Method for updating HD maps in vehicles

Abstract

Methods and systems are provided for managing over-the-air (OTA) updates of a high-definition map used by navigation systems of vehicles of a vehicle management system. Specifically, systems and methods are provided to increase an efficiency of disseminating and distributing HD map updates from the vehicle management system to a plurality of vehicles using a combination of vehicle-to-vehicle communication, infrastructure-to-vehicle communication, and direct OTA updates via a wireless network. A holistic solution comprising various techniques and approaches is described for edge node prioritization and transmission route optimization, based on current and historical driver data, current network topology, and other factors; data preprocessing and compression strategies for reducing transmission time; dynamic reconfiguration based on edge node capability and network conditions; and progressive decentralization of processing and memory resources.

Description

METHOD FOR UPDATING HD MAPS IN VEHICLESTECHNICAL FIELD

[0001] The present description relates to methods and systems for managing over-the-air (OTA) updates of high-definition maps used by vehicle navigation systems.BACKGROUND / SUMMARY

[0002] A vehicle, and in particular, an autonomous vehicle, may include a navigation system that relies on a High-Definition (HD) map of an area in which the vehicle is operating, for vehicle navigation. The HD map may include routes, geographical features, buildings, highways, and the like. Periodically the HD map may be updated at the vehicle, to ensure that the HD map includes the area of operation of the vehicle, and that information presented in the HD map is current. For example, a new version of an HD map may be released, or the vehicle may be navigated to an area not covered by a current HD map of the area, and a new HD map may be requested from a providing service by the vehicle. Additionally, the HD map may be updated to include timesensitive information, such as closed routes and detours, traffic data, and / or weather data. The time-sensitive information may be used by the navigation system to recommend or update a route of the vehicle.

[0003] A new version of an HD map may be transmitted by the vehicle manufacturer to the vehicle via an over-the-air (OTA) update. During an HD map update, a package including the new version may be transmitted from a server of a manufacturer, or from a map provider, to one or more of a plurality of vehicles. The plurality of vehicles may include all vehicles of a vehicle fleet, or a plurality of vehicles operating in a region, or a plurality of vehicles that may be subject to one or more local conditions at a time of the update. For example, during a snow storm, an HD map update of a zone affected by the snow storm may be sent to vehicles operating in the zone. After the new HD map has been downloaded by a vehicle, the new HD map is installed in place of a previous HD map.

[0004] However, due to a size of the HD map, deploying a new version of the HD map to the plurality of vehicles may be time consuming, computationally intensive, and may rely on an available bandwidth for transmission, which may be less than desirable. Additionally, transmittingthe new versions may rely on internet connectivity via a wireless network, which may be inconsistent, of low quality, costly, and / or not be available at certain places and / or times.

[0005] In one embodiment, the problems described above may be addressed by a method for a vehicle management system (VMS) for managing and updating high-definition (HD) maps used by navigation systems of a plurality of vehicles, the method comprising receiving an HD map update; assigning a priority score to the HD map update; in response to the priority score exceeding a threshold priority score, determining a primary geographical area affected by the HD map update and a secondary geographical area affected by the HD map update; identifying a first portion of edge nodes of the VMS located within the primary geographical area, and a second portion of edge nodes of the VMS located within the secondary geographical area; transmitting the HD map update to the first portion of edge nodes, via a wireless network; after transmitting the HD map update to the first portion of edge nodes, transmitting the HD map update to the second portion of edge nodes, via the wireless network; and after transmitting the HD map update to the second portion of edge nodes, transmitting the HD map update to edge nodes of the VMS outside the primary geographical area and secondary geographical area via the wireless network; and in response to the priority score not exceeding the threshold priority score, transmitting the HD map update to a plurality of edge nodes of the VMS via the wireless network.

[0006] It should be understood that the summary above is provided to introduce in simplified form a selection of concepts that are further described in the detailed description. It is not meant to identify key or essential features of the claimed subject matter, the scope of which is defined uniquely by the claims that follow the detailed description. Furthermore, the claimed subject matter is not limited to implementations that solve any disadvantages noted above or in any part of this disclosure.BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Various aspects of this disclosure may be more clearly understood upon reading the following detailed description and upon reference to the drawings in which:

[0008] FIG. 1 is a schematic block diagram of an exemplary vehicle management system, in accordance with one or more embodiments of the present disclosure;

[0009] FIG. 2 is a flowchart that illustrates an exemplary high-level method for a vehicle management system for updating HD maps at a plurality of vehicles, in accordance with one or more embodiments of the present disclosure;

[0010] FIG. 3 is a flowchart that illustrates an exemplary method for assigning a priority to an HD map update, in accordance with one or more embodiments of the present disclosure;

[0011] FIG. 4 is a flowchart that illustrates an exemplary method for receiving (from a second vehicle) and installing an update of vehicle HD map at a first vehicle, in accordance with one or more embodiments of the present disclosure;

[0012] FIG. 5 is a flowchart that illustrates an exemplary method for transmitting an update of vehicle HD map from the second vehicle to the first vehicle, in accordance with one or more embodiments of the present disclosure;

[0013] FIG. 6 is a flowchart that illustrates an exemplary method for an edge node for distributing an update of HD map to a vehicle, in accordance with one or more embodiments of the present disclosure;

[0014] FIG. 7 is a timing diagram showing a wireless transfer of an HD map update package between one or more edge nodes and a vehicle, in accordance with one or more embodiments of the present disclosure;

[0015] FIG. 8A shows an exemplary map of a geographic area including vehicles and edge nodes of a vehicle management system, in accordance with one or more embodiments of the present disclosure; and

[0016] FIG. 8B is an exemplary directed graph that shows connections between the vehicles and edge nodes of FIG. 8A, in accordance with one or more embodiments of the present disclosure.

[0017] The drawings illustrate specific aspects of the described systems and methods. Together with the following description, the drawings demonstrate and explain the structures, methods, and principles described herein. In the drawings, the size of components may be exaggerated or otherwise modified for clarity.DETAILED DESCRIPTION

[0018] The following description relates to systems and methods for managing over-the-air (OTA) updates of High Definition (HD) maps installed at in-vehicle navigation systems. Most modern vehicles today are connected vehicles, equipped with cellular modems, that use a cellularconnection (4G / 5G) to gain access to the internet. The vehicles use Internet connectivity to exchange data with vehicle manufacturers, original equipment manufacturers (OEM), and 3rd party backend systems for enhancing operation of the vehicle, increasing safety, and providing value added services such as entertainment and maintenance to their owner / drivers. The cellular connections may be relied on for performing over-the-air (OTA) updates to software of a vehicle, for example. The OTA updates may include updates to HD maps used by the vehicle navigation systems. The HD map updates may include periodic, general updates, as well as time-sensitive updates such as closed routes, detours, traffic / accident data, weather data, and the like.

[0019] However, the HD map updates may include larger amounts of data than OTA software updates, and unlike OTA updates, the data may change dynamically. As a result, disseminating new versions of HD maps to a plurality of vehicles may be more time consuming, computationally intensive, and more reliant on a wireless connection that may not be stable or available.

[0020] One solution to disseminating and installing HD maps without relying on a wireless network is to use vehicle-to-vehicle (V2V) communication, where a vehicle not including an HD map update may receive the HD map update from a neighboring vehicle rather than via a wireless network. Additionally or alternatively, the HD map updates may be provided via vehicle-to- infrastructure / infrastructure-to-vehicle (V2I / I2V) communication (both referred to herein simply as V2I communication), where the vehicle may download the HD map from an edge node of an edge computing system configured for disseminating HD map updates.

[0021] Edge computing is a computing paradigm in which processing of data transmitted over a network takes place at a computing device, referred to as an edge node, that is physically closer to where the data is being produced or consumed, rather than where the data is transmitted from. An edge node is a computing device that provides a network interface to other nodes of a cluster of client devices in cluster computing. Edge-based computing offers several advantages over cloudbased computing, such as lower access latency, localization, and faster / efficient access to cached information. Modern cellular (4G / 5G) networks offer edge hosting facilities as infrastructure as a service (laaS). An laaS framework provides on-demand / scalable availability of computing, storage, and network resources. Rather than sending data across the Internet to be processed at a data center in a cloud, computing, storage, and network resources are located at edge nodes closer to clients at the edge of a radio network, or in a core network of the operator. Data coming fromand going to the clients can be inspected and / or processed on the edge nodes with near real-time latencies.

[0022] Thus, a vehicle with a less recent version of an HD map may receive an updated version of the HD map from a second vehicle having a more recent version of the HD map via vehicle-to- vehicle (V2V) communication, or from an edge node of a vehicle management system (VMS) via vehicle-to-infrastructure (V2I) communication, rather than receiving the updated version directly from a vehicle management system of a manufacturer of the vehicle via a wireless internet connection. The V2V communication and the V2I communication may be jointly referred to herein as vehicle-to-everything (V2X) communication. By transmitting the HD map updates from a sending vehicle and / or an edge node to a receiving vehicle rather than via a wireless distribution from a central server, HD map updates may be disseminated faster and using less computational resources, while decreasing a reliance on network connectivity and / or availability. As a result, a functioning of a HD map update system may be increased and / or improved.

[0023] However, while the V2X communication may be suitable for OTA software updates, when used to deploy an HD map update to a large body of vehicles, such communication may be more problematic. For example, a receiving vehicle may have to be operating within a close proximity of a transmitting vehicle or edge node for a longer period of time, or download the HD map in batches from a larger number of transmitting vehicles. As a result, the V2X communication may be an effective way to perform HD map updates for some vehicles operating in urban, high traffic areas, but for other vehicles in areas of less traffic, the HD map updates may not be successful, or may not be performed in a timely or predictable manner. In particular, time-sensitive updates may not be distributed to vehicles promptly enough for drivers or autonomous vehicles to avoid hazards such as highway accidents, route closures, dangerous road conditions, etc. An additional problem is that current V2X communication may rely on direct line-of-sight, and may therefore be limited in range (e.g., frequencies used). In rural or sparsely populated areas, the density of vehicles and edge nodes of the VMS may not be sufficient to ensure consistent and reliable updates. Further, as a number of connected vehicles increases in the future, a bandwidth available for V2X communication may become saturated, which may lead to delays in disseminating large HD map updates, for example, during peak traffic times or in densely populated urban areas.

[0024] To address this issue, systems and methods are provided herein to increase an efficiency of disseminating and distributing HD map updates from a vehicle management system to aplurality of vehicles using V2X communication. Specifically, a holistic solution comprising various techniques and approaches is described for edge node prioritization and transmission route optimization, based on current and historical driver data, current network topology, and other factors; data preprocessing and compression strategies for reducing transmission time; dynamic reconfiguration based on edge node capability and network conditions; and progressive decentralization of processing and memory resources.

[0025] Referring now to the figures, FIG. 1 shows a vehicle ecosystem 100, including a vehicle management system 102, a vehicle 130, and an edge node 150. Vehicle management system 102 may be used to manage periodically updating HD maps at one or more of a vehicle 130. For example, a plurality of vehicles 130 may be a member of a vehicle fleet, and HD maps at each vehicle of the vehicle fleet may be periodically updated. Alternatively, HD maps may be updated at a subset of a vehicle fleet, such as, for example, vehicles 130 located within a specific area, or vehicles of a specific make and / or model, or based on different criteria. In various embodiments, vehicle management system 102 may be operated by a manufacturer of vehicle 130 and / or the vehicle fleet. It should be appreciated that as used herein, the word vehicles refers to vehicles 130 that are managed by VMS 102.

