Abstract: Systems and methods for predicting an optimal use level of a driver assist system are provided. The system may collect historical driver behavior and road condition data, and train an optimization prediction model using the historical data, e.g., via machine learning or artificial intelligence. Moreover, the system may collect real-time driver behavior and road condition data, and predict an optimal use level of the driver assist system based on the real-time data using the trained optimization prediction model. The system may then send feedback to the driver assist system when the optimal use level falls outside a predetermined threshold, such that the driver assist system may be unavailable or have reduced functionality until the optimal use level improves.

Abstract: A method includes identifying a broadcast set of access points from among the plurality of access points based on access point interference data associated with a plurality of access points disposed in an environment, where the access point interference data includes current radio frequency (RF) signal interference data associated with each of the plurality of access points, predicted RF signal interference data associated with each of the plurality of access points, or a combination thereof. The method includes partitioning a data packet into a plurality of data subpackets based on the broadcast set of access points and broadcasting the plurality of data subpackets to a vehicle via the broadcast set of access points.

Abstract: A system includes a server computer programmed upon determining that a first portion of software data for updating an operational feature of a first computer is stored in the first computer and a second portion of the software data is stored in a second computer, to encode the first portion and the second portion to generate encoded data, and to send the encoded data via wireless data transfer to the first and second computers. The first computer is programmed to decode the second portion from the received encoded data, to update the operational feature of the first computer based on the stored first portion and the decoded second portion, and to operate the first computer based on the updated operational feature.

Abstract: System and methods for a decentralized hybrid air-ground autonomous last-mile goods delivery are disclosed herein. A drone can include a controller configured to cause the drone to depart from a docking station of a vehicle, transmit a discovery message to available vehicles in an operating area, the discovery message having drone metadata, select at least one of one of the available vehicles based on response codes received from the available vehicle, or a nearest fixed docking station, and dock with a docking station of the one of the available vehicles or the nearest fixed docking station.

Abstract: Systems, methods, and computer-readable media are described for performing real-time vehicular incident risk prediction using real-time vehicle-to-everything (V2X) data. A vehicular incident risk prediction machine learning model is trained using historical V2X data such as historical incident data and historical vehicle operator driving pattern behavior data as well as third-party data such as environmental condition data and infrastructure condition data. The trained machine learning model is then used to predict the risk of an incident for a vehicle on a roadway segment based on real-time V2X data relating to the roadway segment and/or vehicle operators on the roadway segment. A notification of a high risk of incident can then be sent to a V2X communication device of the vehicle to inform an operator of the vehicle.

Abstract: Routes may be presented to a vehicle operator with information about predicted autonomous driving levels. A road network route to a destination is received. The road network route is partitioned into segments. Indicia of the segments are provided to an inference engine. The inference engine predicts levels of autonomous driving for the respective segments generated by the inference engine based on the indicia of the segments. Based on the predicted levels of autonomous driving for the respective segments, a ratio of a level of autonomous driving for the road network route is computed. The ratio corresponds to a proportion of time and/or distance for the road network route during which a driver-system is predicted to be engaged at the level of autonomous driving. A user interface is displayed, which includes a graphic representation of the road network route and a graphic indication of the ratio.

Abstract: A system includes a server computer programmed upon determining that a first portion of software data for updating an operational feature of a first computer is stored in the first computer and a second portion of the software data is stored in a second computer, to encode the first portion and the second portion to generate encoded data, and to send the encoded data via wireless data transfer to the first and second computers. The first computer is programmed to decode the second portion from the received encoded data, to update the operational feature of the first computer based on the stored first portion and the decoded second portion, and to operate the first computer based on the updated operational feature.

Abstract: A method includes identifying a broadcast set of access points from among the plurality of access points based on access point interference data associated with a plurality of access points disposed in an environment, where the access point interference data includes current radio frequency (RF) signal interference data associated with each of the plurality of access points, predicted RF signal interference data associated with each of the plurality of access points, or a combination thereof. The method includes partitioning a data packet into a plurality of data subpackets based on the broadcast set of access points and broadcasting the plurality of data subpackets to a vehicle via the broadcast set of access points.

Abstract: Systems and methods for predicting an optimal use level of a driver assist system are provided. The system may collect historical driver behavior and road condition data, and train an optimization prediction model using the historical data, e.g., via machine learning or artificial intelligence. Moreover, the system may collect real-time driver behavior and road condition data, and predict an optimal use level of the driver assist system based on the real-time data using the trained optimization prediction model. The system may then send feedback to the driver assist system when the optimal use level falls outside a predetermined threshold, such that the driver assist system may be unavailable or have reduced functionality until the optimal use level improves.

Abstract: System and methods for a decentralized hybrid air-ground autonomous last-mile goods delivery are disclosed herein. A drone can include a controller configured to cause the drone to depart from a docking station of a vehicle, transmit a discovery message to available vehicles in an operating area, the discovery message having drone metadata, select at least one of one of the available vehicles based on response codes received from the available vehicle, or a nearest fixed docking station, and dock with a docking station of the one of the available vehicles or the nearest fixed docking station.

