Abstract: A method, apparatus and computer program product are provided for learning to generate maps from raw geospatial observations from sensors traveling within an environment.

Abstract: System and methods for augmenting search results with external/user-provided data not contained in a primary datastore. On demand matching is used at run time to match entries in the primary datastore with information in a secondary datastore that is neither indexed nor pre-matched. A location-based query is received. Candidates are generated by searching a primary datastore. A machine learned entity matching network matches the candidates with information in the secondary datastore. The candidates are augmented with the information and returned as search results.

Abstract: A method, apparatus and computer program product are provided to incentivize crowd sourcing of data by identifying and compensating content contributors based on a value of the content to training a neural network. Methods may include: receiving a request for a machine learning model trained from training data received from a plurality of data contributors, where the training data identified a contributor having provided the respective training data; processing the request for the machine learning model to infer a result based on a subset of training data relevant to the request; identifying one or more data contributors that provided the subset of training data relevant to the request; and providing compensation to the one or more data contributors that provided the subset of training data.

Abstract: An apparatus, method and computer program product are provided for providing autonomous vehicle navigation at intersections. In one example, the apparatus identifies an intersection associated with at least two road signs having the same sign type and facing the same direction. The apparatus updates map data to include a datapoint that provides a representation of a single road sign at the intersection instead of the at least two road signs, thereby enabling an autonomous vehicle relying on the map data to traverse the intersection to adhere to the single road sign in lieu of the at least two road signs.

Abstract: An apparatus, method and computer program product are provided for determining a newly established vehicle-related rule within an area. In one example, the apparatus receives sensor data indicating attributes of an area for a first period. The attributes indicate a number of vehicle-related tickets issued within the area, a parking orientation of each vehicle within the area, or a combination thereof. The apparatus compares the sensor data to historical data associated with the area. The historical data indicate the attributes of the area for one or more second periods preceding the first period. Based on comparison of the sensor data and the historical data, the apparatus determines a likelihood of a vehicle-related rule established for the area, where the vehicle-related rule did not exist during the one or more second periods. The apparatus causes a notification indicating the likelihood at a user interface.

Abstract: A computer-implemented method can provide time-valid location-based information. The method comprises: for an environment, defining a plurality of cells, each cell comprising: a respective location with a range based on a geographical extent of the respective cell within the environment; a respective predetermined validity-time-interval; respective semantic-location-data relevant to the respective location; respective executable code instructions configured to operate on the respective semantic-location-data to enable determination of respective time-valid location-based information upon execution of the respective code instructions by a first edge computing apparatus proximal to the respective location. The respective time-valid location-based information is valid within the respective predetermined validity-time-interval.

Abstract: An apparatus, method and computer program product are provided for providing a map layer of noise levels. In one example, the apparatus receives input data including an image and location data indicating an area in which the image was captured. The apparatus causes a machine learning model to generate a datapoint in a map layer as a function of the input data. The datapoint indicate a predicted decibel level at the area during an instance in which the image was captured. The machine learning model is trained to generate the datapoint as a function of the input data based on training data, where the training data include images and audio data indicating decibel levels of areas in which the images were captured during instances in which the images were captured.

Abstract: A method, apparatus, and computer program product are provided for using confidential computing to execute code on sensitive data in an encrypted area of an apparatus limiting access to data and code to only their respective owners. Methods may include: generating an outer enclave and at least one inner enclave within the outer enclave; providing an outer enclave key and an inner enclave key to a service provider; providing an inner enclave key to a data provider; receiving, from the data provider, a data retrieval location; processing data from the respective retrieval location at the data provider inner enclave using data provider code to generate data provider processed data; providing the data provider processed data to the service provider inner enclave; and processing the data provider processed data with service provider code to generate resultant data; decrypting the resultant data in the outer enclave.

Abstract: Embodiments including a method and apparatus for correction of a global navigation satellite system (GNSS) are described. In one example, the apparatus includes a communication interface and a processor. The communication interface is configured to a plurality of GNSS signals. The GNSS signals may include at least one almanac value and at least one ephemeris value. The processor is configured to generate a spatio-temporal graph model based on the at least one almanac value, the at least one ephemeris value, and a predetermined offset value for a base location. The spatio-temporal graph model analyzes subsequent GNSS signals to determined a predicted offset or a corrected GNSS position.