[0026] Vehicle 130 may be a car, a bus, a truck, or a different type of machinery or vehicle operated by an operator. Vehicle 130 may be powered by an internal combustion engine, or vehicle 130 may be an electric vehicle powered by an electrical power source, or vehicle 130 may be a hybrid vehicle powered by both an internal combustion engine and an electrical power source. Vehicle 130 may also be a specialized vehicle used in a specific environment, such as, for example, a golf cart or transportation vehicle used in certain areas of a private facility such as an indoor facility. Vehicle 130 may be operated on public and / or private roads and highways, or on a set of tracks or rails (e.g., a train). In general, vehicle 130 may be any type of vehicle operated by an operator. In particular, vehicle 130 may be an autonomous or semi-autonomous vehicle, which may operate without a driver or without a direct involvement of the driver.

[0027] Vehicle management system 102 includes one or more processors 106 configured to execute machine readable instructions stored in non-transitory memory 104. As referred to herein, the memory 104 and other memories described herein may include any non-transitory computer readable medium in which programming instructions are stored. For the purposes of this disclosure, the term “tangible computer readable medium” is expressly defined to include any typeof computer readable storage. The example methods and systems may be implemented using coded instruction (e.g., computer readable instructions) stored on a non-transitory computer readable medium such as a flash memory, a read-only memory (ROM), a random-access memory (RAM), a cache, or any other storage media in which information is stored for any duration (e.g. for extended time periods, permanently, brief instances, for temporarily buffering, and / or for caching of the information). Computer memory of computer readable storage mediums as referenced herein may include volatile and non-volatile or removable and non-removable media for a storage of electronic-formatted information such as computer readable program instructions or modules of computer readable program instructions, data, etc. that may be stand-alone or as part of a computing device. Examples of computer memory may include any other medium which can be used to store the desired electronic format of information and which can be accessed by the processor or processors or at least a portion of a computing device.

[0028] Processor(s) 106 and other processors referred to herein may be any suitable processor, processing unit, or microprocessor, for example. Processor(s) 106 may be a multi -processor system, and, thus, may include one or more additional processors that are identical or similar to each other and that are communicatively coupled via an interconnection bus. Processor(s) 106 may be single core or multi-core, and the programs executed thereon may be configured for parallel or distributed processing. In some embodiments, processor(s) 106 may optionally include individual components that are distributed throughout two or more devices, which may be remotely located and / or configured for coordinated processing. In some embodiments, one or more aspects of processor(s) 106 may be virtualized and executed by remotely-accessible networked computing devices configured in a cloud computing configuration.

[0029] Memory 104 may include an HD map update management module 116, which may manage a dissemination of updates to HD maps at one or more vehicles 130. That is, HD map update management module 116 may store one or more master HD maps 118, and copies of the one or more master HD maps 118 may be transmitted by the HD map update management module 116 to the one or more vehicles 130. For example, HD maps used at the one or more vehicles 130 may be periodically updated via an HD map update. HD map update management module 116 may store and keep track of different HD map updates sent out over time. Additionally, HD map update management module 116 may transmit notifications to the one or more vehicles 130 indicating an availability of an HD map update. Each HD map update may be relevant to a specific HD mapinstalled at the one or more vehicles 130. In various embodiments, HD map update management module 116 may include one or more artificial intelligence (Al) models 119, which may be used to classify, prioritize, rank, or otherwise process HD map updates in preparation for dissemination.

[0030] Vehicle management system 102 may include a communication module 107, which may manage wireless communication between vehicle management system 102 and a communication module 138 of vehicle 130. For example, communication module 107 may communicate with communication module 138 via a wireless network 190. In various embodiments, wireless network 190 may be or include the Internet. For example, an HD map update may be sent from vehicle management system 102 to one or more vehicles 130 over the Internet via a cell phone network. In other embodiments, wireless network 190 may be or include one or more private networks.

[0031] Vehicle management system 102 may be operably / communicatively coupled to a user input device 120 and a display device 124. User input device 120 may comprise one or more of a touchscreen, a keyboard, a mouse, a trackpad, a motion sensing camera, or other device configured to enable a user to interact with and manipulate data within vehicle management system 102. In one example, user input device 120 may enable a user to release an update of an HD map to be installed at vehicle 130, which may be downloaded at vehicle 130 via communication module 138.

[0032] Display device 124 may include one or more display devices utilizing virtually any type of technology. In some embodiments, display device 124 may comprise a computer monitor on which data of vehicle 130, vehicle owner(s) 136, and / or vehicle operator(s) 138 may be displayed. Display device 124 may be combined with processor(s) 106, non-transitory memory 104, and / or user input device 120 in a shared enclosure, or may be peripheral display devices and may comprise a monitor, touchscreen, projector, or other display device known in the art, which may enable a user to and / or interact with various data stored in non-transitory memory 104.

[0033] Vehicle 130 may include an electronic control unit (ECU) 132, such as a digital cockpit ECU, which may control various operations of vehicle 130. ECU 132 may include a processor, which may execute instructions stored in a memory of ECU 132 to implement the various operations. In some embodiments, ECU 132 may be powered by a power storage device of vehicle 130, such as a battery 139. Battery 139 may be a dedicated ECU battery, or battery 139 may be a specified battery, for example, for an input-output controller of ECU 132, whereby power to execute the instructions may be available if power is not available via other power sources of vehicle 130. In various embodiments, battery 139 may supply ECU 132 with sufficient power tooperate when a motor or engine of vehicle 130 is not operating, and a main battery of vehicle 130 is not charged. For example, in some embodiments, battery 139 may supply sufficient power to ECU 132 to receive and install an HD map update when the motor or engine of vehicle 130 is not operating and / or and the main battery of vehicle 130 is not charged.

[0034] Communication module 138 may support wireless communication between vehicle 130 and other vehicles 130 managed by vehicle management system 102. The wireless communication may rely on one or more of various wireless technologies (e.g., radio frequency, infrared, near field communication (NFC), etc.). For example, a wireless connection may be established via a radio frequency (RF) link that supports bidirectional communication, whereby RF signals may be transmitted from a first vehicle 130 to a second vehicle 130 via communication module 138, and / or RF signals may be transmitted from the second vehicle 130 to the first vehicle 130 and received at communication module 138. Communication module 138 may communicate via a wireless local area network (LAN) or wide area network (WAN) using any past, present, or future communication protocol (e.g., BLUETOOTH™, USB 2.0, USB 3.0, etc.).

[0035] Vehicle 130 may include an HD map update system 133 stored in a memory 134 of vehicle 130, which may maintain one or more HD maps 135 installed at vehicle 130 up to date. For example, the HD maps 135 may be installed in an in-vehicle navigation system 140. Navigation system 140 may rely on a global positioning system (GPS) to navigate vehicle 130. Navigation system 140 may be used by a driver to navigate vehicle 130, and / or navigation system 140 may be used by ECU 132 to navigate vehicle 130 directly, without the involvement of a driver, in an autonomous driving mode. Periodically, HD map updates to the HD maps 135 may be transmitted to vehicle 130. The HD map updates may be requested and / or received by an OTA client 136 installed in the memory 134 of vehicle 130. In some embodiments, vehicle 130 of vehicle ecosystem 100 may store HD map version information and HD map updates, which may be transmitted to other vehicles 130. The HD map updates transmitted to vehicle 130 may be stored in memory 134 of the HD map update system 133.

[0036] In some embodiments, the HD map update may be downloaded in chunks, where the chunks may be stored in memory 134. After all the chunks of the HD map update have been downloaded, the HD map update may be installed at navigation system 140. After the HD map update is installed, the HD map update may be removed from memory 134.

[0037] As described in greater detail herein, as an alternative to sending an HD map update from vehicle management system 102 to a plurality of vehicles 130 via wireless network 190, the HD map update may be sent from vehicle management system 102 to one vehicle 130, or a small number of vehicles 130, and the HD map update may be transferred from one vehicle 130 or small number of vehicles 130 to the plurality of vehicles 130 via vehi cl e-to- vehicle (V2V) communication. By using the V2V communication, a robustness of vehicle management system 102 to external factors may be increased. For example, a power outage may make vehicle management system 102 unavailable to vehicle 130, whereby vehicle 130 may still be able to perform HD map updates via the V2V communication using HD map update system 133, such that the HD map update may not be affected by the power outage.

[0038] Additionally or alternatively, the HD map update may be sent from vehicle management system 102 to edge node 150. The HD map update may be received by an HD map update system 153 stored in a memory 154 of edge node 150, which may be similar to HD map update system 133 of vehicle 130. HD map update system 153 may send the HD map update to one or more vehicles 130. In this way, edge node 150 may be used to disseminate HD map updates to vehicles 130 in circumstances where the V2V communication is less efficient and / or less desirable. HD map update system 153 may also store one or more HD maps 155 at edge node 150.

[0039] Edge node 150 may be an edge node of an edge computing system used to service the one or more vehicles 130. By connecting to edge node 150 rather than vehicle management system 102, vehicle 130 may avail of greater resources, including processing power and memory. Edge node 150 may include one or more processors 152, which may carry out instructions stored in a memory 154. The one or more processors 152 may be more powerful than processors available at vehicle 130, whereby a greater amount of computation may be performed at edge node 150 than at vehicle 130.

[0040] In various embodiments, edge node 150 may be configured to take advantage of an edge laaS framework 160. Via the laaS framework 160, edge node 150 may take advantage of on- demand computing, storage, and networking resources available on a cloud 192. For example, laaS framework 160 may provide on-demand scaling and pay-per-use pricing, which may help avoid overprovisioning, thereby generating savings in operational and / or capital expenditures. Further, the greater resources may include security resources that may not be available at vehicle 130. For example, edge node 150 may include a plurality of network security components 162. Networksecurity components 162 may be commercially available components that include network security functions, such as a full-scale firewall / IDPS, a secure web gateway, a secure DNS, an access broker, and / or other components.

[0041] For example, vehicle management system 102 may transmit an update to a HD map to edge node 150 via wireless network 190. Vehicle management system 102 may packetize the HD map into a plurality of data packets to transmit to vehicle 130, in accordance with various strategies described in greater detail below in reference to methods 2-5. Edge node 150 may transmit the data packets to vehicle 130. At vehicle 130, the data packets may be received by a communication module 164 of edge node 150. The data packets may be converted into the HD map, which may be installed at vehicle 130 (e.g., by HD map update system 133). Prior to transmitting the data packets to vehicle 130, the data packets may be analyzed at edge node 150 to determine whether a threat may be included in the data packets using one or more of the network security components 162. In a first condition, the data packets may be free of any threats. In response to the network security components not detecting any threats in the data packets, the data packets may be transmitted to vehicle 130. Alternatively, in a second condition, a threat may have been introduced into the data packets. If the threat is detected by network security components 162, the data packets may not be transmitted to vehicle 130.

[0042] Vehicle 130 may discover an edge node 150 to connect to of a plurality of edge nodes 150 of VMS 102. In various embodiments, vehicle 130 may attempt to connect to an edge node 150 that is closest to vehicle 130, using navigation system 140. In other embodiments, vehicle 130 may connect vehicle 130 to a different edge node that is not the closest edge node to vehicle 130, for example, if the different edge node has a stronger signal (e.g., where a signal from the closest edge node may be obstructed). Vehicle 130 may establish a connection via a process that may be repeated periodically in the background, to allow vehicle 130 to remain connected to the nearest edge node while in motion. The selection and prioritization of edge nodes and / or vehicles for distributing an HD map update to a vehicle is described in greater detail below in reference to FIG. 2.

