among them, g M,r The probability gain of road information for operation action M; g M,s The probability gain of objects surrounding the action M; g M,n The probability gain of navigation information for operation action M; and g M,e Let M be the probability gain of environmental knowledge for the action, and let the following equation represent the probability function of the action considering environmental knowledge:

7. The system according to claim 6, wherein, The second control logic also includes: Control logic for normalizing the sensor data, wherein normalizing the sensor data reduces redundant information in the sensor data from a first redundancy level to a second redundancy level lower than the first redundancy level; and Control logic is used to generate confidence calculations, wherein the output of the confidence calculations defines the probability that the master vehicle is performing a specific type of operation action M.

8. The system according to claim 5, wherein, The third control logic also includes: This is used to enhance the control logic of Adaptive Cruise Control (ACC) by leveraging a data-driven classification machine learning (ML) model combined with a shadow planner to predict the driver's expected path and adjust the speed curve of the master vehicle to match the driver's intentions, based on analysis of the driver's input to the master vehicle control system and the behavior of surrounding vehicles. Control logic for performing real-time path classification to assess the vehicle operator of the master vehicle and the type of maneuver the ADAS is actively pursuing; and Control logic for performing enumeration calculations, which limit the real-time path of the classification to at least one of left turns and right turns.

9. The system according to claim 6, wherein, The third control logic also includes: Control logic for generating trajectory predictions based on real-time paths, GPS data, and navigation information, including the selected navigation route.

10. The system according to claim 9, wherein, The fourth control logic also includes: Control logic for generating control commands to the one or more actuators, wherein the control commands include predicted lateral commands and predicted longitudinal commands; and Control logic for activating and controlling one or more actuators of the main vehicle drivetrain control system, engine control system, and braking control system to at least change the longitudinal speed of the main vehicle when a real-time path operation action M of the current classification is being executed according to the predicted lateral command and the predicted longitudinal command.

Images