Computer program, method and system

The system optimizes vehicle behavior by generating and transmitting V2X data to ensure no collisions and satisfy individual preferences, addressing the challenge of coordinating diverse vehicle systems for safe and efficient operation.

JP7589462B2Active Publication Date: 2024-11-26TOYOTA JIDOSHA KK
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
JP2020133731
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2019-08-08
Filing Date
2020-08-06
Publication Date
2024-11-26
Estimated Expiration
2040-08-06

AI Technical Summary

Technical Problem

Existing vehicle systems struggle to adjust behavior optimally to reduce collisions among autonomous, semi-autonomous, and human-driven vehicles with varying destinations and driving preferences, leading to increased collision risks.

Method used

A computer program product that generates and transmits V2X data to determine individual optimal vehicle behavior, ensuring no collisions and satisfying customization needs through vehicle control system modifications based on received behavior data.

Benefits of technology

Achieves overall optimal vehicle behavior in a region by preventing collisions and fulfilling individual vehicle preferences without interfering with others, using machine learning systems and clients to coordinate vehicle operations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007589462000001
    Figure 0007589462000001
  • Figure 0007589462000002
    Figure 0007589462000002
  • Figure 0007589462000003
    Figure 0007589462000003
Patent Text Reader

Abstract

To provide optimum vehicle behaviors in a region.SOLUTION: In some embodiments, a method for a connected vehicle includes transmitting, via Vehicle-to-Everything (V2X) communication, V2X data that includes customized data describing customized need of the connected vehicle. The method includes receiving, via the V2X communication, vehicle behavior data describing an individual optimum behavior for the connected vehicle that is determined at least partly on the basis of the V2X data. The method includes modifying an operation of a vehicle control system of the connected vehicle on the basis of the vehicle behavior data so that the connected vehicle implements the individual optimum behavior. Implementation of the individual optimum behavior by the connected vehicle contributes to achievement of an overall optimum behavior of a region where the connected vehicle is located, while the customized need of the connected vehicle is also satisfied.SELECTED DRAWING: Figure 3
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure is directed to providing optimal vehicle behavior within a region. [Background technology]

[0002] There may be various types of vehicles on the roadway, for example, autonomous vehicles, semi-autonomous vehicles, or human-driven vehicles. Because the vehicles may have different destinations, driving plans, and other driving preferences, coordinating the behavior of these vehicles to reduce the risk of collisions on the roadway may be difficult. Summary of the Invention

[0003] One typical aspect of the embodiments described herein includes a computer program product including a non-transitory memory of an on-board computer system of a connected vehicle that stores computer executable code that, when executed by a processor, causes the processor to generate vehicle-to-everything (V2X) data including customization data describing customization needs of a connected vehicle, transmit via V2X communication a feedback message including the V2X data, receive via the V2X communication a modification message including vehicle behavior data describing an individual optimal behavior of the connected vehicle determined at least in part based on the V2X data, and modify operation of a vehicle control system of the connected vehicle based on the vehicle behavior data such that the connected vehicle performs the individual optimal behavior, wherein performance of the individual optimal behavior by the connected vehicle contributes to achieving an overall optimal behavior of an area in which the connected vehicle is located while also satisfying the customization needs of the connected vehicle. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the respective methods.

[0004] Implementations may include one or more of the following features: A computer program product, wherein performance of the individual optimal behavior by the connected vehicle indicates one or more of: no collision occurs between the connected vehicle and one or more other vehicles in the region, and customization needs of the connected vehicle are satisfied without interfering with one or more customization needs of the one or more other vehicles. A computer program product, wherein the connected vehicle and the one or more other vehicles are included in a group of vehicles in the region, and performance of a global optimal behavior of the region indicates one or more of: no collision occurs within the group of vehicles, and customization needs of each vehicle in the group are satisfied without modifying customization needs of remaining vehicles in the group. A computer program product, wherein customization needs of the connected vehicle are described by one or more of a driving intent of the connected vehicle and a first weighting of the driving intent, and a driving preference of the connected vehicle and a second weighting of the driving preference. Implementations of the techniques described above may include hardware, a method or process, or computer software on a computer-accessible medium.

[0005] One typical aspect includes a method for a connected vehicle including transmitting, via V2X communications, V2X data including customization data describing customization needs including the V2X data, receiving, via the V2X communications, vehicle behavior data describing an individual optimal behavior of the connected vehicle determined based at least in part on the V2X data, and modifying operation of a vehicle control system of the connected vehicle based on the vehicle behavior data such that the connected vehicle performs the individual optimal behavior, wherein performance of the individual optimal behavior by the connected vehicle contributes to achieving an overall optimal behavior of an area in which the connected vehicle is located while also satisfying the customization needs of the connected vehicle. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the respective method.

[0006] Implementations may include one or more of the following features: A method, wherein performance of the individual optimal behavior by the connected vehicle indicates one or more of: no collision occurs between the connected vehicle and one or more other vehicles in the region, and customization needs of the connected vehicle are satisfied without interfering with one or more customization needs of the one or more other vehicles; A method, wherein the connected vehicle and the one or more other vehicles are included in a group of vehicles in the region, and performance of a global optimal behavior for the region indicates one or more of: no collision occurs within the group of vehicles, and customization needs of each vehicle in the group are satisfied without modifying customization needs of remaining vehicles in the group; A method, wherein the customization needs of the connected vehicle are described by one or more customization parameters and one or more weights for the one or more customization parameters; A method, wherein the one or more customization parameters include one or more of a driving intent and a driving preference associated with the connected vehicle. The method further comprising generating V2X data comprising one or more of parameter data describing the one or more customization parameters, weight data describing the one or more weights, sensor data of the connected vehicle, and advanced driver assistance system (ADAS) data of the connected vehicle. The method further comprising predicting a future behavior of the connected vehicle, the V2X data further comprising prediction data describing the future behavior of the connected vehicle. The method, wherein the individual optimal behavior comprises a constraint trajectory, acceleration settings, steering angle settings, and speed settings of the connected vehicle optimized for the connected vehicle. The method, wherein the individual optimal behavior satisfies one or more of traffic regulation requirements and safety requirements in the domain. Implementations of the above techniques may include hardware, methods or processes, or computer software on a computer-accessible medium.

[0007] One typical aspect includes a system including an on-board computer system of a connected vehicle including a non-transitory memory that, when executed by the on-board computer system, causes the on-board computer system to transmit V2X data via V2X communications including customization data describing customization needs including the V2X data, receive via the V2X communications vehicle behavior data describing an individual optimal behavior of the connected vehicle determined based at least in part on the V2X data, and modify operation of a vehicle control system of the connected vehicle based on the vehicle behavior data such that the connected vehicle performs the individual optimal behavior, and performance of the individual optimal behavior by the connected vehicle contributes to achieving an overall optimal behavior of an area in which the connected vehicle is located while also satisfying the customization needs of the connected vehicle. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of a respective method.

[0008] Implementations may include one or more of the following features: A system in which performance of the individual optimal behavior by the connected vehicle indicates one or more of: no collision occurs between the connected vehicle and one or more other vehicles in the region and customization needs of the connected vehicle are satisfied without interfering with one or more customization needs of the one or more other vehicles; A system in which the connected vehicle and the one or more other vehicles are included in a group of vehicles in the region and performance of an overall optimal behavior of the region indicates one or more of: no collision occurs within the group of vehicles and customization needs of each vehicle in the group are satisfied without modifying customization needs of the remaining vehicles in the group; A system in which the customization needs of the connected vehicle are described by one or more customization parameters and one or more weights for the one or more customization parameters; A system in which the one or more customization parameters include one or more of a driving intent and a driving preference associated with the connected vehicle. A system, wherein the computer code, when executed by the on-board computer system, further causes the on-board computer system to generate V2X data including one or more of parameter data describing the one or more customization parameters of the connected vehicle, weight data describing the one or more weights, sensor data, and ADAS data of the connected vehicle. A system, wherein the computer code, when executed by the on-board computer system, further causes the on-board computer system to predict future behavior of the connected vehicle, the V2X data further including prediction data describing the future behavior of the connected vehicle. Implementations of the above techniques may include hardware, methods or processes, or computer software on a computer-accessible medium. [Brief description of the drawings]

[0009] The present disclosure is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings in which like reference symbols refer to similar elements and in which:

[0010] [Figure 1A] FIG. 1 is a block diagram illustrating an operating environment for a machine learning system and a machine learning client according to some embodiments. [Figure 1B] FIG. 1 is a block diagram illustrating an architecture including a machine learning system and various instances of machine learning clients onboard various vehicles according to some embodiments. [Figure 1C] FIG. 1 is a block diagram illustrating an architecture including components of a machine learning system and one or more connected vehicles according to some embodiments. [Figure 1D] FIG. 1 is a block diagram illustrating another architecture including components of a machine learning system and one or more connected vehicles according to some embodiments. [Diagram 2] FIG. 1 is a block diagram illustrating an example computer system including a machine learning client in accordance with some embodiments. [Diagram 3] FIG. 1 illustrates a method for achieving individual optimal behavior of a connected vehicle according to some embodiments. [Figure 4] FIG. 1 illustrates a method for providing a global optimal behavior within a region and individual optimal behaviors of vehicles within the region according to some embodiments. [Figure 5A] 1 is a graphical representation illustrating an example operating scenario for providing optimal vehicle behavior according to some embodiments. [Figure 5B] 1 is a graphical representation illustrating an example operating scenario for providing optimal vehicle behavior according to some embodiments. [Figure 6A] 1 is a graphical representation illustrating another example operating scenario for providing optimal vehicle behavior according to some embodiments. [Figure 6B] 1 is a graphical representation illustrating another example operating scenario for providing optimal vehicle behavior according to some embodiments. [Figure 7A]1 is a graphical representation illustrating an exemplary simulation for providing optimal vehicle behavior according to some embodiments. [Figure 7B] 1 is a graphical representation illustrating an exemplary simulation for providing optimal vehicle behavior according to some embodiments. [Figure 7C] 1 is a graphical representation illustrating another exemplary simulation for providing optimal vehicle behavior according to some embodiments. [Figure 7D] 1 is a graphical representation illustrating another exemplary simulation for providing optimal vehicle behavior according to some embodiments. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] In some areas or situations, it may be difficult to ensure the satisfaction of basic driving requirements of a vehicle. Examples of basic driving requirements include, for example, traffic regulation requirements and safety requirements. For example, it may not be easy to ensure optimal operation of the ADAS system of the vehicle so that the vehicle is provided with an optimal driving trajectory and a vehicle collision does not occur. In another example, in the case of a human-driven vehicle, the driver of the vehicle may need to (1) know the risk of collision and (2) decide how to drive to avoid the possibility of a collision.