[0043] Referring now to FIG. 2, an exemplary method 200 is shown for disseminating an HD map update from a vehicle management system such as VMS 102 of FIG. 1, to a plurality of vehicles such as vehicle 130, using a combination of V2V communication and V2I communication with a plurality of edge nodes of the VMS, such as edge node 150. Method 200 may be carried out by anHD map update management module, such as HD map update management module 116 of VMS 102, based on instructions stored in a memory of the VMS (e.g., memory 104) that are executed on a processor of the VMS, such as processor 106 of VMS 102. It should be appreciated that in some examples, one or more steps of method 200 may be executed in a different order than described below.

[0044] Method 200 begins at 202, where method 200 includes receiving an HD map update. Generation of the HD map update may include aggregating and analyzing data such as satellite imagery, aerial photography, LIDAR scans, ground-level photographs, real-time data streams from vehicles and / or sensors in elements of road infrastructure, public reports of road condi tions / changes, and / or other sources. The data may be processed by the VMS to identify changes in the environment that impact an accuracy of existing HD maps. In some embodiments, the processing may include detection of new road features, changes in roads, updated traffic patterns, or other alternations in map data. Changes to a physical environment like construction projects or natural events that affect a network of roads may also be detected.

[0045] The generated HD map update may comprise various components, including high resolution images that represent terrain and landmarks, precise definitions of roads, lane markings, pedestrian crossings and other road infrastructure, text annotations like street names, traffic signs (regulations), points of interests, etc. In some embodiments, the HD map update may include instructions for adjusting an existing map (e.g., how a map at a vehicle should be modified to reflect the changes). In various embodiments, the generated HD map update may be validated before distribution. In some examples, the HD map update may be generated by the VMS, while in other examples, the HD map update may be generated by a different entity (e.g., a map provider) than the VMS. Once the HD map update has been generated, the HD map update may be staged for distribution at a predetermined location in a memory of the VMS, making it available for dissemination, or the HD map update may be submitted to the VMS in a different manner.

[0046] At 206, method 200 includes assigning a priority score to the HD map update based on a set of predefined heuristics. Prioritizing the HD map update based on the predefined set of heuristics is described in reference to FIG. 3.

[0047] Referring briefly to FIG. 3, an exemplary method 300 for assigning a priority score to the HD map update based on a set of predefined heuristics is described. Method 300 may also be carried out by an HD map update management module, such as HD map update managementmodule 116 of VMS 102, based on instructions stored in a memory of the VMS (e.g., memory 104) that are executed on a processor of the VMS, such as processor 106 of VMS 102.

[0048] At 302, method 300 includes receiving the HD map update. At 304, method 300 includes extracting relevant features from the HD map update, such as changes in road geometry, traffic disruptions, temporary obstructions and / or anomalies, weather and climatic disruptions, and the like. The features may be predefined in accordance with various categories or a hierarchical taxonomy. For example, a first category may include road closures due to natural disasters, weather, or environmental conditions; a second category may include road detours due to construction or other human activities; a third category may include congestion-related map updates; and so on. In various examples, the features may be extracted by a model (e.g., of Al models 119), such as a machine learning model. For example, the model may take as input a first image of an HD map, and a second image of an updated HD map, and the model may extract features of the updated HD map that are different from the HD map. The model may also classify the features to the various categories.

[0049] At 306, method 300 includes evaluating an impact of each extracted feature and assigning an urgency score for each feature based on its assessed importance and urgency. In various embodiments, the urgency score may be calculated by a second Al model based on a severity or extent of the extracted feature. For example, the second Al model may be a rules-based model, such as a decision tree, which may assign urgency scores that are based on a size of an affected geographical area, an amount of danger or risk to vehicles imposed by the extracted feature, and / or other criteria. The urgency score may also be based on factors such as a number and / or proximity of vehicles to the affected geographical area, current traffic conditions in the affected area, and the potential for alternative route availability. For example, if very little traffic is travelling on a rural road that is closed, a lower urgency score may be applied to the road closure than if a large amount of traffic is travelling on the rural road.

[0050] At 308, method 300 optionally includes assigning predefined weights to each extracted feature, using a lookup table, where the predefined weights reflect a relative importance of different types of features. For example, in situations where the HD map update includes a plurality of extracted features that affect vehicles in different areas or in different ways, each extracted feature may be assigned a weight based on a degree to which the vehicles are affected in each area / way.

[0051] At 310, method 300 includes aggregating the urgency scores to calculate an overal priority score for the HD map update. In some embodiments, a weighted aggregation of the scores of each extracted feature may be performed using the weights assigned at 308.

[0052] Returning to FIG. 2, at 208, method 200 includes determining whether the priority score is above a threshold priority score. If at 208 it is determined that the priority score is not above the threshold priority score, it may be inferred that the update is not time-sensitive, and method 200 proceeds to 230. At 230, method 200 includes disseminating the HD map update to all nonprioritized edge nodes, which may be all edge nodes in the event that the priority score is below the threshold priority score. Alternatively, if at 208 it is determined that the priority score is above the threshold priority score, it may be inferred that the update is time-sensitive, whereby method 200 proceeds to 210.

[0053] At 210, method 200 includes determining primary and secondary affected geographic areas (also referred to herein as primary and secondary affected areas, respectively) of the map update. To determine the primary and secondary affected areas, the system may perform a contextual evaluation of how each specific change / feature included in the HD map update affects adjacent areas and an overall network of routes within the HD map. For example, a closure on a major highway may have big impact on traffic flow in surrounding areas. The primary affected area may be defined as an area where the changes to the existing HD map have occurred, where updates should be sent out with a higher priority. The secondary affected area may be defined as an area which may not be directly impacted by the changes, but may be indirectly affected, for example, by redirected traffic, where updates should be sent out with a lower priority. In some examples, the HD map update may have boundaries that correspond to an existing HD map, where the HD map update is used to overwrite map data of the existing HD map, including portions where map data is unchanged. In other examples, the HD map update may have boundaries that correspond to a single portion of the HD map, where the HD map update is used to adjust map data of the single portion, and not adjust map data of areas of the existing HD map outside the single portion.

[0054] At 212, method 200 includes identifying a first portion of edge nodes included in the primary affected area. The first portion of edge nodes may be edge nodes located in physical infrastructure located within the primary affected area. In some embodiments, the first portion of edge nodes may be identified by comparing geographical coordinates of the edge nodes, for example, retrieved from a lookup table, with the boundaries of the primary affected area. In somecases, the first portion of edge nodes may be defined as having a distance from a center point of the primary affected area that is less than a threshold distance.

[0055] The edge nodes may include fixed (e.g., static) infrastructure such as road units, traffic signal controllers, sensors, cameras, cellular towers. The first portion may be updated with availability data and current network connectivity status. Different coverage areas may be defined based on the edge nodes, where each coverage area may have at least one edge node capable of serving as a primary point of communication for the coverage area. Coverage gaps may also be identified. In some examples, a vehicle density analysis across different coverage areas and times of day may be performed to identify patterns of high and low vehicle activity. In some examples, edge node utilization trends may be analyzed to predict nodes that might be overwhelmed in the future, to determine where additional edge nodes may be placed in the future. Strategic nodes of the network may also be identified and ranked, where the ranking may be based on features such as coverage area, historical and predicted vehicle densities, reliability of connection, etc., and other edge nodes could be identified as backup nodes. The nodes list may be continuously updated to reflect changes in the network, such as new nodes being added, existing nodes going offline, etc.

[0056] At 214, method 200 includes sending a request to each edge node of the first portion of edge nodes for location, connectivity, and resource data of vehicles located within a range of the edge node. Each edge node may in turn broadcast a request for such data to vehicles, and one or more vehicles located within the range may respond to the edge node with the location, connectivity, and resource data of the one or more vehicles. When the one or more vehicles respond, the edge node may transmit the location, connectivity, and resource data of the vehicles located within a range of the edge node to the HD map update system.

[0057] The connectivity data may include connectivity status, meaning, information on all current connections of the vehicle, including bandwidth, latency, quality, etc. The connectivity data may include data on a traffic flow managed or monitored by the edge node, which may provide insights into vehicle densities, speeds, and patterns in the vicinity of a relevant edge node / vehicle. The connectivity data may also include information about communication methods with other vehicles and infrastructure, such as supported protocols, bandwidth, current communication capabilities, etc. assessing the current state of network connectivity between the edge nodes, including bandwidth availability, signal strength, and potential interference or congestion within thenetwork. This may include analyzing real-time data from the network infrastructure and feedback (data) from vehicles and edge nodes.

[0058] The resource data may include diagnostic data of the vehicle with respect to memory and processing resources. A capability of each edge node to handle additional data transmission tasks without compromising its operational efficiency or the quality of service for existing functions may be determined.

[0059] In some cases, real-time environmental data collected by sensors of the edge nodes / vehicles may also be transmitted, such as weather conditions, road surface status (wet, icy, dry, etc.), and visibility, which can impact driving conditions and the relevance of map updates. The real-time environmental data may additionally include vehicle-related data, such as vehicle telemetry data (speed, acceleration, gyroscope, braking patterns, direction and other data that can be used to detect patterns and anomalies), location and trajectory based on route planning and navigation systems (useful for assessing vehicles approaching affected areas, etc.), and other information from onboard sensors, such as cameras, LIDAR, and radar, which can provide additional insights into road conditions, nearby vehicle proximity, and potential obstacles or hazards. An amount of the data transmitted may vary, depending on available resources. The vehicle-related data may include information regarding how the vehicle's navigation system is using the current HD map, including any discrepancies or issues encountered, which may inform the prioritization and content of updates.

[0060] In some examples, a data preprocessing stage may be performed on the received data, where a neural network that performs pattern recognition, such as a convolutional neural network (CNN) and / or a recurrent neural network (RNN), could be used to analyze spatial and temporal data patterns from vehicle movements and sensor inputs. Additionally, unsupervised learning algorithms could be used to cluster data points and identify common traffic patterns, road usage trends, and recurrent environmental conditions affecting traffic flow.

[0061] In some embodiments, a hybrid approach for optimal efficiency and responsiveness in receiving the geolocations of the vehicles may be used, where the data may be automatically transmitted from the vehicles for continous monitoring and real-time responsiveness, and may additionally be requested from the vehicles by the VMS. That is, the vehicles may send geolocation and operational data to nearby edge nodes at regular intervals, so that the system can continue to maintain an up-to-date overview of vehicle locations. Additionally or alternatively, a vehicle maytransmit the location data to the edge nodes, for example, when an anomaly or unexpected event is detected by the vehicle. Further, the data may be requested from one or more vehicles by the edge nodes, where a decision to request the data may be made based on limited data received automatically. In other words, the VMS may issue targeted requests for additional data from specific vehicles or edge nodes.

[0062] At 216, method 200 includes ranking the first portion of edge nodes based on a number of vehicles that respond to each edge node. Edge nodes that receive responses from a greater number of vehicles receive a higher ranking than edge nodes that receive responses from a lesser number of vehicles.

[0063] At 218, method 200 includes identifying a second portion of edge nodes included in the secondary affected area. The second portion of edge nodes may be edge nodes located in physical infrastructure located within the second affected area. In some embodiments, the second portion of edge nodes may be identified by comparing geographical coordinates of the second portion of edge nodes with the boundaries of the second affected area, or with a center point of the primary affected area, as described above.

[0064] At 220, method 200 includes sending a request to each edge node of the second portion of edge nodes for location, connectivity, resource, and trip destination data of vehicles located within a range of the edge node. Each edge node of the second portion of edge nodes may in turn broadcast a request for such data to vehicles, and one or more vehicles located within the range may respond to the edge node of second portion of edge nodes with the location, connectivity, resource, and trip destination data of the one or more vehicles. When the one or more vehicles respond, the edge node may transmit the location, connectivity, resource, and trip destination data of the vehicles located within a range of the edge node of the second portion of edge nodes to the VMS.