Abstract: A V2X event-message dictionary and a binning function are received from a cloud server. Sensor data is compared to events specified in the V2X event-message dictionary to identify a best-fit event for the sensor data. A number of bins and a bin number for the event are computed using the binning function. A V2X message is transmitted including the number of bins and the bin number, thereby avoiding including the sensor data in the V2X message.

Abstract: A graph of devices is constructed, each network device serving an amount of bandwidth over the network, each vertex of the graph corresponding to a respective one of the network devices, each edge of the graph connecting network devices by interference weight, such that edges between connected network devices using different channels have an interference weight of zero, and edges between connected network devices using the same channel have an interference weight denoting an amount of interference between the connected network devices. A cell utilization of each of the network devices is determined according to an amount of traffic served by the respective network device compared to a total of the amounts of bandwidth for all network devices. A vehicle requesting bandwidth is assigned to the one of the network devices having a smallest product of cell utilization and maximum interference weight edge.

Abstract: A graph of devices is constructed, each network device serving an amount of bandwidth over the network, each vertex of the graph corresponding to a respective one of the network devices, each edge of the graph connecting network devices by interference weight, such that edges between connected network devices using different channels have an interference weight of zero, and edges between connected network devices using the same channel have an interference weight denoting an amount of interference between the connected network devices. A cell utilization of each of the network devices is determined according to an amount of traffic served by the respective network device compared to a total of the amounts of bandwidth for all network devices. A vehicle requesting bandwidth is assigned to the one of the network devices having a smallest product of cell utilization and maximum interference weight edge.

Abstract: A computer is programmed to divide respective data of a first high-resolution map of a first geographic area and a second high-resolution map of a second geographic area into a plurality of respective subsets, assign one of the subsets of each of the first high-resolution map and the second high-resolution map to a first vehicle and a second vehicle, identify respective locations of the first vehicle and the second vehicle and the one of the first high-resolution map or the second high-resolution map that includes the locations of the first and second vehicles, and send, to the first and second vehicles, a map dataset that is a result of applying an XOR function to (1) the subset of the identified high-resolution map including the location of the first vehicle and (2) the subset of the identified high-resolution map including the location of the second vehicle.

Abstract: A graph of devices is constructed, each device requiring an amount of bandwidth over the network, each vertex of the graph corresponding to a respective one of the devices, each edge of the graph connecting devices according to a weight denoting interference between the connected devices. The vertices of the graph are labeled with labels such that no two vertices sharing the same edge have the same label, each label requiring bandwidth corresponding to the device of that label requiring the most bandwidth. If the sum of the bandwidth required for the labels exceeds the set bandwidth, the edge of the graph having the least interference is deleted, and the perform the construct and label operations are repeated. If the sum of the bandwidth required for the labels is within the set bandwidth, the bandwidth is assigned to the devices.

Abstract: A graph of devices is constructed, each device requiring an amount of bandwidth over the network, each vertex of the graph corresponding to a respective one of the devices, each edge of the graph connecting devices according to a weight denoting interference between the connected devices. The vertices of the graph are labeled with labels such that no two vertices sharing the same edge have the same label, each label requiring bandwidth corresponding to the device of that label requiring the most bandwidth. If the sum of the bandwidth required for the labels exceeds the set bandwidth, the edge of the graph having the least interference is deleted, and the perform the construct and label operations are repeated. If the sum of the bandwidth required for the labels is within the set bandwidth, the bandwidth is assigned to the devices.

Abstract: A computer is programmed to divide respective data of a first high-resolution map of a first geographic area and a second high-resolution map of a second geographic area into a plurality of respective subsets, assign one of the subsets of each of the first high-resolution map and the second high-resolution map to a first vehicle and a second vehicle, identify respective locations of the first vehicle and the second vehicle and the one of the first high-resolution map or the second high-resolution map that includes the locations of the first and second vehicles, and send, to the first and second vehicles, a map dataset that is a result of applying an XOR function to (1) the subset of the identified high-resolution map including the location of the first vehicle and (2) the subset of the identified high-resolution map including the location of the second vehicle.

Abstract: Systems, methods, and computer-readable media are described for performing real-time vehicular incident risk prediction using real-time vehicle-to-everything (V2X) data. A vehicular incident risk prediction machine learning model is trained using historical V2X data such as historical incident data and historical vehicle operator driving pattern behavior data as well as third-party data such as environmental condition data and infrastructure condition data. The trained machine learning model is then used to predict the risk of an incident for a vehicle on a roadway segment based on real-time V2X data relating to the roadway segment and/or vehicle operators on the roadway segment. A notification of a high risk of incident can then be sent to a V2X communication device of the vehicle to inform an operator of the vehicle.