[0012] Different drivers may have different driving intentions, driving preferences, and other driving habits. Some drivers may prefer to drive smoothly and fuel-efficiently, while other drivers may prefer to drive at high speeds to reduce travel time or increase driving enjoyment. Currently, each driver determines his / her driving behavior by himself / herself. This is problematic because allowing drivers to select their driving behavior by themselves may increase the risk of collision with other vehicles or may cause other drivers to give up their driving preferences or intentions.

[0013] Described herein are machine learning systems and machine learning clients that cooperate with each other to achieve optimal vehicle behavior within a domain such that the problem described in the above paragraph is solved. The optimal vehicle behavior includes, for example, a global optimal behavior of the domain and an individual optimal behavior of the vehicles present within the domain. For example, the machine learning system and the machine learning client cooperate with each other to reconcile different driving preferences and intentions of different vehicles. The machine learning system and the machine learning client cooperate with each other to ensure that no collisions occur between the vehicles, and maximize the ability of each vehicle to achieve its own preferences and intentions, as much as possible.

[0014] Exemplary improvements and advantages provided by the machine learning systems and machine learning clients described herein are described below: For example, the machine learning systems and machine learning clients do not allow any particular vehicle to override its own preferences or intent over the preferences or intent of other vehicles in a particular area and the safe operation of those other vehicles.

[0015] In another example, the implementation of the machine learning system and the machine learning client does not require frequent communication between associated endpoints. Thus, the machine learning system and the machine learning client can operate in real time on a vehicle moving at roadway speeds, and as a result, both can be used on vehicles in the real world. In yet another example, the machine learning system and the machine learning client can be applied to manually operated vehicles or non-autonomous vehicles.

[0016] In yet another example, the relative weighting of driving preferences and intent is considered when determining how to modify the behavior of different vehicles. In contrast, existing solutions do not consider the weighting of driving preferences or intent of any particular vehicle, much less the relative weighting of these parameters of one vehicle versus other vehicles. Other exemplary improvements and advantages are possible, but are not limited herein.

[0017] An exemplary overview of a machine learning system and a machine learning client is described below. Assume that the vehicle environment includes (1) an ego vehicle and (2) a set of remote vehicles. Alternatively, or in addition, the roadway environment includes a processing device (e.g., a cloud server, an edge server, a roadside unit, etc.). The ego vehicle is a connected vehicle that includes an instance of a machine learning system. Optionally, the machine learning system is mounted within the processing device. The ego vehicle and the remote vehicles may or may not be autonomous vehicles. The remote vehicles may or may not be connected vehicles. The remote vehicles may or may not include an instance of a machine learning client.

[0018] The machine learning client includes software stored in at least one electronic control unit (ECU) of the remote vehicle. The machine learning client is operable to generate V2X data using on-board sensors and ADAS systems of the remote vehicle and provide feedback messages including the V2X data to the machine learning system via V2X transmission. The V2X data may include customization data describing customization needs of the remote vehicle, where the customization needs of the vehicle (or driver) may include driving intent, driving preferences, and any other needs of the vehicle (or driver).

[0019] The machine learning system collects V2X data received from various vehicles in the region, each of which includes an instance of a machine learning client. Thus, the machine learning system receives digital data describing the driving intent and preferences of the various vehicles in the region. The machine learning system determines region behavior data based at least in part on the V2X data. The region behavior data includes digital data describing an overall optimal behavior of all vehicles contained within the region. For example, the region behavior data describes how these vehicles may interact with each other and with other objects.

[0020] We describe how, collectively and individually, the global optimal behavior of the vehicles in a region ensures that collisions with other vehicles are avoided and allows drivers to satisfy their own customization needs without modifying the customization needs of other drivers.

[0021] The machine learning system analyzes the V2X data to determine each vehicle's driving intentions and preferences and the weighting of those driving intentions and preferences relative to other vehicles. The machine learning system also determines an individual optimal behavior for each vehicle in the domain. Each vehicle's individual optimal behavior contributes to achieving a global optimal behavior for the domain, while also satisfying the customization needs of each vehicle. The machine learning system performs the necessary operations to coordinate each vehicle's individual optimal behavior with the vehicle control system (e.g., ADAS system) of that vehicle such that the global optimal behavior of the domain is achieved.

[0022] The machine learning system transmits vehicle behavior data describing the individual optimal behavior of each vehicle to the corresponding vehicle, and the vehicle control system of the corresponding vehicle then controls the corresponding vehicle to operate in accordance with the individual optimal behavior described by the vehicle behavior data, thus achieving a global optimal behavior of all vehicles in the domain and satisfying the customization needs of all vehicles.

[0023] As described herein, examples of V2X communications include Dedicated Short Range Communications (DSRC) (including Basic Safety Messages (BSM) and Personal Safety Messages (PSM), among other types of DSRC communications). Other examples of V2X communications include Long Term Evolution (LTE), millimeter wave (mmWave) communications, 3G, 4G, 5G, LTE-V2X, 5G-V2X, LTE vehicle-to-vehicle (LTE-V2V), LTE device-to-device (LTE-D2D), or Voice over LTE (VoLTE), etc. In some examples, V2X communications may include V2V communications, vehicle-to-infrastructure (V2I) communications, vehicle-to-network (V2N) communications, or any combination thereof.

[0024] Examples of over-the-air messages (e.g., V2X messages) described herein include, but are not limited to, DSRC messages, BSM messages, and LTE messages. Other examples of over-the-air messages include one or more of LTE-V2X messages (e.g., LTE-V2V messages, LTE-V2I messages, LTE-V2N messages, etc.), 5G-V2X messages, mmWave messages, etc.

[0025] Illustrative Overview 1A, an operating environment 100 of a machine learning client 199 and a machine learning system 191 is shown according to some embodiments. The operating environment 100 may include one or more of the following elements: an ego vehicle 110, and one or more remote vehicles 112 (e.g., a first remote vehicle 112A, ..., an Nth remote vehicle 112N). Optionally, the operating environment 100 may further include a processing device 140. These elements of the operating environment 100 may be communicatively connected to a network 105. In practice, the operating environment 100 may include any number of ego vehicles 110, remote vehicles 112, processing devices 140, and networks 105.

[0026] Network 105 may be of a conventional type, wired or wireless type, and may have a number of different configurations, including star configuration, token ring configuration, or other configurations. Additionally, network 105 may include a local area network (LAN), a wide area network (WAN) (e.g., the Internet), or other interconnected data paths across which multiple devices and / or entities may communicate. In some embodiments, network 105 may include a peer-to-peer network. Network 105 may be connected to or include portions of a telecommunications network that transmit data in a variety of different communication protocols. In some embodiments, network 105 includes a Bluetooth® communication network or a cellular communication network that transmits and receives data, including by way of Short Messaging Service (SMS) and Multimedia Messaging Service (MMS). In some embodiments, network 105 further includes networks for Hypertext Transfer Protocol (HTTP), direct data connection, Wireless Application Protocol (WAP), email, DSRC, full-duplex radio communication, and mmWave. In some embodiments, network 105 further includes networks for WiFi (infrastructure mode), WiFi (ad-hoc mode), visible light communication, TV white space communication, and satellite communication. Network 105 may also include a mobile data network or combination of mobile data networks, which may include 3G, 4G, LTE, LTE-V2X, LTE-D2D, VoLTE, 5G-V2X, or any other mobile data network. Additionally, network 105 may include one or more IEEE 802.11 wireless networks.

[0027] Ego vehicle 110 may be any type of vehicle. For example, ego vehicle 110 may include one or more of the following types of vehicles: car, truck, sport utility vehicle, bus, semi-truck, drone, or any other road-based vehicle. Ego vehicle 110 may be a connected vehicle that includes a communication unit and can communicate with other endpoints connected to network 105.

[0028] In some embodiments, ego vehicle 110 is a DSRC-enabled vehicle that includes a DSRC radio and a DSRC-compliant Global Positioning System (GPS) unit. Ego vehicle 110 may also include other V2X radios in addition to the DSRC radio. DSRC is not a requirement of the embodiments described herein, and any type of V2X communication may be performed.

[0029] Ego vehicle 110 may include one or more of the following elements: processor 125A, memory 127A, communication unit 145A, GPS unit 150, ECU 152, sensor set 154, vehicle control system 156, and machine learning system 191A. In some embodiments, ego vehicle 110 may also include an instance of a machine learning client 199 (e.g., machine learning client 199A). These elements of ego vehicle 110 may be connected to each other via a bus.

[0030] In some embodiments, processor 125A and memory 127A may be elements of an on-board computer system (such as computer system 200 described below with reference to FIG. 2). The on-board computer system may be operable to launch or control the operation of machine learning system 191A (or machine learning client 199A). For example, the on-board computer system may be operable to access and execute data stored in memory 127A to provide functionality described herein with respect to machine learning client 199A or elements thereof (e.g., see FIG. 2).

[0031] Processor 125A includes an arithmetic logic unit, a general purpose controller, or some other processor array that performs calculations and provides electronic display signals to a display device. Processor 125A processes data signals and may include a variety of computing architectures. Examples of computing architectures include Complex Instruction Set Computer (CISC) architectures, Reduced Instruction Set Computer (RISC) architectures, or architectures that implement a combination of instruction sets. Ego-vehicle 110 may include one or more processors 125A. Other processors, operating systems, sensors, displays, and physical configurations are possible.

[0032] The memory 127A stores instructions or data executable by the processor 125A. The instructions or data may include code for performing the techniques described herein. The memory 127A may be a dynamic random access memory (DRAM), a static random access memory (SRAM), a flash memory, or some other memory device. In some embodiments, the memory 127A also includes non-volatile memory or similar persistent storage devices and media. Examples of persistent storage devices include hard disk drives, floppy disk drives, CD-ROM devices, DVD-ROM devices, DVD-RAM devices, DVD-RW devices, and flash memory devices. Examples of additional persistent storage devices may include any other mass storage device for storing information more persistently. The ego-vehicle 110 may include one or more memories 127A.