[0065] At 222, method 200 includes ranking the edge nodes of the second portion of edge nodes based on a number of vehicles that respond to each edge node of the second portion of edge nodes whose location and trip destination data indicate travel towards or to a location within the primary affected area. Edge nodes that receive responses from a greater number of vehicles receive a higher ranking than edge nodes that receive responses from a lesser number of vehicles.

[0066] At 224, method 200 includes ranking vehicles that respond to the edge nodes of the primary and secondary affected areas based on a distance to a closest edge node, connectivity, and available resources of each vehicle. For vehicles in the primary affected area, the closest edge node may bean edge node of the first portion of edge nodes. For vehicles in the secondary affected area, the closest edge node may be an edge node of the second portion of edge nodes. Further, in some embodiments, the vehicles that respond to the edge nodes of the primary and secondary affected areas may be ranked based on a distance of each vehicle to a closest edge node in a direction of travel of the respective vehicle, in a descending order. That is, vehicles that are farther from a closest edge node may receive a higher ranking than vehicles that are closer to the closest edge node. The ranking may be performed taking into consideration all of the distance to the closest edge node, connectivity, and available resources of each vehicle. In various embodiments, the ranking may be performed by an Al model (e.g., of the Al models 119 of FIG. 1), where the Al model may take the distance to the closest edge node, connectivity, and available resources of a vehicle as input, and output the ranking. In some examples, the Al model may be a rules-based model or a statistical model generated based on historical vehicle data and human expertise. In other examples, the Al model may be a ML model trained on historical vehicle data, for example.

[0067] At 226, method 200 includes transmitting the HD map update to the first portion of edge nodes, with instructions to transmit the HD map update to vehicles within a range of edge nodes of the first portion of edge nodes. The HD map update may be transmitted to each edge node of the first portion of edge nodes in an order corresponding to a rank of the edge node. For example, the HD map update may be transmitted to a first edge node of the first portion of edge nodes having a highest rank, due to having a greatest number of vehicles respond to the first edge node; the HD map update may then be transmitted to a second edge node of the first portion of edge nodes having a second highest rank, due to having a second greatest number of vehicles respond to the second edge node; and so on for each of the edge nodes of the first portion of edge nodes. Each edge node of the first portion of edge nodes may transmit the HD map update indiscriminately to vehicles within a threshold range of the respective edge node.

[0068] After transmitting the HD map update to the first portion of edge nodes, method 200 may include transmitting the HD map update directly from the VMS (e.g., meaning, not via an edge node) to a selected number of top-ranking vehicles that responded to an edge node of the first portion of edge nodes. The selected number of top-ranking vehicles may be based on a distance from the top-ranking vehicles to the closest edge nodes of the top-ranking vehicles, as well as a connectivity and resource availability of a respective vehicle. In various embodiments, the selectednumber of top-ranking vehicles may include top-ranking vehicles with distances to a closest edge node that are greater than a threshold distance.

[0069] For example, a farthest vehicle responding to a first edge node of the first portion of edge nodes may be above the threshold distance from the first edge node, and may have sufficient connectivity, processing power, and memory to retransmit the HD map update, whereby the VMS may transmit the HD map update directly from the VMS to the farthest vehicle (e.g., after sending the HD map update to the first edge node). A second farthest vehicle responding to a second edge node of the first portion of edge nodes, which may be the same as the first edge node, may be above the threshold distance from the second edge node, but may not have sufficient connectivity, processing power, and / or memory to retransmit the HD map update, whereby the VMS may not transmit the HD map update directly from the VMS to the second farthest vehicle. A third farthest vehicle responding to a third edge node of the first portion of edge nodes, which may be the same as the first edge node and / or second edge node, may not be above the threshold distance from the third edge node, whereby the VMS may not transmit the HD map update directly from the VMS to the third farthest vehicle.

[0070] In various embodiments, a model-driven approach may be taken to optimize an efficiency of the transmission and distribution of data by leveraging all communication channels and available resources, strategically prioritizing a transmission sequence of updates to ensure quick and widespread coverage with minimal bandwidth consumption. The model-driven approach may rely on various Al or ML models (e.g., Al models 119).

[0071] For example, the various Al or ML models may include a geographic segmentation model that divides the geographic area into segments based on vehicular density, network capacity, and proximity to edge nodes. An Al or ML model may be used that takes as input an HD map, realtime traffic data, edge node locations, and network bandwidth availability, and outputs a segmented HD map identifying priority areas for initial update dissemination.

[0072] The various Al or ML models may include a priority -queue management model, which may rank vehicles and edge nodes within each segment by their potential to propagate the update efficiently, based on connectivity strength, vehicle mobility patterns, and data transmission capabilities of each vehicle / edge node. For example, the priority-queue management model may take as input vehicle and edge node telemetry data, transmission capabilities, connectivity strength,and a segmented geographic map, and may output a prioritized list of transmission sources and targets for each dissemination stage.

[0073] The various Al or ML models may include a bandwidth allocation model that dynamically allocates bandwidth resources across the network, prioritizing initial transmissions to areas and nodes with the highest propagation potential based on an output of the priority-queue management model. For example, such a bandwidth allocation model may take as input current and predicted network load, priority queue of transmission sources and targets, and geographic segmentation, and output an optimized bandwidth allocation plan for each stage of dissemination.

[0074] The various Al or ML models may include a dissemination control model that coordinates the staged dissemination process, initiating transmissions according to the priority queue and bandwidth allocation plan, and adapting to changes in network conditions in real-time. For example, the dissemination control model may take as input a priority queue, the optimized bandwidth allocation plan, and real-time network condition data, and output scheduled transmissions for each stage, with adjustments made as needed to optimize propagation speed and efficiency.

[0075] With respect to training and adaptation, the geographic segmentation model, the priorityqueue management model, the bandwidth allocation model, and the dissemination control model may be trained on historical data sets featuring diverse dissemination scenarios, enabling them to learn and adapt to varying conditions. Machine learning techniques may be employed to refine algorithmic decisions based on outcomes, optimizing the efficiency of future updates. By using these models and algorithms, the system may ensure that HD map updates are propagated more efficiently than traditional methods, such as direct transmission from a central cell tower. This staged approach, coupled with smart bandwidth management and dynamic prioritization, may reduce the time required to achieve comprehensive coverage across the geographic area.

[0076] Additionally, transmission of the HD map update to the edge nodes and vehicles may include determining an optimal chunk size and compression strategy for the HD map update data, based on network conditions and edge node capabilities. The selected chunk size may be adapted to different bandwidth availabilities and latency expectations, possibly resulting in a range of chunk sizes rather than a single chunk size. Compression may also include changing a resolution of the map update. The chunk size determination may be performed accounting for the lowestcommon denominator in device processing power and storage capacity, to ensure compatibility across the board.

[0077] The optimization of chunk handling and distribution of updates may select between direct over-the-air, vehicle-to-vehicle, and vehicle-to-infrastructure communication methods, based on real-time network conditions, vehicle locations, and update priorities, ensuring the rapid and efficient dissemination of time-sensitive updates. Further, a chunk management engine may be used to divide the update into optimized chunks for distribution, and identify network paths for the chunks including local caches for edge nodes and vehicles in the network. Current and predicted conditions, including network availability, vehicle locations, and update priority, may be used to choose a most effective distribution method(s) for each chunk and the target of the chunk (e.g., final vehicle, or an edge node in the network). Chunks may be distributed according to the selected method(s), with distribution paths and methods being dynamically updated in real-time based on feedback, such as vehicle movement, changing network conditions, chunk delivery statuses, and new updates / chunks for distribution. As a result, bandwidth usage may be predicted and optimized to enhance update speed and reliability. Additionally, different types of optimization strategies may be used, such as minimizing data transmission time, optimizing resource utilization, and ensuring timely delivery of high-priority updates.

[0078] At 228, method 200 includes transmitting the HD map update to the second portion of edge nodes, with instructions to transmit the HD map update to vehicles within a range of edge nodes of the second portion of edge nodes. The HD map update may be transmitted to each edge node of the second portion of edge nodes in an order corresponding to a rank of the edge node. For example, the HD map update may be transmitted to a first edge node of the second portion of edge nodes having a highest rank, due to having a greatest number of vehicles respond to the first edge node; the HD map update may then be transmitted to a second edge node of the second portion of edge nodes having a second highest rank, due to having a second greatest number of vehicles respond to the second edge node; and so on for each of the edge nodes of the second portion of edge nodes. Each edge node of the second portion of edge nodes may transmit the HD map update indiscriminately to vehicles within a threshold range of the respective edge node. In some cases, the vehicles may transmit an updated HD map or HD map update to an edge node, and the edge node may be updated based on the updated HD map or HD map update.

[0079] After transmitting the HD map update to the second portion of edge nodes, method 200 may include transmitting the HD map update directly from the VMS (e.g., meaning, not via an edge node) to a selected number of top-ranking vehicles that responded to an edge node of the second portion of edge nodes, as described above with respect to the first portion of edge nodes.

[0080] By sending the HD map update to the edge nodes of the first and second portions of edge nodes, and also to select vehicles outside of a threshold distance to a closest edge node of either of the first or second portions of edge nodes, an efficiency of propagation of the HD map update to vehicles within the primary affected area and to vehicles within the secondary affected area may be increased. When a vehicle receives the HD map update, the vehicle may re-transmit the HD map update to other vehicles within a threshold proximity of the vehicle, in accordance with methods described below in reference to FIGS. 4-7. As a result, the HD map update may be transferred to a greater number of vehicles of the first and second portions of edge nodes, or all of the vehicles of the first and second portions of edge nodes, in a shortest amount of time. In this way, the dissemination of the prioritized HD map update to vehicles within the primary and secondary affected areas may be distributed between edge nodes and vehicles in an even and balanced manner.

[0081] At 230, method 200 includes transmitting the HD map update to non-prioritized edge nodes, meaning, a plurality of edge nodes included in an area of the HD map that are not included in the first and second portions of edge nodes. In the case where the priority score of the HD map update is less than the threshold priority score, and the primary and secondary affected areas are not determined in accordance with method 200, the plurality of edge nodes included in the area of the HD map may include all the edge nodes of the VMS. The edge nodes may retransmit the HD map update to vehicles within or entering within a threshold proximity of each respective edge node, and the vehicles receiving the HD map update may in turn transmit the HD map update to other vehicles (as described herein), until all of the vehicles managed by the VMS have been updated. In some examples, the HD map update may be transmitted directly to one or more vehicles, to decrease a propagation and dissemination time of the HD map update throughout the area of the HD map.

[0082] As described above the HD map update may be transmitted via chunks to target vehicles. The chunks may be routed to the target vehicles via a combination of V2V communication, V2I communication, and transmission via a wireless network such as network 190 of FIG. 1. Forexample, the HD map update may be first transmitted to an edge node via the wireless network in a first stage. The HD map update may be transmitted from the edge node to one or more vehicles of the target vehicles within a proximity of the edge node using V2I communication, in a second stage. The HD map update may then be transmitted from the one or more vehicles to other vehicles outside the proximity of the edge node using V2V communication, in a third stage. Additionally or alternatively, the HD map update may be transmitted to a selected subset of target vehicles via the wireless network, to increase a speed and efficiency of propagation of the HD map update to the target vehicles. The selected subset of target vehicles may transmit the HD map update to other target vehicles, and / or to edge nodes for further transmission.

[0083] FIGS. 4, 5, and 6 describe methods for transmitting the chunks in the different scenarios mentioned above. Specifically, FIG. 4 describes a first method to be performed by a first vehicle to receive chunks of an HD map update from a second vehicle, or from an edge node. FIG. 5 describes a second, reciprocal method to be performed by the second vehicle or edge node to transmit the chunks of the HD map update to the first vehicle. FIG. 6 describes a third method to be performed by a vehicle for selecting an edge node to connect to and connecting to the edge node to download the HD map update.

