Intelligent Road Infrastructure System (IRIS): System and Method

Abstract

To provide a system and method for an Intelligent Road Infrastructure System (IRIS), which facilitates vehicle operations and control for a connected automated vehicle highway (CAVH) system.SOLUTION: An IRIS system and method provide vehicles with individually customized information and real-time control instructions for vehicles to fulfill driving tasks such as car following, lane changing, and route guidance. The IRIS system and method also manage transportation operations and management services for both freeways and urban arterial. In some embodiments, the IRIS comprises or consists of one or more of the following physical subsystems: (1) a roadside unit (RSU) network, (2) a traffic control unit (TCU) and a traffic control center (TCC) network, (3) a vehicle onboard unit (OBU), (4) traffic operations centers (TOCs), and (5) cloud information and computing services. The IRIS manages one or more of the following function categories of sensing, transportation behavior prediction and management, planning and decision making, and vehicle control. The IRIS is supported by real-time wired and / or wireless communication, power supply networks, and cyber safety and security services.SELECTED DRAWING: None

Description

[Technical Field]

[0001] This application claims the benefit of U.S. Provisional Patent Application No. 62 / 627,005, filed February 6, 2018, the contents of which are incorporated herein by reference in their entirety.

[0002] The present invention relates to an intelligent road infrastructure system that provides traffic management and operation and individual vehicle control for connected autonomous vehicles (CAVs), and more particularly to a system that controls CAVs by sending detailed, time-sensitive control commands and traffic information to individual vehicles that are customized for autonomous vehicle operation, such as vehicle following, lane changes, route guidance, and other related information. [Background technology]

[0003] Autonomous vehicles—vehicles that can sense and navigate their surroundings, with or without reduced manual input—are in development. They are currently in the experimental testing stage and have not yet reached widespread commercial use. Existing approaches pose significant challenges for widespread implementation, as they require expensive and complex on-board systems.

[0004] Alternative systems and methods that address these issues are described in U.S. Patent Application No. 15 / 628,331, filed June 20, 2017, and U.S. Provisional Patent Application No. 62 / 626,862, filed February 6, 2018, the disclosures of which are incorporated herein by reference in their entireties (hereinafter referred to as the CAVH system). Summary of the Invention [Problem to be solved by the invention]

[0005] The present invention provides systems and methods for an Intelligent Roadway Infrastructure System (IRIS) that facilitates the operation and control of vehicles in a Connected Autonomous Vehicle Highway (CAVH) system. The IRIS system and method provides vehicles with personalized information and real-time control commands to perform driving tasks such as vehicle following, lane changes, and route guidance. The IRIS system and method also provides traffic operations and management services on both highways and urban arterials. [Means for solving the problem]

[0006] Some embodiments comprise or consist of one or more of the following physical subsystems: (1) Roadside Unit (RSU) network, (2) Traffic Control Unit (TCU) and Traffic Control Center (TCC) network, (3) Vehicle On-Board Unit (OBU), (4) Traffic Operation Centers (TOCs), and (5) Cloud information and computing services. IRIS manages one or more of the following functional categories: Sensing, traffic behavior prediction and management, planning and decision making, vehicle control IRIS is supported by real-time radio and / or wireless communications, power networks, and cyber safety and security services.

[0007] Current technology provides a comprehensive system of full-vehicle operation and control for connected and automated vehicles and highway systems by sending detailed, time-sensitive control commands to individual vehicles, suitable for a portion of a lane or an entire highway. In some embodiments, these control commands are vehicle-specific, optimized, and communicated from the highest-level TCC and transmitted by the lowest-level TCU. These TCCs / TCUs are hierarchical, covering various levels of domain.

[0008] In some embodiments, systems and methods are provided herein that comprise an Intelligent Roadway Infrastructure System (IRIS) that facilitates vehicle operation and control for Connected Autonomous Vehicle Highways (CAVH). In some embodiments, the systems and methods provide customized, detailed information and time-sensitive control commands to individual vehicles to enable them to perform driving tasks such as vehicle following, lane changes, and route guidance, and provide traffic operation and maintenance services to vehicles on both freeways and urban arterials. In some embodiments, the systems and methods are built and managed as an open platform, and in some embodiments, the subsystems described below are physically and / or logically shared among various CAVH systems owned and / or operated by different entities and including one or more of the following physical subsystems: a. Roadside Unit (RSU) network with functions including sensing, communication, control (fast / easy), and range calculation b. Traffic Control Unit (TCU) and Traffic Control Center (TCC) Network c. Vehicle On-Board Unit (OBU) and associated vehicle interfaces d. Traffic Operations Center e. Cloud-based platforms for information and computing services

[0009] In one embodiment, the system and method manage one or more of the following functional categories: a. Sensing b. Traffic behavior prediction and management c. Planning and decision-making d. Vehicle control

[0010] In one embodiment, the system and method are supported by one or more of the following: a. Real-time communications over wired and wireless media b. Power supply network c. Cyber Safety and Security Systems

[0011] In some embodiments, the functional categories and physical subsystems of the IRIS have various configurations with respect to the allocation of functions and physical devices. For example, in some embodiments, the configurations include: a. RSUs provide real-time vehicle environment sensing and traffic behavior prediction, and send instantaneous control commands for individual vehicles via OBUs. b. TCU / TCC and traffic operation centers provide short-term and long-term traffic behavior prediction and management, planning and decision-making, and traffic information collection / processing, with or without cloud information and computing services. c. Vehicle OBUs collect and transmit vehicle-generated data, such as vehicle movement and status, to the RSUs and receive input from the RSUs, as described above. Based on the input from the RSUs, the OBUs facilitate vehicle control. In the event of a failure in the vehicle control system, the OBU can take over quickly and safely bring the vehicle to a halt. In one embodiment, the vehicle OBU includes one or more of the following modules: (1) a communications module, (2) a data collection module, and (3) a vehicle control module. Other modules may also be included.