[0033] Memory 127A may store one or more of the following elements: V2X data 128A, sensor and ADAS data 129, area behavior data 130, intent data 131, preference data 132, vehicle behavior data 133, and vehicle specification data 134.

[0034] In some embodiments, V2X data 128A may include digital data related to the ego-vehicle 110 itself. For example, V2X data 128A may include customization data describing customization needs of the ego-vehicle 110. The customization needs may be described by one or more customization parameters and one or more weights of the customization parameters. The one or more customization parameters may include, for example, driving intentions described by intent data 131 and driving preferences described by preference data 132. For example, V2X data 128A includes customization data describing (1) driving intentions or preferences of a driver of the ego-vehicle 110 and (2) weights of the driving intentions or preferences describing how strong the driving intentions or preferences are.

[0035] In some embodiments, the V2X data 128A further includes sensor and ADAS data 129 generated by the ego vehicle 110. In some embodiments, the V2X data 128A describes some or all of the design specifications of the ego vehicle 110. The machine learning client 199A of the ego vehicle 110 can transmit the V2X data 128A related to the ego vehicle 110 to another end point (e.g., the processing device 140 or the remote vehicle 112) via a feedback message. The feedback message may be a wireless message.

[0036] In some embodiments, the V2X data 128A may include a collective set of V2X data received from a set of remote vehicles 112 within the region in which the ego vehicle 110 resides. For example, each instance of V2X data received from a particular remote vehicle 112 may include customization data describing the customization needs of the remote vehicle (e.g., driving intent, driving preferences, or a combination thereof). An instance of V2X data received from a particular remote vehicle 112 may also include sensor and ADAS data generated by the remote vehicle 112 and data describing some or all of the design specifications of the remote vehicle 112. By collecting the V2X data received from the remote vehicles 112, the customization needs, design specifications, and sensor and ADAS data of the remote vehicles 112 are collected for operation of the machine learning system 191A.

[0037] Sensor and ADAS data 129 may include sensor data and ADAS data. Sensor data includes digital data describing one or more sensor measurements recorded by sensors in sensor set 154. The sensor data may be input to one or more ADAS systems of host vehicle 110 to enable the ADAS systems to provide their functionality. For example, the one or more ADAS systems may perceive the environment of host vehicle 110 and determine the vehicle's response to the environment. ADAS data includes digital data describing the ADAS system's analysis and the vehicle's response.

[0038] In some embodiments, sensor and ADAS data 129 includes digital data describing the behavior of a driver of host vehicle 110 and the context of the behavior. For example, the context of the behavior may include events occurring before, during, or possibly after the behavior. These events may include the behavior of other drivers, the time of day, the day of the week, the weather, whether the driving conditions are urban or rural, etc. Sensor and ADAS data 129 may also include a unique identification of host vehicle 110 (e.g., vehicle identification number (VIN)).

[0039] The sensor and ADAS data 129 may also include behaviors of the driver of the remote vehicle 112 and the context of those behaviors recorded by the sensor set 154 of the host vehicle 110. The sensor and ADAS data 129 may also include a unique identification of the remote vehicle 112. For example, the sensor and ADAS data 129 may include license plate information of the remote vehicle 112, such as the vehicle registration number and the state, province, commonwealth, or other jurisdiction that issues the license plate.

[0040] In some embodiments, the machine learning client 199A can cause the sensor set 154 and ADAS system of the ego vehicle 110 to record sensor and ADAS data 129. In some embodiments, the sensor and ADAS data 129 can include digital data describing the behavior of surrounding vehicles that do not include the machine learning client. The sensor and ADAS data 129 can be included in the V2X data and transmitted to the machine learning system 191 via feedback messages. In this way, the machine learning system 191 receives digital data that can be used to predict future behavior of the vehicles in generating vehicle behavior data for these vehicles, thereby explaining the behavior of these vehicles and avoiding collisions.

[0041] Region behavior data 130 may include digital data describing the overall optimal behavior of vehicles present in the region in which ego-vehicle 110 is located. For example, region behavior data 130 describes how these vehicles can interact with each other and with other objects in the environment.

[0042] The intent data 131 may include digital data describing the driving intent of the host vehicle 110. In some embodiments, the driving intent includes the driver's planned trajectory (e.g., current position, future position, speed, acceleration, steering angle, etc.).

[0043] Preference data 132 may include digital data describing driving preferences for host vehicle 110. Driving preferences may include, for example, a driver's preferences for prioritizing fuel economy, reaching a destination as quickly as possible, or a pleasant experience.

[0044] The vehicle behavior data 133 may include digital data describing an individual optimum behavior of the host vehicle 110. The individual optimum behavior is described in more detail below.

[0045] The vehicle specification data 134 may include digital data describing one or more design specifications of the host vehicle 110. For example, the one or more design specifications may include digital data describing the overall length, overall width, wheelbase, maximum acceleration, maximum steering angle, total weight, and maneuverability of the host vehicle 110. The controllability of the host vehicle 110 indicates whether the host vehicle 110 is controllable by the vehicle control system 156. If the host vehicle 110 is a controllable vehicle, the individual optimal behavior generated by the machine learning system 191 can be automatically performed by the host vehicle 110. If the host vehicle 110 is an uncontrollable vehicle, a human driver can perform the individual optimal behavior according to instructions provided via the user dialogue interface 161 (see FIG. 1B).

[0046] The communication unit 145A transmits and receives data to and from the network 105 or another communication channel. In some embodiments, the communication unit 145A may include a DSRC transceiver, a DSRC receiver, and other hardware or software necessary to make the vehicle 110 a DSRC-enabled device. For example, the communication unit 145A includes a DSRC antenna configured to broadcast DSRC messages over the network. The DSRC antenna can also transmit BSM messages at a user-configurable fixed or variable interval (e.g., every 0.1 seconds, at a time interval corresponding to a frequency range of 1.6 Hz to 10 Hz, etc.).

[0047] In some embodiments, the communication unit 145A may include a port for direct physical connection to the network 105 or another communication channel. For example, the communication unit 145A includes a USB, SD, CAT-5, or similar port for wireless communication to the network 105. In some embodiments, the communication unit 145A includes a wireless transceiver for exchanging data using one or more wireless communication methods with the network 105 or other communication channels. Examples of wireless communication methods may include one or more of IEEE 802.11, IEEE 802.16, and BLUETOOTH. Examples of wireless communication methods may further include EN ISO 14906:2004 Electronic Toll Collection - Application Interface EN 11253:2004 DSRC - Physical Layer using 5.8 GHz microwave (Review). Examples of wireless communication methods may further include EN 12795:2002 DSRC - DSRC Data Link Layer: Media Access and Logical Link Control (Review). Example wireless communication methods may further include EN 12834:2002 DSRC-Application Layer (Review) and EN 13372:2004 DSRC-DSRC Profile for RTTT Applications (Review). Example wireless communication methods may further include the communication methods described in U.S. patent application Ser. No. 14 / 471,387, filed Aug. 28, 2014, and entitled "Full-Duplex Coordination System," or another suitable wireless communication method.

[0048] In some embodiments, the communication unit 145A includes a cellular communication transceiver for transmitting and receiving data over a cellular communication network. For example, data may be transmitted or received via Short Messaging Service (SMS), Multimedia Messaging Service (MMS), Hypertext Transfer Protocol (HTTP), direct data connection, WAP, email, or another suitable type of electronic communication. In some embodiments, the communication unit 145A includes a wired port and a wireless transceiver. The communication unit 145A also provides other conventional connections to the network 105 for distributing files or media objects using standard network protocols including TCP / IP, HTTP, HTTPS, and SMTP, mmWave, DSRC, and the like.

[0049] The communication unit 145A may include a V2X radio 149. The V2X radio 149 may include hardware elements including a DSRC transmitter operable to transmit DSRC messages in the 5.9 GHz band. The 5.9 GHz band is dedicated to DSRC messages. The hardware elements may also include a DSRC receiver operable to receive DSRC messages in the 5.9 GHz band.

[0050] In some embodiments, GPS unit 150 is a conventional GPS unit of host vehicle 110. For example, GPS unit 150 may include hardware for wirelessly communicating with GPS satellites to obtain data describing the geographic location of host vehicle 110. In some embodiments, GPS unit 150 is a DSRC-compliant GPS unit of host vehicle 110. The DSRC-compliant GPS unit is operable to provide GPS data describing the geographic position of host vehicle 110 with lane-level accuracy.

[0051] The ECU 152 may include one or more processors and one or more memories. The ECU 152 may control the operation of the vehicle control system 156, the sensor set 154, and the machine learning client 199A (or the machine learning system 191A) of the ego-vehicle 110. In some embodiments, the machine learning client 199A (or the machine learning system 191A) of the ego-vehicle 110 is mounted within the ECU 152.

[0052] Sensor set 154 includes one or more sensors operable to measure a roadway environment exterior to host vehicle 110. For example, sensor set 154 may include one or more sensors that record one or more physical characteristics of a roadway environment proximate to host vehicle 110. Memory 127 may store sensor data describing the one or more physical characteristics recorded by sensor set 154.

[0053] The sensor set 154 may also include various sensors that record the environment inside the vehicle cabin of the host vehicle 110. For example, the sensor set 154 may include on-board sensors that monitor the environment, whether internal or external, of the host vehicle 110. In another example, the sensor set 154 may include sensors that observe the driver's behavior, such as cameras, LIDAR, radar, infrared sensors, and built-in cameras, biometric identification sensors, etc.

[0054] In some embodiments, sensor set 154 may include one or more of the following vehicle sensors: a camera, a LIDAR sensor, a radar sensor, a laser altimeter, an infrared detector, a motion detector, a thermostat, and an acoustic detector. Sensor set 154 may also include one or more of the following sensors: a carbon monoxide sensor, a carbon dioxide sensor, an oxygen sensor, a mass airflow sensor, and an engine coolant temperature sensor. Sensor set 154 may also include one or more of the following sensors: a throttle position sensor, a crankshaft position sensor, an automobile engine sensor, a valve timer, an air-fuel ratio gauge, and a blind spot gauge. Sensor set 154 may also include one or more of the following sensors: a curb feeler, a fault detector, a Hall effect sensor, a manifold absolute pressure sensor, a parking sensor, a radar gun, a speedometer, and a speed sensor. Sensor set 154 may also include one or more of the following sensors: tire pressure monitoring sensors, torque sensors, transmission fluid temperature sensors, and turbine speed sensors (TSS), variable reluctance sensors, and vehicle speed sensors (VSS). Sensor set 154 may also include one or more of the following sensors: water sensors, wheel speed sensors, and any other type of automotive sensor.