[0084] Referring now to FIG. 4, an exemplary method 400 is shown for performing an HD map update of HD map installed at a first vehicle, such as vehicle 130 of FIG. 1, where the HD map update is received at the first vehicle from a second vehicle located near the vehicle that has a more recent version of the HD map than the first vehicle. By performing the HD map update between the first vehicle and the second vehicle rather than between a vehicle management system (e.g., vehicle management system 102) and the vehicle, and efficiency of the HD map update may be increased. Method 400 may be executed by a ECU of the first vehicle, such as ECU 132 of vehicle 130 of FIG. 1.

[0085] Method 400 begins at 402, where method 400 includes estimating and measuring vehicle operating conditions of the first vehicle. Vehicle operating conditions may be estimated based on one or more outputs of various sensors of the first vehicle (e.g., such as an oil temperature sensor, engine velocity or wheel velocity sensor, torque sensor, pressure sensor, etc). Vehicle operating conditions may include engine velocity and load, vehicle velocity, transmission oil temperature, exhaust gas flow rate, mass air flow rate, coolant temperature, coolant flow rate, engine oil pressures (e.g., oil gallery pressures), operating modes of one or more intake valves and / or exhaustvalves, electric motor velocity, battery charge, engine torque output, vehicle wheel torque, etc., which may be controlled by one or more HD maps installed at the first vehicle. Estimating and / or measuring vehicle operating conditions may include determining whether the one or more HD maps installed at the first vehicle are up to date, and / or whether updated versions of the one or more HD maps are available. For example, estimating and / or measuring vehicle operating conditions may include sending a request to the vehicle management system to determine whether the updated versions of the one or more HD maps are available. Additionally, estimating and / or measuring the vehicle operating conditions may include determining whether the vehicle is being powered by an engine or an electric motor. Estimating and / or measuring the vehicle operating conditions may include determining whether the first vehicle is being operated autonomously, meaning, without active involvement of a driver. Additionally, estimating and measuring vehicle operating conditions of the vehicle may include initializing an OTA client of the vehicle (e.g., OTA client 136). For example, the OTA client may be configured using a configuration file, which may be read during initialization. Using the configuration file, various default values may be established, one or more sanity checks may be performed, etc.

[0086] At 404, method 400 optionally includes determining whether an HD map update notification is received from the vehicle management system, via a wireless network (e.g., wireless network 190). In some embodiments, the vehicle management system may notify the first vehicle that an HD map update has been installed at one or more vehicles (e.g., of a vehicle fleet) and is available for downloading from the one or more vehicles using V2V communication. In other embodiments, the HD map update notification may not be sent, and a determination of whether an HD map update is available may be made during the V2V communication, described below at steps 410-426.

[0087] If the notification of an HD map update is not received by the first vehicle at 404, method 400 proceeds to 406. At 406, method 400 optionally includes continuing operation of first vehicle until an HD map update notification is received, and method 400 proceeds back to 404.

[0088] If the HD map update notification is received by the first vehicle at 404, method 400 proceeds to 408. At 408, method 400 includes broadcasting a request for the HD map update to other vehicles in a vicinity of the first vehicle.

[0089] In various embodiments, the request for the HD map update may be broadcasted on a predetermined radio frequency (RF). Specifically, an RF signal may be generated by the firstvehicle, and the RF signal may be received at one or more second vehicles within a threshold distance of the first vehicle, where the threshold distance may be based on a strength of the RF signal. If a second vehicle is outside the threshold distance, the RF signal may not be received at the second vehicle. If the second vehicle is inside the threshold distance, the RF signal may be received at the second vehicle. Alternatively, in some embodiments, the request for the HD map update may not be broadcasted via the predetermined radio frequency, and the request may be wirelessly streamed by the streaming service over a network and / or via the Internet, such as, for example, via a cell phone network.

[0090] The RF signal (or wireless stream) including the request may be received from the first vehicle by the second vehicle. In various embodiments, the second vehicle may be listening at the predetermined radio frequency for other vehicles in a vicinity of the second vehicle, or may be listening for a predetermined protocol of the wireless stream.

[0091] In some embodiments, the request for the HD map update broadcasted by the first vehicle may include data regarding one or more specific HD maps installed at the first vehicle for which the HD map update is requested. For example, the first vehicle may receive a notification that an HD map update from the vehicle management system for a specific HD map installed at the first vehicle is available, and in response to receiving the notification, the first vehicle may request the notified HD map update from the second vehicle. In other embodiments, the request for the HD map update broadcasted by the first vehicle may be a generic request, and the generic request may include version information of one or more HD maps installed at the first vehicle.

[0092] The specific HD map update may be indicated in the notification received from the vehicle management system. The second vehicle may receive an indication of the specific HD map update, or version information of the specific HD map, and based on the version information of the specific HD map, the second smart contract may determine if an HD map update is available at the second vehicle for the specific HD map. For example, the version information may include a first version number of the HD map of the first vehicle. If the first version number of the HD map of the first vehicle is less than a second version number of a same HD map of the second vehicle, it may be inferred that the second vehicle is operating with a more recent version of the HD map, whereby an HD map update package may be available at the second vehicle to be downloaded by the first vehicle.

[0093] At 410, method 400 includes determining whether a second vehicle has responded to the request broadcasted by the first vehicle. The second vehicle may respond to the request based on information included in the request. For example, the second vehicle may not respond to the request as a result of the second version number of the HD map not being greater than the first version number, from which it may be inferred that the second vehicle is not using a more recent version of the HD map and therefore may not have the HD map update. Alternatively, if a second vehicle does not respond to the request at 410, it may be inferred that a second vehicle is not within the threshold distance of the first vehicle. If the second vehicle does not respond to the request at 410, method 400 proceeds to 412. At 412, method 400 includes continuing operation of the first vehicle until a second vehicle responds to the request of the first vehicle.

[0094] It should be appreciated that in some embodiments, the role of the second vehicle, or other neighboring vehicles described herein, may be performed by a stationary HD map update system of an edge node (e.g., edge node 150) installed within physical infrastructure proximate to the first vehicle. For example, the first vehicle may be parked in a lot, and an HD map update system may be installed at an edge node of a building or communications installation located at the lot. When the first vehicle is parked in the lot, method 400 may be performed to manage an HD map update between the first vehicle and the HD map update system installed at the lot, using vehicle-to- infrastructure (V2X) communication rather than V2V communication.

[0095] If a second vehicle responds to the request at 410, it may be inferred that the second vehicle is within the threshold distance of the first vehicle and includes the HD map update, whereby method 400 proceeds to 414. When the second vehicle responds to the request, the response set by the second vehicle may be received by the first vehicle. In response to receiving the request, the second vehicle may transmit instructions for downloading the HD map update package from the second vehicle.

[0096] At 414, method 400 includes establishing a wireless connection with the second vehicle. In various embodiments, the wireless connection may be established between a first communication module of the first vehicle and a second communication module of the second vehicle. The wireless connection may be established one or more of various wireless technologies (e.g., radio frequency, infrared, near field communication (NFC), BLUETOOTH™, etc.), In accordance with instructions established in the second smart contract.

[0097] At 416, at the 400 includes initiating a download of an HD map update package from the second vehicle to the first vehicle. In various embodiments, the HD map update package may be downloaded by the first vehicle in chunks, in accordance with instructions provided by the second vehicle. For example, a first chunk of the update package may be downloaded. After the first chunk is downloaded, a second chunk of the update package may be downloaded. After the second chunk is downloaded, a third chunk of the update package may be downloaded, and so on. In this way, if the connection is lost during downloading, previously downloaded chunks may be stored, and remaining chunks that have not been downloaded may be downloaded at a later time from a different vehicle. The downloaded chunks may be stored in a first memory (e.g., memory 134) of a first HD map update system of the first vehicle.

[0098] In some cases, a portion of the downloaded chunks may already be downloaded and stored in the first memory. For example, the portion of the downloaded chunks may have been previously downloaded from a different vehicle, and the different vehicle may have moved out of the threshold distance before a remaining portion of the downloaded chunks could be downloaded. Therefore, prior to initiating the download, the first HD map update system may check to see if one or more chunks of the download are already in the first memory. If the one or more chunks of the download are already in the first memory, the HD map update system may determine a last chunk downloaded, and begin the download at a subsequent chunk.

[0099] At 418, method 400 includes determining whether the HD map update package has been entirely downloaded. If at 418 it is determined that the update package has not been entirely downloaded, method 400 proceeds to 420. At 420, method 400 includes determining whether the second vehicle is still within the threshold distance. For example, the first vehicle may send a request for a subsequent chunk of the download. If the request is received at the second vehicle, it may be inferred that the second vehicle is still within the threshold distance. If at 420 it is determined that the second vehicle is still within the threshold distance, method 400 proceeds to 422, A 422, method 400 includes downloading the subsequent chunk, and method 400 proceeds back to 418. In this way, method 400 includes continuing to download additional chunks of the HD map update package until it is determined that the download is finished or the second vehicle is no longer within the threshold distance.

[0100] Alternatively, if at 420 it is determined that the first vehicle is no longer within the threshold distance (e.g., the request for the subsequent chunk is not received), and a subsequentchunk of the HD map update package cannot be downloaded, method 400 proceeds to 424. At 424, method 400 includes recording a last chunk downloaded at the first memory of the first HD map update system of the first vehicle. Method 400 proceeds to 426, where method 400 includes continuing vehicle operation using the current version of the HD map. The HD map update is not installed, and installation of the HD map update may be postponed until a third vehicle including the HD map update is within the threshold distance, and the remaining chunks may be downloaded from the third vehicle.

[0101] If at 418 it is determined that the HD map update package has been entirely downloaded, method 400 proceeds to 428. At 428, method 400 includes installing the HD map update at the first vehicle. In various embodiments, the HD map update may be installed at a navigation system installation (e.g., navigation system 140) of the first vehicle. When the HD map update is installed, one or more files of the HD map update package may be deleted from the first memory. Additionally, a notification may be sent to the vehicle management system that the HD map update was installed.

[0102] FIG. 5 shows an exemplary method 500 for providing an HD map update of a HD map installed at a second vehicle to a first vehicle, where the first vehicle and second vehicle are the same as the first vehicle and second vehicle of FIG. 4, respectively. Method 500 may be executed by a ECU of the second vehicle (ECU 132 of vehicle 130 of FIG. 1).

[0103] Method 500 begins at 502, where method 200 includes estimating and measuring vehicle operating conditions of the second vehicle. Estimating the vehicle operating conditions of the second vehicle may be performed similarly to method 200. Estimating and / or measuring vehicle operating conditions may include determining whether the one or more HD maps installed at the second vehicle are up to date, and / or whether updated versions of the one or more HD maps are available. Estimating and / or measuring the vehicle operating conditions may include determining whether the second vehicle is being operated autonomously, meaning, without active involvement of a driver. Additionally, estimating and measuring vehicle operating conditions of the vehicle may include initializing an OTA client of the vehicle.

[0104] At 504, method 500 includes determining whether an HD map update request is received from the first vehicle. As described above in reference to FIG. 2, the HD map update request may be transmitted from the first vehicle to the second vehicle via V2V communication. The second vehicle may be listening for the HD map update request at a specific radio frequency, or the secondvehicle may identify the HD map update request via header data / metadata and / or protocol data included in the HD map update request.

[0105] If at 504 the HD map update request is not received, method 500 proceeds to 506. At 506, method 500 includes continuing operation of the second vehicle until an HD map update request is received, and method 500 proceeds back to 504. Alternatively, if at 504 the HD map update is received by the second vehicle, method 500 proceeds to 508.