[0012] In one embodiment, the communication module is configured to enable data exchange between the RSUs and the OBUs, and optionally with other vehicle OBUs. Vehicle-sourced data may include, but is not limited to: a. Manually entered data such as origin-destination points, estimated travel time, scheduled departure and arrival times, and service requests; b. Human situation data such as human behavior and human condition (e.g., fatigue level) c. Vehicle status data, such as vehicle ID, type, and data collected by data collection modules

[0013] Data from RSUs may include, but is not limited to: a. Vehicle control commands, such as desired longitudinal and lateral acceleration, desired vehicle direction, etc. b. Route and traffic information, including traffic conditions, incidents, intersection locations, entrances and exits; c. Service data, such as fuel stations and points of interest

[0014] In some embodiments, the data collection module collects data from sensors installed inside and outside the vehicle to monitor vehicle and occupant conditions, including, but not limited to, one or more of the following: a. Vehicle engine condition b. Vehicle speed c. Surrounding objects detected by the vehicle d. Personnel situation

[0015] In some embodiments, a vehicle control module is used to execute control commands from the RSU for driving tasks such as vehicle following, lane changing, etc.

[0016] In some embodiments, IRIS sensing capabilities generate comprehensive real-time, short-term, and long-term information for traffic behavior prediction and management, planning and decision-making, vehicle control, and other functions, including but not limited to: a. Vehicle environment such as distance between vehicles, speed difference, obstacles, lane departure, etc. b. Weather conditions, pavement condition, etc. c. Vehicle attribute data such as speed, location, type, and automation level d. Traffic conditions such as traffic flow rate, occupancy rate, and average speed e. Traffic information such as traffic lights and speed limits f. Incident collection for collisions, traffic jams, etc.

[0017] In some embodiments, IRIS is supported by sensing capabilities that predict conditions across the transportation network at various scales, including but not limited to: a. The microscopic level, which targets individual vehicles, such as longitudinal movement (vehicle following, acceleration / deceleration, parking / stopping) and lateral movement (lane keeping, lane changing). b. Mesoscopic level, covering road corridors and segments, including early notification of special events, incident prediction, interwoven segment merging and splitting, platoon breaking and merging, variable speed limit prediction and response, segment duration prediction, and segment traffic flow prediction. c. Macroscopic level, targeting road networks, such as potential congestion prediction, potential incident prediction, network traffic demand prediction, network state prediction, and network travel time prediction.

[0018] In some embodiments, IRIS is supported by sensing and predictive capabilities, provides planning and decision-making capabilities, and informs target vehicles and entities on a variety of broad scales, including but not limited to: a. Microscopic level such as longitudinal control (vehicle following, acceleration and deceleration) and lateral control (lane keeping, lane changing). b. Mesoscopic level, such as early notification of special events, road construction zones, slow down zones, incident detection, buffer spaces, weather forecast notifications, etc. Planning at this level ensures that vehicles follow all established rules (temporary or permanent), improving safety and efficiency. c. Macro level, such as route planning and guidance, network demand management, etc.

[0019] In some embodiments, the planning and decision-making capabilities of IRIS enhance reactive incident management and support proactive incident prediction and prevention, including but not limited to: As a reactive measure, IRIS will automatically detect incidents as they occur and coordinate with relevant agencies for further action. It will also provide incident warnings and re-route instructions for affected traffic. b. As a proactive measure, IRIS will predict potential incidents, send control commands to guide affected vehicles to safety, and coordinate with relevant agencies for further action.

[0020] In some embodiments, the vehicle control functions of IRIS are supported by sensing, traffic behavior prediction and management, planning and decision making, and further include, but are not limited to: a. Maintaining speed and distance: Maintaining minimum distance and maximum speed in the lane to reach maximum possible traffic capacity. b. Collision Avoidance: Detect potential accidents / collisions on the lane and send warning messages and collision avoidance commands to the vehicle. Under such circumstances, the vehicle must follow the commands from the lane management system. c. Lane Keeping: Maintaining vehicle movement in designated lanes. d. Curvature / Altitude Control: Ensures the vehicle maintains and adapts to appropriate speeds and angles based on factors such as road geometry and pavement conditions. e. Lane Change Control: Coordinating vehicle lane changes in an orderly manner with minimal disruption to traffic flow. f. System Boundary Control: Pre-entry vehicle clearance checks and system takeover and handoff mechanisms for vehicle entry and exit, respectively. g. Platoon Control and Fleet Management h. System Failure Safety Provisions: During a failure, the system shall provide (1) sufficient response time for the driver or vehicle to take over control of the vehicle, or (2) other provisions to safely bring the vehicle to a stop. i. Task Priority Management: Provides a mechanism for prioritizing various control goals.