[0055] Vehicle control system 156 may control the operation of host vehicle 110. For example, vehicle control system 156 may provide all or a portion of the autonomous functionality to host vehicle 110. In some embodiments, vehicle control system 156 may include one or more ADAS systems, autonomous driving systems, or a combination thereof.

[0056] Examples of ADAS systems included in the host vehicle 110 include one or more of an automatic cruise control (ACC) system, an adaptive high beam system, an adaptive lighting control system, and an automatic parking system. Other examples of ADAS systems include an automotive night vision system, a blind spot monitor, a collision avoidance system, a crosswind stabilization system, a driver drowsiness detection system, and a driver monitoring system. Other examples of ADAS systems include an emergency driver assistance system, a forward collision warning system, an intersection assistance system, an intelligent speed adaptation system, a lane departure warning system (also called a lane keeping assistant). Other examples of ADAS systems include a pedestrian protection system, a traffic sign recognition system, a turning assistant, and a wrong-way driving warning system, etc. The features and functions provided by these exemplary ADAS systems are also referred to herein as "autonomous features" or "autonomous functions", respectively. In fact, an in-vehicle system includes the features of any vehicle that has the capability to monitor and track driving data and route data, and is not limited to an ADAS system.

[0057] The remote vehicles 112 present on the roadway of the ego vehicle 110 may have a similar structure to the ego vehicle 110, and a similar description will not be repeated here. The operating environment 100 may include multiple remote vehicles 112 (e.g., a first remote vehicle 112A, ..., an Nth remote vehicle 112N). In some embodiments, the first remote vehicle 112A may include an instance of a machine learning client (e.g., a machine learning client 199B) and a communication unit 145B. The remote vehicle 112A may store V2X data 128B in its local storage device. The remote vehicle 112N may have a similar structure to the remote vehicle 112A, and therefore a similar description will not be repeated here.

[0058] V2X data 128A and V2X data 128B have similar content and may be referred to herein individually or collectively as “V2X data 128.” A similar description of V2X data 128B will not be repeated here.

[0059] Machine learning clients 199A and 199B may have similar structures and provide similar silks, and may be referred to herein individually or collectively as "machine learning clients 199."

[0060] In some embodiments, the machine learning client 199 includes software operable, when executed by the processor 125, to cause the processor 125 to perform one or more steps of a method 300, described below with reference to FIG. 3. In some embodiments, the machine learning client 199 can be implemented using hardware, including a field programmable gate array ("FPGA") or an application specific integrated circuit ("ASIC"). In some other embodiments, the machine learning client 199 can be implemented using a combination of hardware and software. The machine learning client 199 can be stored on a combination of devices (e.g., a server or other device) or on one of the devices.

[0061] The machine learning client 199 is described in more detail below with reference to Figures 1B, 2-3 and 5A-7D.

[0062] The processing device 140 may be one of a cloud server, an edge server, a roadside unit, or any other processing device. The processing device 140 includes a machine learning system (e.g., the machine learning system 191B), a processor 125B, a memory 127B, a communication unit 145C, and any other suitable components.

[0063] Processors 125A and 125B may have similar structures and provide similar functionality and may be referred to herein, individually or collectively, as "processor 125." Memories 127A and 127B may have similar structures and provide similar functionality and may be referred to herein, individually or collectively, as "memory 127." Communications units 145A, 145B, and 145C may have similar structures and provide similar functionality and may be referred to herein, individually or collectively, as "communications unit 145." Similar descriptions of processor 125B, memory 127B, and communications units 145B and 145C will not be repeated here.

[0064] Machine learning systems 191A and 191B may have similar structures and provide similar functionality, and may be referred to herein individually or collectively as "machine learning systems 191."

[0065] In some embodiments, the machine learning system 191 includes software operable, when executed by the processor 125, to cause the processor 125 to perform one or more steps of a method 400, described below with reference to FIG. 4. In some embodiments, the machine learning system 191 may be implemented using hardware, including an FPGA or an ASIC. In some other embodiments, the machine learning system 191 may be implemented using a combination of hardware and software. The machine learning system 191 may be stored on one of the devices or a combination of devices (e.g., a server or other device).

[0066] In some embodiments, machine learning system 191 may include one or more of predictor module 195 and negotiator module 197. Machine learning system 191, predictor module 195, and negotiator module 197 are described in more detail below with reference to Figures 1B-1D and 4-7D.

[0067] 1B, an architecture 160 is shown that includes a machine learning system 191 and various instances of a machine learning client 199 on board various connected vehicles 114 according to some embodiments. For example, two connected vehicles 114A and 114B are shown in FIG. 1B. Connected vehicle 114A includes a vehicle control system 156 and may be an autonomous vehicle or a semi-autonomous vehicle. For example, connected vehicle 114A may be controlled by one or more ADAS systems and driven by a system driver. Connected vehicle 114B may or may not include a vehicle control system 156. Connected vehicle 114B may be driven by a human driver.

[0068] Each connected vehicle 114 includes an instance of a machine learning client 199. Each connected vehicle 114 may be the ego vehicle 110 or a remote vehicle 112 shown in FIGURE 1A. For each connected vehicle 114, the machine learning client 199 causes its sensor set 154 and ADAS system to collect sensor data and ADAS data.

[0069] In some embodiments, the predictor module 195 resides within a machine learning client 199 of the connected vehicle 114. In this case, the machine learning client 199 stores data structures describing the past behavior of the connected vehicle 114. The predictor module 195 analyzes current sensor data and current ADAS data of the connected vehicle 114 to estimate the future behavior of the connected vehicle 114 by comparison with the past behavior of the connected vehicle 114. The machine learning client 199 may include learning algorithms that continually improve the ability of the predictor module 195 to predict the behavior of the connected vehicle 114.

[0070] In some embodiments, the predictor module 195 is implemented within the machine learning system 191. The operation of predicting the future behavior of the connected vehicle 114 is performed by the machine learning system 191 after receiving V2X data from the connected vehicle 114 and analyzing the V2X data (described in more detail below).

[0071] In other words, the prediction of the future behavior of the connected vehicle 114 is performed by the machine learning client 199 or the machine learning system 191, but not by both. However, in some embodiments, the future behavior of the connected vehicle 114 can be provided by a system operator (e.g., an ADAS system or an autonomous driving system, etc.) or a human driver of the connected vehicle 114. The future behavior of the connected vehicle 114 may include a planned trajectory of the connected vehicle 114 (e.g., current position, future position, speed, acceleration settings, steering angle settings, etc.). The future behavior of the connected vehicle 114 can be included as part of the driving intent of the connected vehicle 114.

[0072] The machine learning client 199 generates the V2X data 128 based on the sensor and ADAS data, and optionally generates other data, such as user input. The V2X data 128 describes the sensor and ADAS data, and optionally describes the future behavior of the connected vehicle 114. For example, the V2X data 128 includes customization data describing customization needs of the connected vehicle 114 (e.g., driving preferences and weights of the driving preferences, driving intent, and weights of the driving intent). In another example, the V2X data 128 further includes data describing the roadway environment surrounding the connected vehicle 114 and data describing the vehicle state of the connected vehicle 114. In yet another example, the V2X data 128 further includes data describing the design specifications of the connected vehicle 114.

[0073] The machine learning client 199 causes the communication unit 145 to transmit a feedback message over the V2X network, with the feedback message including the V2X data 128 as its payload. The machine learning system 191 then receives the feedback message from the connected vehicles 114. Thus, the machine learning system 191 receives multiple feedback messages, one for each connected vehicle 114. The machine learning system 191 parses the V2X data from each of the feedback messages.

[0074] In some embodiments, the machine learning system 191 stores a data structure that describes the past behavior of the connected vehicles 114. For example, the data structure describes the behavior of connected vehicles in general, rather than the behavior of a specific connected vehicle. The predictor module 195 of the machine learning system 191 analyzes the current sensor and ADAS data of a particular connected vehicle 114 to estimate the future behavior of the connected vehicle 114 by comparison with the past behavior of other connected vehicles 114. The predictor module 195 may include learning algorithms that continually improve the ability of the predictor module 195 to predict the behavior of connected vehicles based on the V2X data received from these connected vehicles 114.

[0075] The negotiator module 197 of the machine learning system 191 determines customization needs for each connected vehicle 114 based on the V2X data 128 received from that connected vehicle 114. For example, the negotiator module 197 determines one or more of driving preferences, weights of the driving preferences, driving intentions, weights of the driving intentions, etc. for each connected vehicle 114.

[0076] The negotiator module 197 determines the relative weighting of driving preferences and intentions for a group of connected vehicles 114 located within the region.

[0077] The negotiator module 197 determines region behavior data describing optimal overall behavior of the connected vehicles 114 located within the region. For example, the negotiator module 197 determines the region behavior data based on one or more determining factors. Examples of determining factors include V2X data of the connected vehicles 114 (including driving intent and driving preferences) and traffic or safety standards within the region. Another example of determining factors includes past crash data describing real-world incidents within the region, and preferences of vehicle manufacturers, roadway authorities, or design engineers. Another example of determining factors includes traffic regulation requirements within the region (e.g., speed limits, traffic signals, direction of travel, etc.), and safety requirements within the region (e.g., minimum distance and relative speed between adjacent vehicles on two levels). Another example of determining factors includes a digital simulation generated based at least in part on the other determining factors listed above.

[0078] The negotiator module 197 generates, for each connected vehicle 114 providing V2X data, vehicle behavior data 133 describing the vehicle's individual optimal behavior. The individual optimal behavior is optimized by the machine learning system 191 to maximize the vehicle's ability to satisfy customization needs (e.g., its driving preferences or intentions) as long as doing so is consistent with the domain's optimal overall behavior. For example, the individual optimal behavior of each connected vehicle 114 contributes to one form of achieving the domain's optimal overall behavior. Additionally, the individual optimal behavior of each connected vehicle does not interfere with the ability of other vehicles to satisfy their customization needs.