[0106] At 508, method 500 includes identifying a HD map associated with the HD map update request, and comparing a first version number of the HD map installed at the first vehicle with a second version number of the HD map installed at the second vehicle. In some embodiments, the first version number may be transmitted within the HD map update request. In other embodiments, either or both of the first version number and the second version number may be retrieved from the second vehicle.

[0107] At 510, method 500 includes determining whether an HD map update package is available at the second vehicle for downloading by the first vehicle. If the first version number is equal to the second version number, it may be inferred that the first vehicle and a second vehicle may both have the same version of the HD map, whereby an HD map update package may not be available at the second vehicle for downloading by the first vehicle. Alternatively, if the first version number is less than the second version number, it may be inferred that the second vehicle has a more recent version of the HD map than the first vehicle, whereby an HD map update package may be available at the second vehicle for downloading by the first vehicle.

[0108] If at 510 it is determined that the HD map update package is not available at the second vehicle, method 500 proceeds to 512. At 512, method 500 includes sending a response to the first vehicle indicating that the HD map update package is not available at the second vehicle for downloading by the first vehicle, and method 500 ends. If at 510 it is determined that the HD map update package is available at the second vehicle, method 500 proceeds to 514.

[0109] At 514, method 500 includes sending a response to the first vehicle indicating that the HD map update package is available at the second vehicle for downloading by the first vehicle. Method 500 proceeds to 516.

[0110] At 516, method 500 includes establishing a wireless connection between the first vehicle and the second vehicle, and providing download access to the HD map update package to the first vehicle. Specifically, the wireless connection may be established between a first communicationmodule of the first vehicle and a second communication module of the second vehicle (e.g., communication module 138 of FIG. 1). Establishment of the wireless connection may be initiated by the first vehicle, or by the second vehicle. After the wireless connection between the first communication module and a second communication module is established, the first vehicle may proceed to download the HD map update package to the first vehicle. As described above in reference to FIG. 2, the HD map update package may be downloaded in chunks. Method 500 ends.

[0111] Referring now to FIG. 6, a flowchart is shown illustrating an exemplary method 600 for a vehicle such as vehicle 130 to connect to an infrastructural edge node of a connected vehicle management system, such as edge node 150 of VMS 102 of FIG. 1. Method 600 may be executed by a controller of the vehicle, in accordance with instructions stored in a memory of the vehicle. In various embodiments, method 600 may be implemented by a control unit of the vehicle, such as ECU 132.

[0112] At 602, method 600 includes estimating and measuring vehicle operating conditions of the vehicle, as described above in reference to methods 200 and 300. Estimating and / or measuring vehicle operating conditions may include determining whether the one or more HD maps installed at the second vehicle are up to date, and / or whether updated versions of the one or more HD maps are available.

[0113] At 604, method 600 includes determining a current location of the vehicle. The current location of the vehicle may be obtained, for example, from a vehicle guidance system and / or a navigation system of the vehicle (e.g., navigation system 140 of FIG. 1).

[0114] At 606, method 600 includes discovering a nearest edge node to the current location. In some embodiments, the edge nodes may be servers running at cellular towers of a wireless, cellular network that the vehicle is connected to, whereby the nearest edge node to the current location may be a nearest cellular tower to the current location. In other embodiments, the wireless network may not be a cellular tower, and the nearest edge node may be located within different infrastructure (e.g., in a building).

[0115] At 608, method 600 includes determining whether the vehicle is already connected to an edge node. If at 608 it is determined that the vehicle is already connected to an edge node, method 600 proceeds to 614. If at 608 it is determined that the vehicle is not connected to an edge node, method 600 proceeds to 610.

[0116] At 610, method 600 includes connecting to the edge node. Once connected, at 612, method 600 includes changing a network configuration of the vehicle to route network traffic to the vehicle through the edge node, and method 600 proceeds to 614.

[0117] At 614, method 600 includes collecting HD map information of the vehicle. The HD map information may include, for example, versions of various HD maps stored at the vehicle.

[0118] At 616, method 600 includes determining whether an updated HD map is available at the edge node. Determining whether the updated HD map is available at the edge node may comprise determining whether a first HD map installed at a navigation system of the vehicle has a lower version number than a second HD map stored at the edge node.

[0119] If the first HD map has a higher version number than the second HD map, it may be inferred that the second HD map is not a more recent (e.g., updated) version of the HD map, whereby method 600 proceeds to 626. At 626, method 600 includes continuing operation of the vehicle using the HD map installed at the vehicle (e.g., the first HD map), and method 600 ends. Alternatively, if the first HD map has a lower version number than the second HD map, it may be inferred that the second HD map is a more recent (e.g., updated) version of the HD map, whereby method 600 proceeds to 618.

[0120] At 618, method 600 includes initiating a download of the second HD map from the edge node. In various embodiments, steps 618-626 of method 600 may be similar to or the same as corresponding steps of method 200 of FIG. 2. As with the V2V communication, the second HD map may be downloaded in batches or chunks. If the download completes while the vehicle is connected to the edge node at 620, method 600 proceeds to 622 and the second HD map is installed at the vehicle. Alternatively, if the download does not complete, a last chunk downloaded from the edge node may be recorded at 624, and the vehicle may continue operation using the first HD map of the vehicle. For example, the vehicle may leave a vicinity of the edge node before the download is complete, and when the vehicle leaves the vicinity, the connection between the edge node and the vehicle may be lost. In such cases, the vehicle may finish downloading the updated HD map at a later time, at a second edge node or via one or more other vehicles via the V2V communication.

[0121] FIG. 7 shows a timeline 700 during updating of HD map installed at a first vehicle 704, a second vehicle 706, and a third vehicle 708 of a vehicle fleet, by a vehicle management system (VMS) 702, which may be non-limiting versions of vehicle 130 and VMS 102 of FIG. 1. It should be appreciated that while timeline 700 describes the transferring of discreet chunks of an HD mapupdate to a series of vehicles, each of first vehicle 704, second vehicle 706, and third vehicle 708 may be replaced with an infrastructural edge node (e.g., edge node 150) without departing from the scope of this disclosure. For example, while timeline 700 describes transferring the HD map update from first vehicle 704 to third vehicle 708 via second vehicle 706, in other embodiments, a similar series of events may transpire when transferring the HD map update from first vehicle 704 to third vehicle 708 via an edge node.

[0122] First vehicle 704, second vehicle 706, and third vehicle 708 may be using a version 1.0 of the HD map. An update package is released by VMS 702 may be disseminated to first vehicle 704, second vehicle 706, and third vehicle 708 in accordance with interactions between the vehicles described on timeline 700, where time is indicated on a left side of timeline 700. The interactions between the vehicles are indicated by arrows between dotted lines associated with VMS 702, first vehicle 704, second vehicle 706, and third vehicle 708.

[0123] At a time t=l, first vehicle 704 is using version 1.0 of the HD map. At a time t=2, the HD map update package is downloaded from VMS 702 by first vehicle 704. The HD map update package may be downloaded via a wireless network, such as wireless network 190 of FIG. 1. In some embodiments, vehicle 704 may be located at an establishment of a manufacturer of vehicle 704, and the HD map update package may be downloaded from VMS 702 to first vehicle 704 via a private wireless network. In other embodiments, first vehicle 704 may be a vehicle in circulation in a region of VMS 702, and the HD map update package may be downloaded from VMS 702 via a public wireless network, over the Internet. At a time t=3, the update package is installed at first vehicle 704. At time t=4, first vehicle 704 is using version 2.0 of the HD map.

[0124] At a time t=5, second vehicle 706 enters into proximity with first vehicle 704, and a connection is made between first vehicle 704 and second vehicle 706. For example, the connection may be made between a first communication module 138 of first vehicle 704 and a second communication module 138 of second vehicle 706.

[0125] At times t=6, 7, and 8, the HD map update package may be downloaded from first vehicle 704 to second vehicle 706 in chunks. For example, a first chunk of the HD map update package may be downloaded at time t=6; a second chunk of the HD map update package may be downloaded at time t=7; and a third chunk of the HD map update package may be downloaded at time t=8.

[0126] At a time t=9, the wireless connection between first vehicle 704 and second vehicle 706 may be lost. As a result of the connection being lost, downloading of the HD map update package from first vehicle to second vehicle 706 may end, where the update package may not have been fully downloaded. For example, the HD map update package may be divided into five chunks, and only three chunks may be downloaded prior to losing the connection. The three chunks may be stored in a memory of an HD map update system of second vehicle 706, such as memory 134 of vehicle 130 of FIG. 1.

[0127] At a time t=10, third vehicle 708 enters into proximity with first vehicle 704, and a connection is made between first vehicle 704 and third vehicle 708. For example, the connection may be made between the first communication module 138 of first vehicle 704 and a second communication module 138 of third vehicle 708.

[0128] At times t=l 1-15, the HD map update package may be downloaded from first vehicle 704 to third vehicle 708 in chunks. By time t=l 5, all five of the chunks of the HD map update package may be downloaded to third vehicle 708. As a result of all five of the chunks of the HD map update package being downloaded, at time t=l 6, the HD map update package is installed at third vehicle 708.

[0129] At a time t=l 7, as a result of installing the HD map update package, version 2.0 of the HD map is installed at third vehicle 708.

[0130] At a time t=18, second vehicle 706 enters into proximity with third vehicle 708, and a connection is made between third vehicle 708 and second vehicle 706. For example, the connection may be made between first communication module 138 of third vehicle 708 and a second communication module 138 of second vehicle 706.

[0131] At times t= 19 and 20, the remaining chunks four and five of the HD map update package may be downloaded from second vehicle 706 to third vehicle 708. As a result of all five of the chunks of the HD map update package being downloaded at second vehicle 706, at time t=21 , the HD map update package is installed at second vehicle 706. At a time t=22, version 2.0 of the HD map is installed at second vehicle 706.

[0132] In this way, the HD map update package may be transmitted from the VMS 702 to each of first vehicle 704, second vehicle 706, and third vehicle 708 via V2V communication between the vehicles. By transmitting the HD map update package via the V2V communication rather thanfrom VMS 702, a use of resources of VMS 702 may be reduced, increasing an efficiency of VMS 702.

[0133] It should be appreciated that VMS 702 may be further decentralized and / or extended by taking advantage of greater data processing and storage resources closer to the edge of the network, which may minimize latency, reduce central server load, and ensure that vehicles receive the most relevant and up-to-date information with minimal delays. In other words, edge processing units (EPUs) of the edge nodes and vehicles are strategically (based on network and storage needs) located computing units equipped with the capability to process and temporarily store HD map updates, as part of road infrastructure. Data of an HD map update can thus be advantageously routed to specific EPUs or between specific EPUs to maximize an efficiency of the dissemination and decrease an amount of time taken to install the HD map update at a plurality of vehicles within a geographical region.

[0134] When the VMS receives a new HD map update, the VMS may segment the update by geographical relevance and route specific update segments to EPUs serving those regions. The EPUs may evaluate real-time traffic conditions, vehicle densities, and specific route requests to prioritize time-sensitive updates for immediate distribution. The updates may be prioritized based on urgency and impact, focusing on safety and high-impact updates for immediate attention. The updates may be segmented according to geographic impact areas to streamline dissemination. Segmented updates may then directed to EPUs located in the corresponding geographic areas. The EPUs may evaluate current traffic conditions, vehicle densities, and specific route requests to refine dissemination strategies. High-priority updates may be cached for quick access and rapid dissemination. In some examples, preprocessing needs may be assessed for each update segment, considering network conditions and receiving device capabilities. Adjustments, including compression and segmentation, may be applied to prepare updates for efficient transmission. This may include adjusting the preprocessing and optimization plan for compressing update files or segmenting the update files into smaller, more easily downloadable chunks tailored for the network conditions and device capabilities in the specific area.