[0021] In some embodiments, the RSU includes one or more of the following modular configurations, but is not limited to: a. Sensing module that detects the driving environment b. A communication module that communicates with the vehicle, TCUs, and the cloud via wired or wireless media. c. A data processing module that processes data from the sensing and communication module. d. An interface module that communicates between the data processing module and the communication module. e. Adaptive power modules that provide backup redundancy to adjust power supply based on local grid conditions

[0022] In some embodiments, the sensing module includes one or more of the following flow-type sensors, but is not limited to: a. Radar-based sensors that work in conjunction with vision sensors to sense driving environment and vehicle attribute data, including but not limited to: i. LiDAR ii. Microwave radar iii. Ultrasonic radar iv. Millimeter wave radar b. Vision-based sensors that work in conjunction with radar-based sensors to provide driving environment data, including but not limited to: i. Color camera ii. Nighttime infrared camera iii. Night-time thermal camera Satellite navigation systems that cooperate with inertial navigation systems to assist in vehicle location, including but not limited to: i. DGPS ii. BeiDou System d. Inertial navigation systems, including but not limited to inertial reference units, that cooperate with satellite navigation systems to support vehicle position determination. e. Vehicle identification devices, including but not limited to RFID.

[0023] In some embodiments, RSUs are installed and deployed based on functional requirements, environmental factors such as road type, geometry, and safety considerations, including but not limited to: a. Some modules are not necessarily installed in the same physical location as the core module of the RSUs. To archive maximum coverage and eliminate detection blind spots, RSU spacing, deployment, and installation methods may vary depending on road geometry, including but not limited to highways, roadsides, highway on / off ramps, intersections, roadside buildings, bridges, tunnels, roundabouts, transfer stations, parking lots, railroad crossings, and school zones. c. RSUs are located at: i. Fixed location for long-term deployment ii. Mobile platforms for short-term or flexible deployment, including, but not limited to, cars, trucks, and unmanned aerial vehicles (UAVs).

[0024] In some embodiments, RSUs are deployed in special locations and periods requiring additional system coverage, and RSU configurations may vary. Special locations include, but are not limited to: a. Construction area b. Special events such as sporting events, street fairs, outdoor parties, and concerts c. Special weather conditions such as storms and heavy snow

[0025] In some embodiments, the TCCs and TCUs, in conjunction with the RSUs, may have a hierarchical structure that includes, but is not limited to: a. Traffic Control Center (TCC) realizes comprehensive traffic operation optimization, data processing, archiving functions, and provides human operation interface. Based on coverage area, TCC may be further classified into macro TCC, regional TCC, and corridor TCC. b. Traffic Control Units (TCUs) provide highly automated, real-time vehicle control and data processing functions based on pre-installed algorithms. Based on their coverage area, TCUs are further classified into segment TCUs and point TCUs. c. A network of roadside units (RSUs) that receive data flows from connected vehicles, detect traffic conditions, and send targeted commands to vehicles, and point or segment TCUs can be physically coupled or integrated with the RSUs.

[0026] In some embodiments, the cloud-based platform provides information and computing services to a network of RSUs and TCCs / TCUs, including but not limited to: a. Storage as a Service (STaaS) for IRIS's additional storage needs b. Control as a Service (CCaaS), which provides additional control capabilities as a service to IRIS c. Computing as a Service (CaaS) for providing an IRIS entity or group of entities that require additional computing resources d. Sensing as a Service (SEaaS), which provides additional sensing capabilities as a service to IRIS.

[0027] The system and method may include or be integrated with the features and components described in U.S. Provisional Patent Application No. 62 / 626,862, filed February 6, 2018, the contents of which are incorporated herein by reference in their entirety.

[0028] In some embodiments, the system and method provide a virtual traffic light control function. In some such embodiments, a cloud-based traffic light control system is characterized by including roadside sensors, such as sensing devices, control devices, and communication devices. In some embodiments, the sensing components of the RSU are provided on the road (e.g., at an intersection) and detect road vehicular traffic, and the sensing components are connected to a cloud system and upload the information to the cloud system. The cloud system analyzes the sensed information and transmits the information to the vehicles via the communication devices.

[0029] In some embodiments, the system and method provide traffic state estimation functionality. In some embodiments, the cloud system includes traffic state estimation and prediction algorithms. A weighted data fusion approach is applied to estimate traffic states, with data fusion weights depending on the quality of information provided by sensors in the RSUs, TCCs / TCUs, and TOCs. In the event of sensor unavailability, the method estimates traffic states based on forecast and inferential information, ensuring the system provides reliable traffic states even in situations with transmission and / or vehicle shortage challenges.

[0030] In some embodiments, the systems and methods provide fleet maintenance capabilities. In some such embodiments, the cloud system uses traffic state estimation and data fusion methods to support fleet maintenance applications such as remote vehicle diagnostics, intelligent fuel-efficient driving, and intelligent charging / refueling.

[0031] In some embodiments, IRIS is equipped with high performance computing capabilities and allocates computing power to achieve sensing, prediction, planning and decision making, and control, specifically at three levels: a. Microscopic level, typically 1 to 10 milliseconds, such as calculating vehicle control commands b. Mesoscopic level, typically 10 to 1000 milliseconds, such as incident detection and pavement condition notification c. Macroscopic level, such as root computing, typically exceeding 1 second

[0032] In some embodiments, IRIS performs traffic and lane management to facilitate traffic operations and control on various types of roadway facilities, including but not limited to: a. Highways, including but not limited to the following: i. Main lane lane change management ii. Traffic merging / diverging management, such as on-ramps / off-ramps iii. High Occupancy / Toll (HOT) Lanes iv. Dynamic Shoulder Lane v. Express Lane vi. Managing the adoption rate of autonomous vehicles across vehicles with various levels of automation vii. Lane closure management for construction zones, incidents, etc. b. Urban highways, including but not limited to the following: i. Basic Lane Change Management ii. Intersection management iii. Urban Street Lane Closure Management iv. Mixed traffic management to accommodate various modes of transport, including bicycles, pedestrians, and buses