[0079] In some embodiments, the individual optimal behavior of each connected vehicle 114 is a safe behavior that satisfies one or more of the traffic regulation requirements and safety requirements in the domain while maximizing the fulfillment of the vehicle's customization needs. In some embodiments, the individual optimal behavior includes one or more of the constrained trajectory, acceleration setting, steering angle setting, and speed setting of the connected vehicle 114 that are optimized for the connected vehicle 114.

[0080] The negotiator module 197 generates modification messages that are sent to the connected vehicles 114. The modification messages are wireless messages and may each include vehicle behavior data 133 for said connected vehicles 114. Different instances of the vehicle behavior data 133 tend to be different from each other because different vehicles may have different customization needs (e.g., different driving preferences and intentions). The negotiator module 197 sends the modification messages to the connected vehicles 114, respectively.

[0081] The machine learning client 199 of each connected vehicle 114 then receives that particular modification message and parses the corresponding vehicle behavior data 133 contained within the received modification message.

[0082] For a connected vehicle 114 (e.g., vehicle 114A) that has a system driver and is controllable by a vehicle control system 156, the machine learning client 199 modifies one or more operations of the vehicle control system 156 based on the vehicle behavior data 133. For example, the machine learning client 199 modifies settings of the vehicle's ADAS system based on the vehicle behavior data 133. The connected vehicle 114A is then controlled to operate in accordance with the individual optimal behavior described by its vehicle behavior data 133.

[0083] For connected vehicles 114 that have a human driver and are not controlled by vehicle control system 156 (e.g., vehicle 114B), machine learning client 199 provides notifications to the driver via user interaction interface 161. The notifications may include instructions to operate the vehicle to conform to the individual optimal behavior described by vehicle behavior data 133. User interaction interface 161 may include a display (e.g., a touch screen), a sound reproduction device (e.g., a speaker), a camera, and any other interface that can be used to interact with the driver.

[0084] 1C is a block diagram illustrating an architecture 170 including a machine learning system 191 and one or more components of a connected vehicle 114 according to some embodiments. Here, in this example, the machine learning system 191 includes a predictor module 195 and a negotiator module 197.

[0085] The predictor module 195 includes code and routines that, when executed by the processor, cause the processor to analyze V2X data 128 received from the connected vehicle 114 and predict the future behavior of the connected vehicle 114. The predictor module 195 may include machine learning algorithms so that its predictions can automatically improve over time.

[0086] In some embodiments, when the V2X data 128 received from a particular connected vehicle 114 does not include data describing its customization needs, the predictor module 195 can estimate the customization needs of the vehicle. Additionally, when the V2X data 128 received from a particular connected vehicle 114 does not include data describing its future behavior, the predictor module 195 can generate forecast data 171 describing the future behavior of the vehicle. For example, the predictor module 195 can estimate one or more of driving preferences and weights of the driving preferences, driving intentions and weights of the driving intentions, and future behavior of the vehicle based on the V2X data 128 of the vehicle.

[0087] The negotiator module 197 includes code and routines that, when executed by the processor, cause the processor to analyze V2X data 128 received from the connected vehicle 114 and determine vehicle behavior data for the connected vehicle 114. In some embodiments, the negotiator module 197 provides its functionality based on the predicted future behavior output by the predictor module 195.

[0088] In some embodiments, the negotiator module 197 does not include machine learning algorithms because behaviors generated from machine learning algorithms are empirical and not suitable for safety applications. Thus, the generation of a global optimal behavior for the domain and an individual optimal behavior for the connected vehicle 114 is not a by-product of running a machine learning algorithm. Conversely, the negotiator module 197 can use an optimization mechanism to generate behaviors that ensure safety. An example application of the optimization mechanism is shown with reference to Figures 7A-7D.

[0089] The predictor module 195 and the negotiator module 197 have been described above with reference to FIG. 1B and so a similar description will not be repeated here.

[0090] 1D is a block diagram illustrating another architecture 175 including a machine learning system 191 and one or more components of a connected vehicle 114 according to some embodiments. Here, the machine learning system 191 further includes an offline teacher 194 and a safety checker 193.

[0091] In some embodiments, the optimization mechanism may take a long time to generate an optimized solution. Therefore, an offline teacher 194 may be added offline to apply the optimization mechanism as an offline teacher of the negotiator module 197. In this case, the negotiator module 197 may be implemented with a machine learning algorithm. To ensure the safety of the generated vehicle behavior, a safety checker 193 may be added to check whether the global optimal behavior or any individual optimal behavior of the domain generated by the negotiator module 197 is safe. For example, the safety checker 193 may determine whether the individual optimal behavior of the particular connected vehicle 114 satisfies the safety requirements in the domain. If the individual optimal behavior satisfies the safety requirements, the safety checker 193 may send the vehicle behavior data 133 describing the individual optimal behavior to the particular connected vehicle 114. If not, the safety checker 193 does not transmit the vehicle behavior data 133 and instructs the negotiator module 197 to regenerate an individual optimal behavior for this particular connected vehicle 114 .

[0092] Predictor module 195 and negotiator module 197 have been described above with reference to FIGS. 1B-1C, and similar description will not be repeated here.

[0093] With combined reference to Figures 1A-1D above and Figures 2-7D below, the machine learning system 191 and the machine learning client 199 can minimize the overall risk of violation of basic driving requirements. The machine learning system 191 and the machine learning client 199 can maximize the fulfillment of different vehicle customization needs (e.g., maximize the overall vehicle optimization). Additionally, the machine learning system 191 and the machine learning client 199 can customize the priorities of vehicles based on their driving intents (or driving preferences) and weights of the driving intents (or driving preferences).

[0094] Exemplary Computer System 2, a block diagram illustrates an example computer system 200 including a machine learning client 199 according to some embodiments. In some embodiments, computer system 200 may include a special-purpose computer system programmed to perform one or more steps of a method 300 described below with reference to FIG.

[0095] In some embodiments, computer system 200 may be a component of connected vehicle 114 (e.g., host vehicle 110 or remote vehicle 112). In some embodiments, computer system 200 may be an on-board computer of connected vehicle 114. In some embodiments, computer system 200 may include an engine control unit, a head unit, or some other processor-based computing device of connected vehicle 114.

[0096] Computer system 200 may include one or more of the following elements according to some embodiments: machine learning client 199, processor 125, and communication unit 145. Computer system 200 may further include one or more of the following elements: sensor set 154, GPS unit 150, memory 127, vehicle control system 156, and storage device 241. The components of computer system 200 are communicatively coupled to bus 220.

[0097] In a preferred embodiment, the processor 125 is communicatively connected to the bus 220 via signal line 237. The vehicle control system 156 is communicatively connected to the bus 220 via signal line 238. The communication unit 145 is communicatively connected to the bus 220 via signal line 246. The sensor set 154 is communicatively connected to the bus 220 via signal line 232. The GPS unit 150 is communicatively connected to the bus 220 via signal line 230. The storage device 241 is communicatively connected to the bus 220 via signal line 242. The memory 127 is communicatively connected to the bus 220 via signal line 244.

[0098] The following elements have been described above with reference to Figure 1A: processor 125, communication unit 145, sensor set 154, GPS unit 150, vehicle control system 156, and memory 127. Such description will not be repeated here.

[0099] Storage device 241 may be a non-transitory storage medium that stores data to provide the functionality described herein. Storage device 241 may be a DRAM device, an SRAM device, a flash memory, or some other device. In some embodiments, storage device 241 further includes non-volatile memory or similar persistent storage devices or media (e.g., hard disk drives, floppy disk drives, flash memory devices, etc.) for storing information in a more permanent form.

[0100] 2, the machine learning client 199 includes a communications module 202, a custom processing module 204, a V2X data module 206, and an execution module 208. In some embodiments, the machine learning client 199 further includes a predictor module 195. These components of the machine learning client 199 are communicatively coupled to each other via a bus 220. In some embodiments, the components of the machine learning client 199 may be stored within a single server or device. In some other embodiments, the components of the machine learning client 199 may be distributed and stored across multiple servers or devices.

[0101] The communications module 202 may be software that includes routines that handle communications between the machine learning client 199 and other components of the computer system 200. In some embodiments, the communications module 202 may be stored in the memory 127 of the computer system 200 and may be accessed and executed by the processor 125. The communications module 202 may be configured to cooperate and communicate over signal lines 222 between the processor 125 and other components of the computer system 200.

[0102] The communications module 202 transmits and receives data to and from one or more elements of the operating environment 100 via the communications unit 145. For example, the communications module 202 receives or transmits V2X data of other vehicles or V2X data of the connected vehicle 114 via the communications unit 145. The communications module 202 can transmit or receive any of the data or messages described above with reference to Figures 1A-1D via the communications unit 145.

[0103] In some embodiments, the communications module 202 receives data from other components of the machine learning client 199 and stores the data in one or more of the storage device 241 and the memory 127. The other components of the machine learning client 199 may cause the communications module 202 to communicate with other components of the computer system 200 or the operating environment 100 (via the communications unit 145). For example, the V2X data module 206 may use the communications module 202 to communicate with the sensor set 154 and cause the sensor set 154 to record sensor data.

[0104] The custom processing module 204 may be software that includes routines for generating customization data for the connected vehicle 114. In some embodiments, the custom processing module 204 may be stored in the memory 127 of the computer system 200 and may be accessed and executed by the processor 125. The custom processing module 204 may be configured to cooperate and communicate over signal lines 224 between the processor 125 and other components of the computer system 200.

[0105] In some embodiments, the customization processing module 204 is operable to generate customization data describing customization needs of the connected vehicle 114. For example, the customization processing module 204 can determine the customization needs of the connected vehicle 114 based on one or more user inputs or historical travel data of the connected vehicle 114. The customization needs of the connected vehicle 114 can be described by one or more customization parameters and one or more weights of the customization parameters. For example, the one or more customization parameters include one or more of a driving intent and a driving preference of the connected vehicle 114. The customization needs can be described by one or more of (1) a driving intent and a weight of the driving intent, and (2) a driving preference and a weight of the driving preference.

[0106] In some embodiments, the custom processing module 204 transmits the generated customization data to the V2X data module 206.

[0107] The V2X data module 206 may be software including routines for generating V2X data for the connected vehicle 114. In some embodiments, the V2X data module 206 may be stored in the memory 127 of the computer system 200 and may be accessed and executed by the processor 125. The V2X data module 206 may be configured to cooperate and communicate over signal lines 226 between the processor 125 and other components of the computer system 200.