[0135] In some examples, a bottom-up approach may be taken to decentralizing the dissemination, where EPUs and vehicles to contribute data and partake in decision-making with respect to how HD map update data is routed throughout a network of vehicles. This may ensure efficient, localized processing and distribution of map updates, adapting dynamically to real-time vehicularand environmental conditions. Vehicles may continuously transmit data to nearby edge nodes, providing "up-to-the-minute" insights on traffic patterns, road conditions, and map accuracy (existing map data vs real time measurements). Each edge node may process the received data to identify an applicability of map updates. Utilizing localized decision-making algorithms, edge nodes may determine the priority and scope of updates applicable to their specific geographic area. The edge nodes may also identify patterns and notify the VMS about potential permanent changes in HD maps used by the vehicles. Based on the analysis, the edge nodes can independently optimize map updates for their region, segmenting large updates into manageable chunks, compressing data for efficient transmission, and even generating localized updates based on the collected vehicle data. After performing the analysis, the edge nodes may distribute the optimized updates directly to vehicles within their coverage area through V2X communication, prioritizing updates based on real-time conditions reported by the vehicles.

[0136] As an example of how the HD map updates might be performed in a real world situation, an exemplary map update scenario in which an HD map update package may be propagated to vehicles in a given region (as described above in reference to FIGS. 5, 6, and 7) is shown in reference to FIGS. 8A and 8B. Referring to FIG. 8A, an exemplary simplified HD map 800 of a geographical area is shown. HD map 800 includes a first residential area 815 and a second residential area 817, which are connected via a highway 801. The depicted geographical area additionally includes a first road 802, a second road 803, and a third road 804. First road 802 and third road 804 intersect with highway 801, and third road 804 connects first road 802 and second road 803, such that second residential area 817 is encircled by first road 802, second road 803, third road 804, and highway 801. Third road 804 connects to a fourth road 805, leading away from second residential area 817.

[0137] A major accident has occurred at a point 819 on highway 801, indicated by an X, at an intersection to an entrance to residential area 817. As a result of the major accident, highway 801 is closed at point 819, and vehicles traveling on highway 801 are not able to pass point 819. At a time of the accident, a plurality of vehicles are travelling on routes within the geographical area. A first set of vehicles 820, 821, and 822 are travelling on highway 801. A second set of vehicles 841, 842, and 843 are travelling within first residential area 815. A third set of vehicles 830, 831, 832, 833, and 834 are travelling on first road 802 towards second residential area 817 and highway 801. A fifth set of vehicles 852, 853, 854, 855, and 856 are travelling within second residentialarea 817. Additionally, a vehicle 857 is traveling on fourth road 805, and a vehicle 858 is travelling on third road 804.

[0138] As a consequence of the major accident, an HD map update may be generated for immediate dissemination within an affected area of the accident, meaning the geographical area. The HD map update may be disseminated to vehicles travelling within the affected area, including the first, second, third, fourth, and fifth sets of vehicles described above. The HD map update indicates that highway 801 is unpassable at point 819. Once the HD map update is installed, navigation systems of vehicles travelling on highway 801 or on a road merging with highway 801 may navigate the vehicles along a detour route around point 819, as indicated by a plurality of dashed arrows 825.

[0139] The dissemination of the HD map update may be performed by a vehicle management system, such as VMS 102 of FIG. 1, in accordance with one or more of the methods of FIGS. 2-6. The dissemination may be performed using a combination of V2V communication between the vehicles; V2I communication between the vehicles and one or more edge nodes of the VMS, such as a first edge node 812, a second edge node 814, and a third edge node 816; and transmission via a wireless network (e.g., network 190), for example, via a cell phone tower 810; all in a manner that may be conceptualized and modeled as a dynamic graph, as shown in FIG. 8B. This graph representation facilitates an intuitive and efficient way of visualizing the complex network of interactions and flow of data within the system, as the HD map update is distributed in chunks by a chunk management engine, as described above.

[0140] Referring now to FIG. 8B, a dynamic graph 870 is shown of the vehicles and edge nodes described above in reference to FIG. 8A. In the graph-based model shown in FIG. 8B, each entity — whether a central server, an EPU, or a vehicle — is represented as a node. These nodes are interconnected by edges (e.g., lines) that symbolize available communication paths between the nodes. A distinctive feature of this model is the dynamic nature of each node with its current available resources, such as computing power, storage capacity, and connectivity status. Additionally, each node's expected time within the communication range of adjacent nodes may also be dynamically updated, reflecting the highly mobile nature of the vehicular network. Edges within this graph may be attributed with properties that represent the network bandwidth and the quality of connectivity between nodes. This allows for a granular and real-time understanding of the communication capacity and potential bottlenecks within the network.

[0141] For example, in response to the accident, an HD map update may be generated by a third party responsible for supplying maps to vehicle navigation systems. The HD map update may be transmitted to the VMS by the third party. When the VMS receives the HD map update, the VMS may compare the HD map update to an existing HD map, and may extract a feature that is different between the two maps using a first model, where the feature is the closure of the road at point 819 of FIG. 8A. The first model may classify the feature as a road closure. The first model, or a different model may assign a priority score to the HD map update, based on a predicted impact of the road closure on local traffic. Due to point 819 being located on a route with historically heavy traffic, the priority score may be higher than a threshold value, indicating that the HD map update is a time-sensitive update.

[0142] The VMS may determine that a primary affected area comprises a section of highway 801 within a threshold distance of point 819 (e.g., one miles), and a secondary affected area comprises roads 802, 803, and 804, which lead to and / or connect with highway 801. The VMS may determine that edge node 812 is in the primary affected area, and that edge nodes 814 and 816 are in the secondary affected area.

[0143] The VMS may transmit a request to edge node 812 to provide location, connectivity, and resource data of vehicles within a range of edge node 812. In response to receiving the request, edge node 812 may broadcast a request for location, connectivity, and resource data of vehicles in a proximity of edge node 812. Edge node 812 may receive the location, connectivity, and resource data of vehicles 820 and 821 travelling on highway 801.

[0144] The VMS may also transmit a request to edge nodes 814 and 816 to provide location, connectivity, and resource data of vehicles within a range of edge nodes 814 and 816. In response to receiving the request, edge nodes 814 and 816 may broadcast a request for location, connectivity, and resource data of vehicles in a proximity of edge nodes 814 and 816. Edge nodes 814 and 816 may additionally broadcast a request for trip destination data of vehicles in the proximity of edge nodes 814 and 816 that are navigating to a destination. Edge node 814 may receive the location, connectivity, resource, and trip destination data of vehicles 830, 831, 832, and 833 travelling on road 802. Edge node 816 may receive the location, connectivity, resource, and trip destination data of vehicles 857 and 858 travelling on road 804. The requests sent from VMS to edge nodes 812, 814, and 816 may be sent via a wireless (e.g., cell phone) network, for example, through cell phone tower 810. The edge nodes 812, 814, and 816 may includetransmitters, receivers, and computational / memory resources installed in a physical infrastructure near one or more roads, such as a building.

[0145] The VMS may rank edge nodes 814 and 816 based on the number of responding vehicles, and edge node 814 may be assigned a higher rank than edge node 816, as a result of four vehicles responding to edge node 814 as opposed to two vehicles responding to edge node 816. Additionally, vehicles 830, 831, 832, and 833 may be ranked according to a distance from edge node 814, and vehicles 857 and 858 may be ranked according to a distance from edge node 816.

[0146] The VMS may first transmit the HD map update to edge node 812, with instructions for edge node 812 to transmit the HD map update to vehicles 820 and 821. The VMS may then transmit the HD map update to higher-ranked edge node 814, with instructions for edge node 814 to transmit the HD map update to vehicles 830, 831, 832, and 833. However, from vehicle data received from top-ranked vehicle 833 (e.g., the farthest responding vehicle from edge node 814), the VMS may determine that vehicle 833 is close to an edge of a range of edge node 814, and may move out of the range before receiving the HD map update. Based on the determination, the VMS may transmit the HD map update directly to vehicle 833, via the wireless network and cell tower 810.

[0147] The VMS may then transmit the HD map update to edge node 816, with instructions for edge node 816 to transmit the HD map update to vehicles 857 and 858. However, from vehicle data received from top-ranked vehicle 858 (e.g., the farthest responding vehicle from edge node 816), the VMS may determine that vehicle 858 is close to an edge of a range of edge node 816 and moving away from edge node 816. Based on the determination, the VMS may transmit the HD map update directly to vehicle 858, via the wireless network and cell tower 810.

[0148] As a result of receiving the HD map update, vehicles 820 and 821 may be rerouted along road 803, around the accident at point 819. On road 803, vehicles 820 and 821 may transmit the HD map update to vehicle 851 via V2V communication. For example, a first part of the HD map update may be transmitted from vehicle 821 to vehicle 851, and a second part of the HD map update may be transmitted from vehicle 820 to vehicle 851, in the manner described above in reference to FIG. 7. Concurrently, as a result of receiving the HD map update, vehicles 830, 831, 832, and 833 may be rerouted along road 803, around the accident at point 819. On road 803, vehicle 833 may be travelling in a same direction as vehicle 834, whereby vehicle 833 maytransmit the HD map update to vehicle 834 via V2V communication. Vehicle 822 may receive the HD map update from vehicle 858 in a similar manner.

[0149] In this way, the HD map update may be disseminated to all of the vehicles travelling on each of highway 801, road 802, road 803, and road 804 (e.g., the primary and secondary affected areas) in a manner that minimizes an amount of time taken to update all of the vehicles. That is, by disseminating the HD map update to the vehicles in this manner, the amount of time may be less than a second amount of time taken to update the vehicles travelling on each of highway 801, road 802, road 803, and road 804 directly via the wireless network, which may experience bandwidth limitations and slow update speeds. The amount of time may also be less than a third amount of time taken to update the vehicles travelling on each of highway 801, road 802, road 803, and road 804 via edge nodes 812, 814, and 816 without relying on the V2V communication. Further, edge nodes 812, 814, and 816 may not be positioned in a manner such that the HD map update can be disseminated to vehicles travelling on each of highway 801, road 802, road 803, and road 804 without relying on the V2V communication.

[0150] After the VMS transmits the HD map update to edge nodes 812, 814, and 816 and vehicles 833 and 858, the VMS may then proceed to update remaining vehicles of FIG. 8A that are not included in the primary and secondary affected areas. The remaining vehicles include vehicles 841, 842, and 843 operating in first residential area 815, and vehicles 852, 853, 854, 855, and 856 operating in second residential area 817, which may not be immediately affected by the road closure. The HD map update may be transmitted to the remaining vehicles via edge nodes 812, 814, and 816 and / or via one or more of the vehicles of the primary and secondary affected areas that have received the HD map update. Additionally or alternatively, The HD map update may be transmitted directly to the remaining vehicles via the wireless network and cell tower 810. In some embodiments, the VMS may determine, via vehicle data received from the remaining vehicles, a portion of the remaining vehicles that may more efficiently receive the HD map update from other vehicles via V2V communication and / or from edge nodes 812, 814, and / or 816 via I2V communication. It should be appreciated that the HD map update may be received and updated at the remaining vehicles while vehicles of the primary and secondary affected areas are still being updated. In this way, the HD map update may be applied at all of the vehicles of FIG. 8A in a more or less concurrent manner.