[0033] In some embodiments, IRIS provides additional safety and efficiency measures for vehicle operation and control in adverse weather conditions, including but not limited to: a. High-definition map services provided by local RSUs, without the need for vehicle-mounted sensors, that include lane width, lane approach (left / straight / right), gradient (uphill / downhill slope), arc, and other geometric information. b. Site-specific road weather information provided by RSUs to support TCC / TCU network and cloud services c. Vehicle control algorithms designed for severe weather conditions, supported by site-specific road weather information

[0034] In some embodiments, IRIS includes security, redundancy, and resiliency measures to enhance system reliability, including but not limited to: a. Security measures such as network security and physical equipment security i. Network security measures, such as firewalls at various levels and regular system scans ii. Physical equipment security, including secure hardware installations, access controls, and identification trackers b. System redundancy: Additional hardware and software resources are on standby to compensate for a failed system. c. System Backup and Restore. The IRIS system is backed up at various intervals, from the overall system level to the individual device level. If a failure is detected, recovery is performed on a corresponding scale, restoring to the backup that most closely resembles the original state. d. If a failure is detected, the system fail-over mechanism is activated and a higher level system unit identifies the failure and performance response procedures to replace and / or restore the failed unit.

[0035] Also provided herein are methods employing any of the systems described herein for managing one or more aspects of traffic control, including processes undertaken by individual participants in the system (e.g., drivers, public or private jurisdictions, local or national transportation facilities, government agencies, etc.) as well as collective efforts by one or more participants, either cooperatively or individually.

[0036] In some portions of this specification, embodiments of the invention are described in terms of algorithms and symbolic representations of information operations. These algorithmic descriptions and representations are commonly used by those skilled in the data processing arts to effectively convey the substance of their work to others skilled in the art. While these operations are described functionally, computationally, or logically, they are understood to be implemented by computer programs or equivalent electrical circuits, microcode, or the like. Further, it has proven convenient at times to refer to arrangements of these operations as modules, without loss of generality. The described operations and their associated modules may be embodied in software, firmware, hardware, or any combination thereof.

[0037] Certain steps, operations, or processes described herein may be implemented or performed by one or more hardware or software modules, alone or in combination with other devices. In one embodiment, software modules are implemented with a computer program product comprising a computer-readable medium containing computer program code that can be executed by a computer processor to perform any or all of the steps, operations, or processes described herein.

[0038] Embodiments of the present invention may also relate to apparatus for performing the operations described herein. This apparatus may be specially constructed for the required purposes, and / or may comprise a general-purpose computing device selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored on a non-transitory, tangible, readable record storage medium or any type of medium suitable for storing electronic instructions that is connectable to a computer system bus. Furthermore, any computing system referred to in this specification may include a single processor or may be an architecture employing a multiple processor design for increased computing power.

[0039] Embodiments of the present invention may also relate to products produced by the computing processes described herein. Such products include information obtained by the computing processes, which information is stored on a non-transitory, tangible computer-readable medium, and may include any embodiment of the computer program product or other data combination described herein. [Brief explanation of the drawings]