[0108] In some embodiments, the V2X data module 206 causes the sensor set 154 and the ADAS system of the connected vehicle 114 to collect sensor and ADAS data. Optionally, the predictor module 195 is stored in the machine learning client 199 of the connected vehicle 114. The V2X data module 206 causes the predictor module 195 to analyze the sensor and ADAS data of the connected vehicle 114 to estimate the future behavior of the connected vehicle 114 by comparison with the past behavior of the connected vehicle 114. The predictor module 195 transmits data describing the future behavior of the connected vehicle 114 to the V2X data module 206.

[0109] The V2X data module 206 is operable to generate V2X data for the connected vehicle 114. In some embodiments, the V2X data includes customization data describing customization needs of the connected vehicle 114. For example, the V2X data includes one or more of parameter data describing one or more customization parameters (e.g., driving intent or driving preferences, etc.) and weight data describing one or more weights of the one or more customization parameters.

[0110] In some embodiments, the V2X data further includes sensor and ADAS data. For example, the V2X data further includes data describing a roadway environment surrounding the connected vehicle 114 and data describing a vehicle state of the connected vehicle 114. In some embodiments, the V2X data further includes one or more of data describing a design specification of the connected vehicle 114 and a future behavior of the connected vehicle 114.

[0111] The V2X data module 206 causes the communication unit 145 to transmit a feedback message over the V2X network, the feedback message including the V2X data of the connected vehicle 114 as its payload.

[0112] The predictor module 195 may be software that includes routines for predicting future behavior of the connected vehicle 114. In some embodiments, the predictor module 195 may be stored in the memory 127 of the computer system 200 and may be accessed and executed by the processor 125. The predictor module 195 may be configured to cooperate and communicate over signal lines 228 between the processor 125 and other components of the computer system 200.

[0113] The predictor module 195 was described above with reference to Figure 1B and will not be repeated here.

[0114] The execution module 208 may be software including routines for executing an individual optimal behavior of the connected vehicle 114. In some embodiments, the execution module 208 may be stored in the memory 127 of the computer system 200 and may be accessed and executed by the processor 125. The execution module 208 may be configured to cooperate and communicate over signal lines 223 between the processor 125 and other components of the computer system 200.

[0115] In some embodiments, the execution module 208 receives a modification message from the machine learning system 191 and parses vehicle behavior data for the connected vehicle 114 from the modification message. The vehicle behavior data describes an individual optimal behavior of the connected vehicle 114. The execution module 208 modifies operation of the vehicle control system 156 of the connected vehicle 114 based on the vehicle behavior data such that the connected vehicle 114 executes the individual optimal behavior. For example, the execution module 208 modifies operation of the vehicle control system 156 based on the vehicle behavior data such that the connected vehicle 114 is controlled to operate in accordance with the individual optimal behavior.

[0116] In some embodiments, performance of an individual optimal behavior by a connected vehicle 114 contributes to achieving a global optimal behavior for the area in which the connected vehicle 114 is located. Meanwhile, performance of an individual optimal behavior satisfies customization needs of the connected vehicle.

[0117] In some embodiments, performance of the individual optimal behavior by connected vehicle 114 indicates that no collision will occur between connected vehicle 114 and one or more other vehicles in the area. Performance of the individual optimal behavior by connected vehicle 114 also indicates that the customization needs of connected vehicle 114 will be satisfied without interfering with one or more customization needs of one or more other vehicles.

[0118] In some embodiments, the connected vehicle 114 and one or more other vehicles are included in a group of vehicles in a region, and achieving global optimal behavior for the region exhibits one or more of: (1) no collisions occur within the group of vehicles, and (2) the customization needs of each vehicle in the group are satisfied without modifying the customization needs of any other vehicle in the group.

[0119] In some embodiments, the individual optimal behavior includes one or more of a constrained trajectory, acceleration settings, steering angle settings, and speed settings of the connected vehicle 114 that are optimized for the connected vehicle 114. In some embodiments, the individual optimal behavior satisfies one or more of traffic regulation requirements and safety requirements in the domain.

[0120] Example Process Referring to Figure 3, a flow chart of an example method 300 for performing individual optimal behavior in a connected vehicle 114 is shown in accordance with some embodiments. The steps of the method 300 may be performed in any order, not necessarily in the order shown in Figure 3. The connected vehicle 114 may be the ego vehicle 110 or a remote vehicle 112. The method 300 may be performed by a machine learning client 199 of the connected vehicle 114.

[0121] In step 301, the V2X data module 206 generates V2X data including customization data describing the customization needs of the connected vehicle 114.

[0122] In step 303, the V2X data module 206 transmits a feedback message including the V2X data via V2X communication.

[0123] At step 305, the execution module 208 receives a modification message via V2X communication that includes vehicle behavior data describing an individual optimal behavior of the connected vehicle 114. For example, the individual optimal behavior is determined by the machine learning system 191 based at least in part on the V2X data of the connected vehicle 114.

[0124] In step 307, the execution module 208 determines whether the connected vehicle 114 is a controllable vehicle. If the connected vehicle 114 is a controllable vehicle, the method 300 proceeds to step 309. If not, the method 300 proceeds to step 311.

[0125] For example, the execution module 208 determines whether the connected vehicle 114 is controlled by the vehicle control system 156. If the connected vehicle 114 is controlled by the vehicle control system 156, the method 300 proceeds to step 309. If the connected vehicle 114 is controlled by a human driver, the method 300 proceeds to step 311.

[0126] In step 309, the execution module 208 modifies the operation of the vehicle control system 156 of the connected vehicle 114 based on the vehicle behavior data so that the connected vehicle 114 executes the individual optimal behavior, where the execution of the individual optimal behavior by the connected vehicle 114 contributes to achieving a global optimal behavior of the area in which the connected vehicle 114 is located, while the customization needs of the connected vehicle 114 are also satisfied by the execution of the individual optimal behavior.

[0127] In step 311, the execution module 208 provides a notification of the individual optimal behavior to the driver of the connected vehicle 114 via the user interaction interface 161. The human driver can then follow the instructions provided by the notification such that the individual optimal behavior of the connected vehicle 114 is also executed.

[0128] Figure 4 illustrates a method 400 for providing a global optimal behavior within a region and an individual optimal behavior for a vehicle within the region, according to some embodiments. The steps of method 400 may be performed in any order, not necessarily the order shown in Figure 4. Method 400 may be performed by a machine learning system 191, which may be installed on a cloud server, an edge server, a roadside unit, the ego vehicle 110, or any other end point of a roadway.

[0129] In step 401, the negotiator module 197 of the machine learning system 191 receives V2X data from vehicles in the area and parses customization data for each vehicle from the V2X data.

[0130] In step 403, negotiator module 197 determines, based on the customization data, customization needs for each vehicle described by one or more customization parameters and one or more weights for the one or more customization parameters. For example, based on the customization data for each vehicle, negotiator module 197 determines one or more of driving preferences and weights for the driving preferences and driving intentions and weights for the driving intentions.

[0131] In step 405, the negotiator module 197 determines safe areas for uncontrollable vehicles within the region. An uncontrollable vehicle may be a vehicle that is not controlled by the vehicle control system. For example, an uncontrollable vehicle does not automatically execute the individual optimal behavior generated by the machine learning system 191.

[0132] For example, the negotiator module 197 may reserve the area in which the out-of-control vehicle is located as a safe area for the out-of-control vehicle, in some embodiments, other out-of-control vehicles are barred from the out-of-control vehicle's safe area.

[0133] In step 407, the negotiator module 197 determines region behavior data describing an overall optimal behavior of the region based at least in part on the V2X data. The overall optimal behavior includes constrained trajectories for vehicles (e.g., controllable vehicles, uncontrollable vehicles, or both) in the region. The execution of these constrained trajectories (1) satisfies traffic regulation requirements and safety requirements in the region, and (2) maximizes the fulfillment of vehicle customization needs.

[0134] For example, the negotiator module 197 determines whether the safety requirements in the region can be satisfied by modifying the trajectories of the controllable vehicles in the region. If the safety requirements can be satisfied by modifying the trajectories of the controllable vehicles in the region, the negotiator module 197 modifies the trajectories of the controllable vehicles in the region to generate forced trajectories for the controllable vehicles. In this case, the global optimal behavior includes the forced trajectories for the controllable vehicles in the region. However, if the safety requirements cannot be satisfied by modifying the trajectories of the controllable vehicles in the region, the negotiator module 197 modifies the trajectories of all vehicles (whether controllable or uncontrollable) in the region. The negotiator module 197 generates forced trajectories for all vehicles. In this case, the global optimal behavior includes the forced trajectories for the controllable and uncontrollable vehicles in the region.

[0135] In step 409, the negotiator module 197 determines vehicle behavior data describing an individual optimal behavior of each vehicle based on the regional behavior data. The individual optimal behavior of each vehicle includes a forced trajectory of the vehicle that (1) satisfies traffic regulation requirements and safety requirements, and (2) satisfies the customization needs of the vehicle without modifying the customization needs of other vehicles.

[0136] For example, the negotiator module 197 may modify the trajectory included in the vehicle's driving intent based on the region behavior data to generate an enforced trajectory for the vehicle. The execution of the enforced trajectory by the vehicle contributes to one form of achieving an overall optimal behavior of the region while also satisfying the vehicle's customization needs.

[0137] In step 411, the negotiator module 197 transmits corresponding vehicle behavior data to each vehicle, describing the individual optimal behavior of that vehicle.

[0138] 5A-5B are graphical representations 500 and 550 illustrating an example operating scenario for providing optimal vehicle behavior according to some embodiments. In this operating scenario, there are four vehicles on a roadway (vehicles A, B, C, and D). Each of the vehicles may include an instance of the machine learning client 199.

[0139] Referring to FIG. 5A, vehicle A has a driving preference of fuel economy (e.g., “eco”) with a “low” weight. Vehicle A has a planned trajectory to move into the correct lane where vehicles D and B currently reside. Vehicle B has a driving preference of time saving with a “high” weight and a planned trajectory to go straight. Vehicle C has a driving preference of fuel economy with a “low” weight and a planned trajectory to merge into the left lane. Vehicle D has a driving preference of comfort with a “high” weight and a planned trajectory to go straight. These vehicles can each transmit V2X data (including data describing the driving preferences, the weights of the driving preferences, the planned trajectory, etc.) to the machine learning system 191 via feedback messages.