[0151] By not transmitting the HD map update directly to vehicles 841-843 and vehicles 851-856 via the wireless network, and using available bandwidth of the wireless network to send the HD map update to edge node 812, edge node 814, vehicle 833, edge node 816, vehicle 858 in that order, the HD map update may be more efficiently propagated throughout the vehicles of the geographic area. Because transmissions of the HD map update to vehicles within each stage are performed concurrently, the HD map update may reach all of the vehicles in the geographic area faster than if the HD map update were transmitted to each of the vehicles in the geographic area via cell phone tower 810.

[0152] Thus, as a result of a small number of initial downloads via a wireless network, the HD map update may be propagated through and installed at a plurality of vehicles, without each vehicle of the plurality of vehicles having to download the HD map update directly from the VMS via the wireless network. As a result, network traffic on the wireless network may be reduced. Additionally, a first amount of time taken to install the HD map update at the plurality of vehicles via the V2X communication (e.g., from a vehicle or from an edge node) may be faster than a second amount of time taken to install the HD map update at the plurality of vehicles via direct downloads from the VMS system, as a result of a greater number of HD map update downloads being performed via the V2X communication in parallel. Further, because processing resources used for managing the downloads may be consumed at each vehicle of the plurality of vehicles as opposed to the VMS, processing resources of the VMS may be freed up to perform other tasks, thereby increasing a performance and / or functioning of the VMS.

[0153] The following claims particularly point out certain combinations and sub-combinations regarded as novel and non-obvious. These claims may refer to “an” element or “a first” element or the equivalent thereof. Such claims should be understood to include incorporation of one or more such elements, neither requiring nor excluding two or more such elements. Other combinations and sub-combinations of the disclosed features, functions, elements, and / or properties may be claimed through amendment of the present claims or through presentation of new claims in this or a related application. Such claims, whether broader, narrower, equal, or different in scope to the original claims, also are regarded as included within the subject matter of the present disclosure.

Claims

CLAIMS:

1. A method for a vehicle management system (VMS) for managing updating high-definition (HD) maps used by navigation systems of a plurality of vehicles, the method comprising: receiving an HD map update; assigning a priority score to the HD map update; in response to the priority score exceeding a threshold priority score: determining a primary geographical area affected by the HD map update and a secondary geographical area affected by the HD map update; identifying a first portion of edge nodes of the VMS located within the primary geographical area, and a second portion of edge nodes of the VMS located within the secondary geographical area; transmitting the HD map update to the first portion of edge nodes, via a wireless network; after transmitting the HD map update to the first portion of edge nodes, transmitting the HD map update to the second portion of edge nodes, via the wireless network; and after transmitting the HD map update to the second portion of edge nodes, transmitting the HD map update to edge nodes of the VMS outside the primary geographical area and secondary geographical area via the wireless network; and in response to the priority score not exceeding the threshold priority score, transmitting the HD map update to a plurality of edge nodes of the VMS via the wireless network.

2. The method of claim 1, wherein assigning the priority score to the HD map update further comprises: comparing the HD map update to an existing HD map; extracting features of the HD map update that are different between the HD map update and the existing HD map, using a first artificial intelligence (Al) model; assigning an urgency score to each extracted feature; and aggregating the urgency scores of each extracted feature to generate an overall priority score.

3. The method of claim 2, wherein determining the primary geographical area affected by the HD map update and the secondary geographical area affected by the HD map update further comprises defining the primary geographical area where changes to the existing HD map have occurred, and defining the secondary geographical area as an area where traffic may be redirected as a result of the changes.

4. The method of claim 2, further comprising assigning weights to each of the urgency scores, the weights retrieved from a lookup table, and performing a weighted aggregation of the urgency scores.

5. The method of claim 1, wherein: transmitting the HD map update to the first portion of edge nodes further comprises: sending a first request to the first portion of edge nodes for location, connectivity, and resource data of a first plurality of vehicles within a range of the first portion of edge nodes; ranking each edge node of the first portion of edge nodes based on a number of vehicles responding to the request; and transmitting the HD map update to the first portion of edge nodes in an order of the rank of each edge node of the first portion of edge nodes; and transmitting the HD map update to the second portion of edge nodes further comprises: sending a second request to the second portion of edge nodes for location, connectivity, resource, and trip destination data of a second plurality of vehicles within a range of the second portion of edge nodes; ranking each edge node of the second portion of edge nodes based on a number of vehicles responding to the request that are traveling to or towards the primary geographical area; and transmitting the HD map update to the second portion of edge nodes in an order of the rank of each edge node of the second portion of edge nodes.

6. The method of claim 5, further comprising: ranking each vehicle of the first plurality of vehicles and the second plurality of vehicles using a second Al model, where the second Al model takes a closest edge node, connectivity, and available resources of the vehicle as input, and outputs the ranking; after transmitting the HD map update to the first portion of edge nodes and before transmitting the HD map update to the second portion of edge nodes, transmitting the HD map update to one or more top-ranked vehicles of the first plurality of vehicles directly via the wireless network; and after transmitting the HD map update to the second portion of edge nodes and before transmitting the HD map update to the edge nodes of the VMS outside the primary geographical area and secondary geographical area, transmitting the HD map update to one or more top-ranked vehicles of the second plurality of vehicles directly via the wireless network.

7. The method of claim 5, wherein the connectivity data includes: a bandwidth, latency, and quality of a current connection of a respective vehicle; a supported communication protocol of the respective vehicle; current communication capabilities and available bandwidth of the respective vehicle.

8. The method of claim 5, wherein the resource data includes diagnostic data of a respective vehicle with respect to memory and processing resources of the vehicle.

9. The method of claim 1, further comprising strategically prioritizing a transmission sequence of HD map updates to ensure quick and widespread coverage with minimal bandwidth consumption, using one or more of: a geographic segmentation model that divides a geographic area of the HD map updated into segments based on vehicular density, network capacity, and proximity to edge nodes; a priority-queue management model that ranks vehicles and edge nodes within each segment by a potential to propagate the update efficiently, based on connectivity strength, vehicle mobility patterns, and data transmission capabilities of each vehicle / edge node;a bandwidth allocation model that dynamically allocates bandwidth resources across the wireless network, prioritizing initial transmissions to areas and nodes with a highest propagation potential based on an output of the priori ty-queue management model; and a dissemination control model that coordinates a staged dissemination process, initiating transmissions according to the priority queue and the allocated bandwidth resources, and adapting to changes in network conditions in real-time.

10. The method of claim 1, further comprising determining an optimal chunk size and compression strategy for data of the HD map update, based on network conditions of the wireless network and edge node capabilities.

11. The method of claim 1 , wherein the HD map update is transmitted to vehicles managed by the VMS via a combination of vehicle-to-vehicle (V2V) communication with other vehicles and vehicle-to-infrastructure / infrastructure-to-vehicle (V2I / I2V) communication with edge nodes of the VMS.

12. The method of claim 1, further comprising: after transmitting the HD map update to the second portion of edge nodes, transmitting the HD map update to one or more vehicles outside the primary geographical area and secondary geographical area via the wireless network.

13. A vehicle management system (VMS), comprising: a processor, and a memory storing instructions that when executed, cause the processor to: receive a high-definition (HD) map update; assign a priority score to the HD map update; in response to the priority score exceeding a threshold priority score: determine a primary geographical area affected by the HD map update and a secondary geographical area affected by the HD map update; identify a first portion of edge nodes of the VMS located within the primary geographical area, and a second portion of edge nodes of the VMS located within the secondary geographical area;transmit the HD map update to the first portion of edge nodes, via a wireless network; after transmitting the HD map update to the first portion of edge nodes, transmit the HD map update to the second portion of edge nodes, via the wireless network; and after transmitting the HD map update to the second portion of edge nodes, transmit the HD map update to edge nodes of the VMS outside the primary geographical area and secondary geographical area via the wireless network; and in response to the priority score not exceeding the threshold priority score, transmitting the HD map update to a plurality of edge nodes of the VMS via the wireless network.

14. The VMS of claim 13, wherein further instructions are stored in the memory that when executed, cause the processor to: compare the HD map update to an existing HD map; extract features of the HD map update that are different between the HD map update and the existing HD map, using a first artificial intelligence (Al) model; assign an urgency score to each extracted feature; aggregate the urgency scores of each extracted feature to generate an overall priority score of the HD map update.

15. The VMS of claim 13, wherein further instructions are stored in the memory that when executed, cause the processor to: when transmitting the HD map update to the first portion of edge nodes: send a first request to the first portion of edge nodes for location, connectivity, and resource data of a first plurality of vehicles within a range of the first portion of edge nodes; rank each edge node of the first portion of edge nodes based on a number of vehicles responding to the request; and transmit the HD map update to the first portion of edge nodes in an order of the rank of each edge node of the first portion of edge nodes; and when transmitting the HD map update to the second portion of edge nodes:send a second request to the second portion of edge nodes for location, connectivity, resource, and trip destination data of a second plurality of vehicles within a range of the second portion of edge nodes; rank each edge node of the second portion of edge nodes based on a number of vehicles responding to the request that are traveling to or towards the primary geographical area; and transmit the HD map update to the second portion of edge nodes in an order of the rank of each edge node of the second portion of edge nodes.

16. The VMS of claim 15, wherein further instructions are stored in the memory that when executed, cause the processor to: rank each vehicle of the first plurality of vehicles and the second plurality of vehicles using a second Al model, where the second Al model takes a closest edge node, connectivity, and available resources of the vehicle as input, and outputs the rank; after transmitting the HD map update to the first portion of edge nodes and before transmitting the HD map update to the second portion of edge nodes, transmit the HD map update to one or more top-ranked vehicles of the first plurality of vehicles directly via the wireless network; and after transmitting the HD map update to the second portion of edge nodes and before transmitting the HD map update to the edge nodes of the VMS outside the primary geographical area and secondary geographical area, transmit the HD map update to one or more top-ranked vehicles of the second plurality of vehicles directly via the wireless network.

17. The VMS of claim 13, wherein further instructions are stored in the memory that when executed, cause the processor to determine an optimal chunk size and compression strategy for data of the HD map update, based on conditions of the wireless network and edge node capabilities.

18. The VMS of claim 13, wherein the HD map update is transmitted to vehicles managed by the VMS via a combination of vehicle-to-vehicle (V2V) communication with other vehicles and vehicle-to-infrastructure / infrastructure-to-vehicle (V2I / I2V) communication with edge nodes of the VMS.

19. The VMS of claim 13, wherein further instructions are stored in the memory that when executed, cause the processor to: after transmitting the HD map update to the second portion of edge nodes, transmit the HD map update to one or more vehicles outside the primary geographical area and secondary geographical area via the wireless network.

20. A method for a vehicle management system (VMS) for updating a high-definition (HD) map used by a navigation system of a vehicle, the method comprising: receiving the HD map update; comparing the HD map update to an existing HD map of the VMS; extracting features of the HD map update that are different between the HD map update and the existing HD map using an artificial intelligence (Al) model; assigning an urgency score to each extracted feature; aggregating the urgency scores of each extracted feature to generate an overall priority score of the HD map update; in response to the priority score being greater than a threshold priority score: defining a primary geographical area where changes to the existing HD map have occurred, and defining a secondary geographical area where traffic may be redirected as a result of the changes; identifying a first portion of edge nodes of the VMS located within the primary geographical area, and a second portion of edge nodes of the VMS located within the secondary geographical area; transmitting the HD map update to the first portion of edge nodes, via a wireless network; after transmitting the HD map update to the first portion of edge nodes, transmitting the HD map update to the second portion of edge nodes, via the wireless network; and after transmitting the HD map update to the second portion of edge nodes, transmitting the HD map update to edge nodes of the VMS outside the primary geographical area and secondary geographical area via the wireless network; andin response to the priority score not exceeding the threshold priority score, transmitting the HD map update to a plurality of edge nodes of the VMS via the wireless network.

Images