[0040] [Figure 1]Figure 1 shows examples of OBU components. 101: Communication module: Can transfer data between RSU and OBU. 102: Data collection module: Can collect vehicle dynamic and static state and human generated data. 103: Vehicle control module: Can execute control commands from RSU. In case of damage to the vehicle's control system, it can take over control and safely stop the vehicle. 104: Vehicle and human data. 105: RSU data. [Figure 2] Figure 2 shows an example of the IRIS sensing framework. 201: Vehicles send data collected within their sensing range to RSUs. 202: RSUs collect lane traffic information based on vehicle data on lanes, and RSUs share / broadcast the collected traffic information to vehicles within their range. 203: RSUs collect road incident information from vehicle reports within their coverage range. 204: RSUs in an incident segment send incident information to vehicles within their coverage range. 205: RSUs share / broadcast the collected lane information to segment TCUs within their range. 206: RSUs collect weather information, road information, and incident information from segment TCUs. 207 / 208: RSUs in different segments share information with each other. 209: RSUs send incident information to segment TCUs. 210 / 211: Various segment TCUs share information with each other. 212: Information sharing between RSUs and CAVH cloud. 213: Information sharing between segment TCUs and CAVH cloud. [Figure 3] Figure 3 shows an example of the IRIS prediction framework. 301: Data sources including vehicle sensors, roadside sensors, and cloud. 302: Data fusion module. 303: Prediction module based on learning, statistical, and empirical algorithms. 304: Data output at microscopic, mesoscopic, and macroscopic levels. [Figure 4]Figure 4 shows an example of planning and decision-making functionality. 401: Raw data and processed data for three levels of planning. 402: Planning module for macroscopic, mesoscopic and microscopic level planning. 403: Decision-making module for vehicle control commands. 404: Macroscopic level planning. 405: Mesoscopic level planning. 406: Microscopic level planning. 407: Data input for macroscopic level planning: raw data and processed data for macroscopic level planning. 408: Data input for mesoscopic level planning: raw data and processed data for mesoscopic level planning. 409: Data input for microscopic level planning: raw data and processed data for microscopic level planning. [Figure 5] Figure 5 shows an example of vehicle control flow components. 501: Planning and prediction module sends information to control strategy calculation module. 502: Data fusion module receives calculation results from various sensing devices. 503: The integrated data is sent to communication modules of RSUs. 504: RSUs send control commands to OBUs. [Figure 6] FIG. 6 shows an example of a flow chart for longitudinal control. [Figure 7] FIG. 7 shows an example of a flowchart for lateral control. [Figure 8] FIG. 8 shows an example of a flowchart of fail-safe control. [Figure 9] Figure 9 shows an example of RSU physical components. 901 Communication module. 902 Sensing module. 903 Power supply unit. 904 Interface module: Module for communicating between data processing module and communication module. 905 Data processing module: Module for performing data processing. 909: Physical connection of communication module to data processing module. 910: Physical connection of sensing module to data processing module. 911: Physical connection of data processing module to interface module. 912: Physical connection of interface module to communication module. [Figure 10]Figure 10 shows an example of an internal data flow in an RSU. 1001: Communication module. 1002: Sensing module. 1004: Interface module: A module that communicates between the data processing module and the communication module. 1005: Data processing module. 1006: TCU. 1007: Cloud. 1008: OBU. 1013: Data flow from the communication module to the data processing module. 1014: Data flow from the data processing module to the interface module. 1015: Data flow from the interface module to the communication module. 1016: Data flow from the sensing module to the data processing module. [Figure 11] Figure 11 shows an example of a TCC / TCU network structure. 1101: Overall object and system information provided by a macroscopic TCC to a regional TCC. 1102: Local system and traffic information provided by a regional TCC to a macroscopic TCC. 1103: Object and local information provided by a regional TCC to a corridor TCC. 1104: Corridor system and traffic information provided by a corridor TCC to a regional TCC. 1105: Object and corridor system information provided by a corridor TCC to a segment TCU. 1106: Segment system and traffic information provided by a segment TCU to a corridor TCC. 1107: Object and segment system information provided by a segment TCU to a point TCU. 1108: Point system and traffic information provided by a point TCU to a corridor TCU. 1109: Object and local traffic information provided by a point TCU to an RSU. 1110: RSU status and traffic information provided by the RSU to the point TCU. 1111: Customized traffic information and control commands from the RSU to the vehicle. 1112: Information provided by the vehicle to the RSU. 1113: Services provided by the cloud to the RSU / TCC-TCU network. [Figure 12] FIG. 12 shows an example of the architecture of a cloud system. [Figure 13]Figure 13 shows an example of an IRIS computation flowchart. 1301: Data collected from the RSU, including but not limited to image data, video data, radar data, and onboard unit data. 1302: Data allocation module, allocating computational resources for various data processing. 1303: Computational resource module for real data processing. 1304: GPU (Graphics Processing Unit) mainly for large parallel data. 1305: CPU (Central Processing Unit) mainly for advanced control data. 1306: Prediction module for IRIS prediction function. 1307: Planning module for IRIS planning function. 1308: Decision making for IRIS decision making function. 1309: Data for processing by computational resource allocation. 1310: Processing data for prediction module, computation module, and decision making module. 1311: Results from prediction module to planning module. 1312 Results from planning module to decision making module. [Figure 14] Figure 14 shows an example of a traffic and lane management flowchart. 1401 Lane management related data collected by RSU and OBU. 1402 Traffic information from controlled object and upper level IRIS TCU / TCC network. 1403 Lane management and control command. [Figure 15] Figure 15 shows an example of vehicle control in a severe weather component. 1501: Vehicle status, position, and sensor data. 1502: Comprehensive weather and pavement condition data and vehicle control commands. 1503: Wide-area weather and traffic information obtained by the TCU / TCC network. [Figure 16] Figure 16 shows an example of an IRIS system security design. 1601: Network firewall. 1602: Internet and external services. 1603: Data center for data services, such as data storage and processing. 1604: Local server. 1605: Data transmission flow. [Figure 17]Figure 17 shows an example of IRIS system backup and recovery components: 1701: Cloud for data and other services. 1702: Intranet. 1703: Local storage for backups. 1704: Any IRIS device, i.e., RSU, TCU, or TCC. [Figure 18] FIG. 18 shows an example of a system fault management component. [Figure 19] FIG. 19 shows a cross-sectional view of an example RSU deployment. [Figure 20] FIG. 20 shows a top view of an example RSU deployment. [Figure 21] FIG. 21 shows an example of RSU lane management on a highway segment. [Figure 22] FIG. 22 shows an example of RSU lane management at a typical urban intersection. DETAILED DESCRIPTION OF THE INVENTION

[0041] Examples of this technique are described below, but these are exemplary embodiments and the invention is not limited to these specific examples.

[0042] 1 shows an embodiment of an OBU, which includes a communication module 101, a data collection module 102, and a vehicle control module 103. The data collection module 102 collects vehicle and person related data 104 and transmits it to the RSU via the communication module 101. The OBU can also receive data from the RSU 105 via the communication module 101. Based on the data from the RSU 105, the vehicle control module 103 helps control the vehicle.

[0043] FIG. 2 shows an example of a lane management sensing system and its data flow framework.

[0044] The RSU exchanges information between the vehicle and the road and communicates with the TCUs, including weather information, road condition information, lane traffic information, vehicle information, and incident information.

[0045] Figure 3 shows an example workflow of a basic prediction process for a lane management sensing system and its data flow. In some embodiments, fused multi-source data collected from vehicle sensors, roadside sensors, and the cloud is processed through models, including but not limited to learning models, statistical models, and empirical models. New models, including learning models, statistical models, and empirical models, are then used to predict microscopic and Predictions are made at various levels, including the mesoscopic and macroscopic levels.