[0140] As shown in FIG. 5A, if vehicles A to D move according to their respective planned trajectories, these vehicles may encounter each other in the same lane and a collision may occur.

[0141] The machine learning system 191 can determine an overall optimal behavior for the vehicles A-D based at least in part on the V2X data received from these vehicles. One form of achieving the overall optimal behavior can ensure that (1) no collisions occur between any of the vehicles A-D, and (2) driving preferences of the vehicles A-D are satisfied, while still satisfying the overall optimal behavior. The machine learning system 191 determines an individual optimal behavior for each of the vehicles such that the overall optimal behavior is achieved within the domain. The machine learning system 191 transmits vehicle behavior data via a correction message that describes the individual optimal behavior of each respective vehicle.

[0142] 5B, the individual optimal behavior of vehicle A includes modifying its planned trajectory ("move to right lane") to a forced trajectory of "go straight." When vehicle behavior data describing the individual optimal behavior is received, the result is that vehicle A operates in compliance with the forced trajectory of "go straight."

[0143] The individual optimal behavior of vehicle B includes modifying its planned trajectory ("go straight") to a forced trajectory of "move to the left lane and increase the moving speed." When vehicle behavior data describing the individual optimal behavior is received, the result is that vehicle B operates in accordance with the forced trajectory of "move to the left lane and increase the moving speed."

[0144] The individual optimal behavior of vehicle C includes smoothly merging into the lane to its left. When vehicle behavior data describing the individual optimal behavior is received, the result is that vehicle C smoothly merges into the lane to its left.

[0145] The individual optimal behavior of vehicle D includes moving on its planned trajectory with acceleration turned off. When vehicle behavior data describing the individual optimal behavior is received, vehicle D turns off acceleration.

[0146] 6A-6B are graphical representations 600 and 650 illustrating another example operating scenario for providing optimal vehicle behavior according to some embodiments. In this operating scenario, there are three vehicles at an intersection: vehicles A, B, and C. Each of the vehicles may include an instance of the machine learning client 199.

[0147] Referring to FIG. 6A, vehicle A has a driving preference of "large" weighted time saving. Vehicle B has a driving preference of "small" weighted time saving. Vehicle C has a driving preference of "small" weighted fuel efficiency. Vehicles A-C each have a planned trajectory of going straight. These vehicles can each transmit V2X data (including data describing the driving preferences, the weights of the driving preferences, the planned trajectory, etc.) to machine learning system 191 via feedback messages.

[0148] As shown in FIG. 6A, when vehicles A to C move according to their respective planned trajectories, there is a possibility that these vehicles will encounter each other at an intersection and a collision will occur.

[0149] The machine learning system 191 can determine an overall optimal behavior for the vehicles A-C based at least in part on the V2X data received from these vehicles. One form of achieving the overall optimal behavior can ensure that (1) no collisions occur between any of the vehicles A-C, and (2) the driving preferences of the vehicles A-C are satisfied.

[0150] 6B , because the driving preferences of vehicles A and B are “save time,” while the driving preference of vehicle C is “eco,” the machine learning system 191 can instruct vehicles A and B to pass through the intersection before vehicle C. Because the weight of the driving preference of vehicle A is greater than the weight of the driving preference of vehicle B, the machine learning system 191 can instruct vehicle A to pass through the intersection before vehicle B.

[0151] The machine learning system 191 then determines individual optimal behaviors of the vehicles A to C such that the vehicle A passes through the intersection first, the vehicle B passes through the intersection next, and the vehicle C passes through the intersection last. The machine learning system 191 transmits vehicle behavior data describing the individual optimal behaviors of the vehicles to the vehicles A to C, respectively, so that the vehicles A to C can perform their respective individual optimal behaviors.

[0152] 7A-7B are graphical representations 700 and 710 illustrating an example simulation for providing optimal vehicle behavior according to some embodiments. There are three vehicles in the simulation (vehicles A, B, and C).

[0153] In some embodiments, the machine learning system 191 negotiator module 197 uses an optimization mechanism to determine the vehicle's individual optimal behavior. Within the optimization mechanism, one or more constraints are configured such that basic driving requirements are satisfied. The objective of the optimization mechanism includes maximizing the fulfillment of the vehicle's customization needs. For example, the objective of the optimization mechanism may include one or more of minimizing arrival time, minimizing fuel consumption, minimizing sudden movements, minimizing comfort, etc.

[0154] For example, the driving preference of vehicle A is weighted W A The travel time of vehicle B is reduced, and the driving preference of vehicle B is weighted W B The fuel efficiency of vehicle C and the driving preference of vehicle C are weighted W C Then, the objective of the optimization mechanism is to A Minimize the arrival time of vehicle A weighted by weight W B Minimizing the fuel consumption of vehicle B weighted by and weight W C This may include maximizing the comfort experience of vehicle C weighted by

[0155] The negotiator module 197 considers one or more of the following factors (1)-(7) when applying the optimization mechanism: Factor (1) includes the trajectory planned by each vehicle ("planned trajectory"); Factor (2) includes one or more fundamental rules that the vehicle's trajectory must adhere to (e.g., equations of motion, speed limits, power limits, etc.); Factor (3) includes one or more traffic rules that must be adhered to (e.g., each vehicle must be in the correct lane and facing the correct direction, each vehicle must observe the speed limit, etc.). Traffic rules may describe, for example, the shape of the lane, the physical boundaries of the lane, and the direction of travel; Factor (4) includes that the vehicle must not collide with adjacent vehicles (e.g., sufficient space must be maintained between adjacent vehicles); Factor (5) includes whether a forced trajectory of the vehicle is possible without colliding with other vehicles; Factor (6) includes whether the vehicle is controllable or uncontrollable; Factor (7) includes a requirement that there be no collisions with any vehicle in the area.

[0156] Referring to the simulation of FIG. 7A, the negotiator module 197 considers the above factors (1)-(4) using an objective function that minimizes the arrival time of the vehicles when applying the optimization mechanism. The planned trajectory of each vehicle is shown in FIG. 7A. The optimization mechanism uses a mixed integer linear programming (MILP) as the optimization problem formulation and applies an optimization solver (e.g., OR tool) to solve the optimization problem. The simulation results with each vehicle's forced trajectory are shown in FIG. 7B. The negotiator module 197 determines the forced trajectory of each vehicle so that collisions do not occur on the roadway.

[0157] 7C-7D are graphical representations 730 and 750 illustrating another exemplary simulation to provide optimal vehicle behavior according to some embodiments. Planned trajectories of vehicles A-C are shown in FIG. 7C. The negotiator module 197 considers the above factors (1)-(4) with an objective function that minimizes the arrival times of the vehicles when applying the mechanism. Furthermore, for a particular vehicle group considered, the optimization mechanism defines the planned trajectories of other groups as no-go areas for that particular vehicle group. The optimization mechanism uses MILP as the optimization problem formulation and applies an optimization solver (e.g., OR tool) to solve the optimization problem.

[0158] For example, when applying the optimization problem, the negotiator module 197 determines which vehicles have intersecting planned trajectories (e.g., vehicles B and C have intersecting planned trajectories). The negotiator module 197 classifies the vehicles with intersecting planned trajectories into the same group. For example, the negotiator module 197 classifies vehicle A into group A because vehicle A does not have intersecting planned trajectories with other vehicles. The negotiator module 197 classifies vehicles B and C into group B because vehicles B and C have intersecting planned trajectories. The negotiator module 197 then excludes the planned trajectories of other groups from consideration when determining the forced trajectory of a particular group (e.g., the planned trajectories of other groups are considered obstacles for the particular group). For example, the negotiator module 197 excludes the planned trajectory of vehicle A from consideration when determining the forced trajectory of group B.

[0159] The simulation result of the forced trajectories of the vehicles is shown in FIG. 7D. For example, the negotiator module 197 determines the forced trajectory of vehicle A in group A. The negotiator module 197 also determines the forced trajectories of vehicles B and C in group B, respectively. No vehicle collision occurs. By excluding the planned trajectories of other unrelated vehicles (e.g., vehicle A), the time to execute the optimization mechanism of FIG. 7C-7D is shorter compared to the simulation shown in FIG. 7A-7B.

[0160] In the above description, for ease of understanding, many specific details are set forth in order to provide a thorough understanding of the present specification. However, it will be apparent to those skilled in the art that the present disclosure can be practiced without these specific details. In some instances, structures and devices are shown in block diagram form in order to avoid obscuring the description. For example, the present embodiment may be described above primarily with reference to a user interface and specific hardware. However, the present embodiment may be applicable to any type of computer system capable of receiving data and commands, and any peripheral device that provides services.

[0161] Reference herein to "some embodiments" or "some instances" means that a particular feature, structure, or characteristic described in connection with that embodiment or instance may be included in at least one embodiment herein. The appearances of the phrase "in some embodiments" in various places in the specification are not necessarily all referring to the same embodiments.

[0162] Some portions of the detailed descriptions which follow are presented in terms of algorithms and symbolic representations of operations on data bits within a computer memory. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. An algorithm is herein, and generally conceived to be, a self-consistent sequence of steps leading to a desired result. The steps are those requiring physical manipulations of physical quantities. Usually, though not necessarily, these physical quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has proven convenient at times, primarily for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like.

[0163] It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless otherwise indicated, and as will be apparent from the discussion that follows, discussions throughout the description including "processing" or "calculating" or "calculating" or "determining" or "displaying" or the like will be understood to refer to operations and processes of a computer system or similar electronic computing device that manipulates and converts data represented as physical (electronic) quantities in the computer system's registers and memory into other data similarly represented as physical quantities in the computer system's memory or registers or other such information storage, transmission, or display devices.

[0164] The embodiments herein also relate to an apparatus for performing the operations described herein. This apparatus may be specially constructed for the required purposes or may include a general-purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored on a computer-readable storage medium, including any type of disk including, but not limited to, floppy disks, optical disks, CD-ROMs, and magnetic disks, each coupled to a computer system bus, read-only memory (ROM), random access memory (RAM), EPROM, EEPROM, magnetic or optical cards, flash memory including USB keys with non-volatile memory, or any type of medium suitable for storing electronic instructions.