[0046] Figure 4 shows an example of the planning and decision-making process in IRIS. Data 401 is fed into planning module 402 according to three planning modules 407, 408, and 409. The three planning sub-modules retrieve and process the corresponding data for their own planning tasks. At macroscopic level 404, route planning and guidance optimization is performed. At mesoscopic level 405, special events, construction zones, deceleration zones, incidents, buffer spaces, and extreme weather are handled. At microscopic level 406, longitudinal and lateral control are generated based on internal algorithms. After calculation and optimization, all planning outputs from these three levels are generated and sent to decision-making module 403 for further processing, including steering, throttle control, and braking.

[0047] Figure 5 shows an example of data flow for an infrastructure automation-based control system. The control system calculates the results from all sensing detectors, performs data fusion, and exchanges information between RSUs and vehicles. The control system includes a) a control strategy calculation module 501, b) a data fusion module 502, c) a communication module (RSU) 503, and d) a communication module (OBU) 504.

[0048] Figure 6 shows an example of a process for longitudinal control of a vehicle. As shown in the figure, the vehicle is monitored by RSUs. If a relevant control threshold (e.g., minimum distance, maximum speed, etc.) is reached, the necessary control algorithm is triggered. The vehicle then follows the new control command and drives accordingly. If the command is not confirmed, a new command is sent to the vehicle.

[0049] Figure 7 shows an example of a process for lateral control of a vehicle. As shown in the figure, the vehicle is monitored by the RSUs. If the relevant control threshold (e.g., lane keeping, lane change, etc.) is reached, the necessary control algorithm is triggered. The vehicle then follows this new control command and drives accordingly. If the command is not confirmed, a new command is sent to the vehicle.

[0050] Figure 8 shows an example of a process for vehicle fail-safe control. As shown in the figure, the vehicle is monitored by RSUs. If an error occurs, the system sends a warning message to alert the driver and takes control of the vehicle. If the driver does not respond or does not have enough time to make a decision, the system sends a control threshold to the vehicle. If the relevant control threshold (e.g., stopping, hitting a safety device, etc.) is reached, the necessary control algorithm is triggered. The vehicle then drives according to the new control command. If the command is not confirmed, a new command is sent to the vehicle.

[0051] 9 shows an example of the physical components of a typical RSU, including a communication module, a sensing module, a power supply unit, an interface module, and a data processing module. The RSU may have any of a variety of modular configurations. For example, in the case of a sensing module, a low-cost RSU may include only a vehicle ID recognition unit for vehicle tracking, while a typical RSU includes various sensors such as LiDAR, a camera, microwave radar, etc.

[0052] Figure 10 shows an example of internal data flow within the RSU. The RSU exchanges data with vehicle OBUs, higher-level TCUs, and the cloud. The data processing module includes two processors: an external object computation module (EOCM) and an AI processing unit. The EOCM is responsible for detecting traffic objects based on input from the sensing module, while the AI processing unit is more focused on the decision-making process.

[0053] Figure 11 shows an example of a TCC / TCU network structure. A macro TCC, which may or may not collaborate with an external TOC, manages a certain number of regional TCCs in its coverage area. Similarly, a regional TCC manages a certain number of corridor TCCs, which manage a certain number of segment TCUs, which manage a certain number of point TCUs, which manage a certain number of RSUs. The RSUs send customized traffic information and control commands to vehicles and receive information provided by vehicles. This network is supported by services provided by the cloud.

[0054] 12 illustrates how an embodiment of a cloud system communicates with sensors in the RSU, TCC / TCU (1201), and TOC via a communication layer (1202). The cloud system includes a cloud infrastructure (1204), a platform (1205), and application services (1206), which also support applications (1203).

[0055] 13 shows an example of data collected from a sensing module 1301, such as image data, video data, and vehicle status data. The data is separated into two groups by a data allocation module 1302: large parallel data and advanced control data. The data allocation module 1302 determines how to allocate data 1309 using computational resources 1303, such as graphics processing units (GPUs) 1304 and central processing units (CPUs) 1305. The processed data 1310 is sent to a prediction module 1306, a planning module 1307, and a decision-making module 1308. The prediction module provides results 1311 to the planning module, and the planning module provides results 1312 to the decision-making module.

[0056] 14 shows how an example of data collected from OBUs and RSUs is provided to the TCU along with object and traffic information 1402 from the higher level IRIS TCC / TCC network. The TCU's lane management module generates lane management and vehicle control instructions 1403 for the vehicle control and lane control modules.

[0057] Figure 15 shows an example of data flow for vehicle control in bad weather. Table 1 below shows an approach for measuring bad weather scenarios. [Table 1] The number of "*" indicates the degree of decline.

[0058] Figure 16 shows an example of IRIS security measures, including network security and physical equipment security. Network security is implemented by firewalls 1601, which complete periodic system scans at various levels. These firewalls protect data transmissions 1605, either between the system and the Internet 1601 or between a data center 1603 and local servers 1604. For physical equipment security, hardware is securely installed, secured with identification trackers, and potentially isolated.

[0059] 17, periodically, IRIS system component 1704 backs up data through firewall 1601 to local storage 1703 on the same intranet 1702. In some embodiments, it also uploads the backup copy through firewall 1601 to cloud 1701, which is logically located on intranet 1702.

[0060] Figure 18 shows an example of a routine check for system failures in the IRIS system. In the event of a failure, the system fail-over mechanism is activated. First, the failure is detected and the failed node is identified. The functionality of the failed node is handed over to the shadow system, and if there are no abnormalities, success feedback is sent back to the higher-level system. Meanwhile, the failed system / subsystem is restarted and / or restored from the most recent backup. If successful, feedback is reported to the higher-level system. If the failure is addressed, the functionality is moved back to the original system.

[0061] Examples of hardware and parameters used in embodiments of the current technology include, but are not limited to: OBU: a) Communication module technical specifications Standards Compliant: IEEE 802.11p-2010 Bandwidth: 10 MHz Data rate: 10 Mbps Antenna diversity CDD transmit diversity Environmental operating range: -40°C to +55°C Frequency band: 5 GHz Doppler spread: 800 km / h Delay spread: 1500ns ·Power: 12 / 24V b) Data Acquisition Module Hardware Technical Specifications Intuitive PC user interface for configuration, tracing, sending, filtering, logging, and other functions High-speed data transfer c) Software Technical Specifications Tachograph driver alerts and remote analysis Real-time CAN BUS statistics ·CO2 emissions report d) Vehicle Control Module Technical Specifications ·Low power consumption Reliable longitudinal and lateral vehicle control RSU design a) A communication module containing three communication channels: Vehicle communications including DSRC / 4G / 5G (e.g., Coda Wireless MK5V2X) o Communication with point TCUs, including wired and wireless communication (e.g., Cabresys fiber optics) Communication with the cloud, including wired and wireless communications with a combined bandwidth of at least 20M b) A data processing module containing two processors: External Object Computation Module (EOCM) Process object detection using data from the sensing module and other required routine computations (e.g., a low-power fully custom ARM / X86-based processor) AI Processing Unit Machine learning Decision-making / planning and forecasting processes c) Interface Module: FPGA-based interface unit An FPGA processor that acts as a bridge between the AI processor and the external object computing module processor, sending instructions to the communication module. RSU deployment a. Deployment location RSU deployment is based on functional requirements and road type. RSUs are used to sense, communicate, control, and provide automation for vehicles on roadways. LiDAR and other sensors (such as loop detectors) require different specialized locations, so some of these may be located separately from the RSU's core processor. Two examples of RSU location deployment types: i. Fixed Location Deployment: The location of this type of RSU is fixed and is routinely used to serve regular roadways with fixed traffic demand. ii. Mobile Deployment. Mobile RSUs can be moved and quickly established in new locations and situations, and are used to serve stochastic and unstable demand, special events, collisions, etc. When an event occurs, these mobile RSUs can move to that location and perform their function. b. Coverage Method RSUs may be connected underground (e.g., by wire). To function properly, RSUs are mounted on poles facing downwards. The pole blades are T-shaped. The roadway lanes requiring CAVH functionality are covered by the sensing and communication devices of the RSUs. There is overlap between the coverage areas of the RSUs to ensure functionality and performance. c. Deployment density The density of deployment depends on the type and requirements of the RSUs. Typically, the minimum distance between two RSUs is determined by the RSU sensor with the smallest coverage range. d. Blind spot response Vehicles can obstruct each other, creating blind spots that cannot be sensed. This is a common problem, especially when vehicles are in close proximity. The solution is to utilize the collaboration of different sensing technologies from both infrastructure-based RSUs and vehicle-based OBUs. The purpose of this type of deployment is to improve traffic conditions and control performance under certain special circumstances. Mobile RSUs can be moved to the deployment location by an agent. In most cases, special RSUs are used temporarily, and mounting poles are not always available. Therefore, these RSUs may be installed on temporary frames, roadside buildings, or even overpasses where the location is appropriate.

[0062] An example of an RSU configuration is shown in FIGS. 19-22. FIG. 19 shows a cross section of an example RSU deployment. FIG. 20 shows a top view of an example RSU deployment. In this road segment, sensing is covered by two types: 901 RSU A: a camera group, which is the most commonly used sensor for object detection, and 902 RSU B: a LiDAR group, which provides a 3D view of objects and higher accuracy. The camera sensor group employs a lower range than LiDAR, e.g., less than 150 m in this particular case, so the camera groups are spaced 150 m apart along the road. Other types of RSUs have fewer requirements for density (e.g., some, such as LiDAR or ultrasonic sensors, may require larger spacing).

[0063] 21 illustrates an example of an RSU lane management configuration for a highway segment. RSU sensing and communication covers each lane of the road segment and performs example lane management functions (denoted by red arrows in the figure). Example lane management functions include, but are not limited to: 1) Changing lanes from one lane to another 2) Merging from an on-ramp 3) Diversion from main roads to off-ramps 4) Inclusion zone management to ensure safety 5) Reversible lane management

[0064] 22 shows an example of a lane management configuration for a typical urban intersection. RSU sensing and communication covers each corner of the intersection and performs example lane management functions (shown in red in the figure). Example lane management functions include, but are not limited to: 1) Changing lanes from one lane to another 2) Traffic management (left-turn lanes) 3) Lane closure management on this leg 4) Bicycle lane management