[0165] The specification can take the form of some entirely hardware embodiments, some entirely software embodiments, or some embodiments containing both hardware and software elements. In some preferred embodiments, the specification is implemented in software, including but not limited to firmware, resident software, microcode, etc.

[0166] Furthermore, the present specification may take the form of a computer program product accessible from a computer usable or computer readable medium providing program code for use by or in connection with a computer or any instruction execution system. For purposes of this specification, a computer usable or computer readable medium may also be any apparatus that can contain, store, communicate, propagate, or transfer a program for use by or in connection with an instruction execution system, apparatus, or device.

[0167] A data processing system suitable for storing or executing program code will include at least one processor coupled directly or indirectly through a system bus to memory elements. The memory elements may include local memory used during the actual execution of the program code, mass storage devices, and cache memory that provides temporary storage of at least some of the program code to reduce the number of times the code must be retrieved from mass storage devices during execution.

[0168] Input / output or I / O devices (including but not limited to keyboards, displays, pointing devices, etc.) can be coupled to the system either directly or through intervening I / O controllers.

[0169] Network adapters may also be coupled to the system that enable the data processing system to become coupled to other data processing systems or remote printers or storage devices through intervening private or public networks. Modems, cable modems and Ethernet cards are just a few of the currently available types of network adapters.

[0170] Additionally, the algorithms and displays presented herein are not inherently related to any particular computer or other apparatus. Various general-purpose systems may be used in conjunction with programs in accordance with the teachings herein, or it may prove more convenient to construct a specialized apparatus to perform the required method steps. The required structure for a variety of these systems will appear from the description that follows. Moreover, the invention is not described with reference to any particular programming language. It will be understood that a variety of programming languages ​​can be used to implement the teachings herein as set forth.

[0171] The above description of the embodiments of the present specification has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the specification to the precise form disclosed. Many modifications and variations are possible in light of the above teachings. It is intended that the scope of the present disclosure be limited not by this detailed description, but by the scope of the claims of this application. As will be appreciated by one of ordinary skill in the art, the specification can be embodied in other specific forms without departing from its spirit or essential characteristics. Similarly, the specific names and divisions of modules, routines, features, attributes, methods, and other aspects are not essential or critical, and the mechanisms for implementing the specification or its features may have different names, divisions, or formats. Moreover, as will be appreciated by one of ordinary skill in the art, the modules, routines, features, attributes, methods, and other aspects of the present disclosure may be implemented as software, hardware, firmware, or any combination of the three. Also, wherever a component (e.g., a module) herein is implemented in software, the component may be implemented as a stand-alone program, as part of a larger program, as multiple separate programs, as a statically or dynamically linked library, as a kernel loadable module, as a device driver, or in any other manner known now or in the future to those skilled in the art of computer programming. Furthermore, the present disclosure is in no way limited to implementation in any specific programming language, or for any specific operating system or environment. Accordingly, the present disclosure is intended to be illustrative, without limiting the scope of the present disclosure, which is set forth in the appended claims.

[0172] The present disclosure includes the following examples. [Example 1] 1. A computer program product comprising: 1. A non-transitory memory of an on-board computer system of a connected vehicle, comprising: When executed by a processor, the processor: generating vehicle-to-everything (V2X) data including customization data describing customization needs of the connected vehicle; Transmitting a feedback message including the V2X data via V2X communication; receiving, via the V2X communication, a modified message including vehicle behavior data describing an individual optimal behavior of the connected vehicle determined based at least in part on the V2X data; a non-transitory memory storing computer executable code for modifying operation of a vehicle control system of the connected vehicle based on the vehicle behavior data such that the connected vehicle achieves the individual optimal behavior; 23. A computer program product, wherein performance of the individual optimal behavior by the connected vehicle contributes to achieving a global optimal behavior of an area in which the connected vehicle is located while also satisfying the customization needs of the connected vehicle. [Example 2] 2. The computer program product of Example 1, wherein the performance of the individual optimal behavior by the connected vehicle indicates one or more of: no collision occurs between the connected vehicle and one or more other vehicles in the area; and customization needs of the connected vehicle are satisfied without interfering with one or more customization needs of the one or more other vehicles. [Example 3] 3. The computer program product of Example 2, wherein the connected vehicle and the one or more other vehicles are included in a group of vehicles in the region, and achieving an overall optimal behavior of the region indicates one or more of: no collisions occurring within the group of vehicles; and the customization needs of each vehicle in the group being satisfied without modifying the customization needs of the remaining vehicles in the group. [Example 4] 2. The computer program product of Example 1, wherein the customization needs of the connected vehicle are described by one or more of driving intentions of the connected vehicle and a first weighting of the driving intentions, and driving preferences of the connected vehicle and a second weighting of the driving preferences. [Example 5] 1. A method for a connected vehicle, comprising: transmitting, via vehicle-to-everything (V2X) communication, V2X data including customization data describing customization needs of the connected vehicle; receiving vehicle behavior data via the V2X communication describing an individual optimal behavior of the connected vehicle determined based at least in part on the V2X data; and modifying operation of a vehicle control system of the connected vehicle based on the vehicle behavior data such that the connected vehicle performs the individual optimal behavior; A method, wherein performance of the individual optimal behavior by the connected vehicle contributes to achieving an overall optimal behavior of an area in which the connected vehicle is located while also satisfying the customization needs of the connected vehicle. [Example 6] 6. The method of example 5, wherein the performance of the individual optimal behavior by the connected vehicle indicates one or more of: no collision occurs between the connected vehicle and one or more other vehicles in the area; and customization needs of the connected vehicle are satisfied without interfering with one or more customization needs of the one or more other vehicles. [Example 7] 7. The method of example 6, wherein the connected vehicle and the one or more other vehicles are included in a group of vehicles in the region, and achieving an overall optimal behavior of the region indicates one or more of: no collisions occurring within the group of vehicles, and the customization needs of each vehicle in the group being satisfied without modifying the customization needs of the remaining vehicles in the group. [Example 8] 6. The method of example 5, wherein the customization needs of the connected vehicle are described by one or more customization parameters and one or more weights of the one or more customization parameters. [Example 9] 9. The method of example 8, wherein the one or more customization parameters include one or more of a driving intent and a driving preference associated with the connected vehicle. [Example 10] 9. The method of example 8, further comprising generating V2X data comprising one or more of: parameter data describing the one or more customization parameters, weight data describing the one or more weights, sensor data of the connected vehicle, and advanced driver assistance system (ADAS) data of the connected vehicle. [Example 11] 6. The method of example 5, further comprising predicting a future behavior of the connected vehicle, wherein the V2X data further comprises prediction data describing the future behavior of the connected vehicle. [Example 12] 6. The method of example 5, wherein the individual optimal behavior includes a constrained trajectory, acceleration settings, steering angle settings, and speed settings of the connected vehicle that are optimized for the connected vehicle. [Example 13] 6. The method of example 5, wherein the individual optimal behavior satisfies one or more of traffic regulation requirements and safety requirements within the region. [Example 14] 1. A system comprising: An on-board computer system of a connected vehicle including a non-transitory memory, The memory, when executed by the vehicle-mounted computer system, causes the vehicle-mounted computer system to: transmitting, via vehicle-to-everything (V2X) communication, V2X data including customization data describing customization needs of the connected vehicle; receiving, via the V2X communication, vehicle behavior data describing an individual optimal behavior of the connected vehicle determined based at least in part on the V2X data; storing computer code for modifying operation of a vehicle control system of the connected vehicle based on the vehicle behavior data such that the connected vehicle performs the individual optimal behavior; The system, wherein performance of the individual optimal behavior by the connected vehicle contributes to achieving an overall optimal behavior of the area in which the connected vehicle is located while also satisfying the customization needs of the connected vehicle. [Example 15] The system of Example 14, wherein the achievement of the individual optimal behavior by the connected vehicle indicates one or more of: no collision occurs between the connected vehicle and one or more other vehicles in the area; and customization needs of the connected vehicle are satisfied without interfering with one or more customization needs of the one or more other vehicles. [Example 16] The system of Example 15, wherein the connected vehicle and the one or more other vehicles are included in a vehicle group within the area, and achieving an overall optimal behavior of the area indicates one or more of: no collisions occurring within the vehicle group; and the customization needs of each vehicle in the group being satisfied without modifying the customization needs of the remaining vehicles in the group. [Example 17] 15. The system of example 14, wherein the customization needs of the connected vehicle are described by one or more customization parameters and one or more weights of the one or more customization parameters. [Example 18] 20. The system of example 17, wherein the one or more customization parameters include one or more of a driving intent and a driving preference associated with the connected vehicle. [Example 19] 18. The system of Example 17, wherein the computer code, when executed by the on-board computer system, further causes the on-board computer system to generate V2X data including one or more of: parameter data describing the one or more customization parameters of the connected vehicle, weight data describing the one or more weights, sensor data, and advanced driver assistance system (ADAS) data for the connected vehicle. [Example 20] 15. The system of Example 14, wherein the computer code, when executed by the on-board computer system, causes the on-board computer system to further predict future behavior of the connected vehicle, and the V2X data further includes prediction data describing the future behavior of the connected vehicle.

Claims

1. A computer program comprising:

1. A non-transitory memory of an on-board computer system of a connected vehicle, comprising: When executed by a processor, the processor: generating vehicle-to-everything (V2X) data including customization data describing customization needs of the connected vehicle; transmitting a feedback message including the V2X data via V2X communication; receiving, via the V2X communication, a modified message including vehicle behavior data describing an individual optimal behavior of the connected vehicle determined based at least in part on the V2X data; a non-transitory memory storing computer executable code for modifying operation of a vehicle control system of the connected vehicle based on the vehicle behavior data such that the connected vehicle achieves the individual optimal behavior; The performance of the individual optimal behavior by the connected vehicle contributes to achieving a global optimal behavior of the area in which the connected vehicle is located while also satisfying the customization needs of the connected vehicle; 11. A computer program product comprising: a computer program product, the computer program product comprising: a computer program product configured to program a customization need for the connected vehicle;

Citation Information

Patent Citations

  • Cellular network-based auxiliary driving method and traffic control unit

    CN108022450A

  • Driving support device and driving support system

    JP2007241729A

  • Driving support device and driving support method

    JP2018026065A

  • Driving device for vehicle

    JP2018105235A