Claims

1. An autonomous vehicle cloud system (AVCS), comprising: a) a roadside unit (RSU) network having a plurality of roadside units (RSUs); b) Hierarchical Traffic Control Units (TCUs) and Traffic Control Centers (TCCs); c) Vehicle On-Board Unit (OBU) and Vehicle Interface; d) a cloud-based platform that supports the computer functions of the RSU, the TCC, and the TCU and provides information and computing services; and e) Communication module Including, The TCU and the TCC are for collecting and processing traffic information, and the TCU has segment TCUs and point TCUs based on the area it covers, and the TCC is configured to realize traffic operation optimization, data processing and archiving functions, or provide a human operation interface, and the TCC has macroscopic TCCs, regional TCCs, and corridor TCCs based on the area it covers, and the macroscopic TCC manages a certain number of regional TCCs, the regional TCC manages a certain number of corridor TCCs, the corridor TCC manages a certain number of segment TCUs, and the segment TCUs manage a certain number of point TCUs; the cloud-based platform comprises one or more of a sensing subsystem, a prediction subsystem, a planning and decision-making subsystem, and a control subsystem; the communication module communicates with one or more of another autonomous vehicle (AV), the roadside unit (RSU), and the traffic control center / traffic control unit (TCC / TCU), and receives vehicle-specific information from one or more of another autonomous vehicle (AV), the roadside unit (RSU), and the traffic control center / traffic control unit (TCC / TCU), and transmits the vehicle-specific information to a vehicle OBU; the sensing subsystem is configured to acquire data from one or more of other vehicles, the RSU, and the TCC / TCU to generate comprehensive information at a microscopic level; the prediction subsystem is configured to predict individual vehicle behavior and traffic conditions or traffic demands at a microscopic level based on data from one or more of other vehicles, the RSU, and the TCC / TCU; the planning and decision-making subsystem is configured to provide decision instructions to individual vehicles based on data from one or more of other vehicles, the RSU, and the TCC / TCU; the control subsystem is configured to generate or receive time-sensitive control commands for a particular vehicle to perform a driving task based on data from other vehicles and one or more of the RSUs; An autonomous driving vehicle cloud system characterized by:

2. f) Cloud infrastructure; and g) Application Services Subsystem Further provided with the cloud infrastructure comprises a computing module and a storage module; the application services subsystem provides information services and computing services; h) Information Services (IaaS) that provide additional information retrieval functionality as a service to the AVCS system; i) Storage as a Service (STaaS) to meet additional storage needs for the AVCS system; j) Control Services as a Service (CCaaS) that provide additional control functions as a service to the AVCS system; k) Computing as a Service (CaaS) that provides an entity or group of entities to the AVCS system that requires the additional computing resources; l) Sensing Services (SEaaS) that provide additional sensing capabilities as a service to the AVCS system; and / or m) Operation and Maintenance Services (OMaaS) that provide additional operation and maintenance capabilities as a service to said AVCS system.

2. The autonomous vehicle cloud system of claim 1, further comprising:

3. The comprehensive information includes vehicle information, weather information, vehicle attribute data, traffic condition information, road information, and / or incident information. The autonomous driving vehicle cloud system according to claim 1 .

4. The decision instructions at the microscopic level include longitudinal control (car following, acceleration and deceleration) and lateral control (lane keeping, lane changing), The autonomous driving vehicle cloud system according to claim 1 .

5. The time-sensitive control commands include vehicle longitudinal acceleration and velocity, vehicle lateral acceleration and velocity, and vehicle heading and orientation. The autonomous driving vehicle cloud system according to claim 1 .

6. the cloud-based platform providing vehicle-specific information consisting of weather information, pavement conditions, and / or estimated travel times to individual vehicles; The autonomous driving vehicle cloud system according to claim 1 .

7. the cloud-based platform having high computing power is configured to allocate computing power and storage resources to realize one or more functions of sensing, predicting, planning, decision-making, and / or control at microscopic, mesoscopic, and macroscopic levels; The microscopic level is typically in the range of 1 to 10 milliseconds, The mesoscopic level is in the range of 10 to 1000 milliseconds, The macroscopic level is greater than 1 second. The autonomous driving vehicle cloud system according to claim 1 .

8. The cloud-based platform includes: fusing multi-source data collected from one or more other vehicles, the RSU, and the TCC / TCU; Implement learning-based, statistical, and empirical models; optimizing the output of one or more of the sensing subsystem, the prediction subsystem, the planning and decision-making subsystem, and the control subsystem; The autonomous driving vehicle cloud system according to claim 1 .

9. The cloud-based platform provides a virtual traffic light control function using a sensing device and roadside sensors including a control device, a communication device, and a sensing device associated with a cloud system; The sensed information is analyzed in a cloud system and transmitted to the vehicle via communication equipment. The autonomous driving vehicle cloud system according to claim 1 .

10. Fleet maintenance features include remote vehicle diagnostics, intelligent fuel-saving driving, and intelligent charging / refueling. The autonomous driving vehicle cloud system according to claim 1 .

11. The cloud-based platform provides vehicle operation and maintenance services; and / or Providing security, redundancy, and resilience measures; The autonomous driving vehicle cloud system according to claim 1 .

12. The cloud-based platform uses a weighted data fusion approach and traffic condition estimation and forecasting algorithms to estimate traffic conditions based on predicted and estimated information, ensuring that the autonomous vehicle cloud system provides reliable traffic conditions in transmission and / or vehicle shortage situations. The autonomous driving vehicle cloud system according to claim 1 .

13. the cloud-based platform communicates with one or more of other AVs, the RSU, and the TCC / TCU, and receives vehicle-specific information from one or more of the AVs, RSUs, and TCC / TCUs; The autonomous driving vehicle cloud system according to claim 1 .

14. The cloud-based platform provides safety and efficiency measures for vehicle operation and control in adverse weather conditions. The autonomous driving vehicle cloud system according to claim 1 .

15. The cloud-based platform is configured to distribute vehicle-specific driving tasks and functions. The autonomous driving vehicle cloud system according to claim 1 .

Images