Abstract: A method, apparatus, and computer program product are provided for real-time map-matching of probe data. Methods may include: receiving a plurality of probe data points; receiving an indication of changes to the road network, where the indication of changes includes an addition of new road segments and a removal of closed road segments; map-matching the plurality of probe data points to one or more road segments of the road network based on closest candidate road segments to form map-matched probe data; in response to a closest candidate road segment for one or more probe data points of the plurality of probe data points corresponding to a closed road segment, map-matching the one or more probe data points to a next closest candidate road segment in the map-matched probe data; and updating dynamic map data with the map-matched probe data.

Abstract: A system, a method and a computer program product are provided to determine a left turn decision for a user or autonomous vehicle attempting to turn at an intersection of a road. For example, the system obtains contextual features and/or sensor data related to the intersection of the road. An optimal location on the road is determined for the user or autonomous vehicle to position the vehicle to make the left turn decision, based on the contextual features and/or the sensor data. A left turn risk factor is determined based on the location on the road for the user or autonomous vehicle to position the vehicle to make a left turn. An alert is presented to the user or autonomous vehicle regarding the left turn risk factor with a left turn advisory message.

Abstract: A system, a method and a computer program product are provided to predict vehicle parking bunching in a geographic region. For example, the system is configured to obtain a plurality of contextual features and/or a plurality of sensor data related to parking information in the geographic region. The system is configured to predict a vehicle parking bunching based on a vehicle unparking threshold, a vehicle separation distance cluster threshold and/or unparking vehicle information. The system may also be configured to alert a vehicle of the vehicle parking bunching with a vehicle parking bunching notification.

Abstract: A system, a method and a computer program product to compute a pedestrian real-time vulnerability index are disclosed. For example, the system is configured to obtain static information related to the geographic region near the pedestrian and/or dynamic information related to the geographic region near the pedestrian. The system is configured to compute the real-time vulnerability index for the pedestrian based on the static information related to the geographic region near the pedestrian and/or the dynamic information related to the geographic region near the pedestrian. The system may alert the pedestrian to the real-time vulnerability index with a pedestrian advisory indication.

Abstract: A system, a method and a computer program product are provided, for example, to assist vehicle turns during limited visibility of incoming vehicles based on a turning confidence index for a left turn decision. For example, the system may obtain a plurality of traffic features related to the intersection of the road based on historical road features and/or real-time road features associated with the road. Using a trained machine learning model, a turning confidence index for the intersection of a road and/or a driving decision associated with attempting to turn at the intersection of the road may be determined.

Abstract: An apparatus, method and computer program product are provided for predicting tire temperature levels. In one example, the apparatus receives input data indicating attributes of the target vehicle and attributes of target routes for the target vehicle. The apparatus causes a machine learning model to generate output data as a function of the input data, where the output data indicate a subset of the target routes of which the target vehicle can successfully traverse while using the at least one spare tire. The machine learning model is trained to generate the output data as a function of the input data based on training data, where the training data indicate events in which vehicles used spare tires to traverse routes. The training data indicate attributes of the vehicles and attributes of the routes.

Abstract: A method is provided to using a machine learning model to predict lane geometry where incorrect or missing lane line geometry is detected. Methods may include: receiving a representation of lane line geometry including a masked area representing an area of broken lane line geometry; processing the representation using an inpainting model including a generator to produce a corrected representation, where the masked area is inpainted to correct the broken lane line geometry, where the generator processes the representation through one or more convolution layers and at least eight dilated convolution layers to generate inpainting of corrected lane line geometry in the corrected representation; processing the corrected representation through a discriminator of the inpainting model, where the discriminator discerns a score reflecting a quality measure of the corrected representations; and updating a map database based on the corrected representation in response to the score satisfying a predetermined threshold.

Abstract: Methods described herein relate to self-localization. Methods may include: receiving sensor data from a vehicle traveling along a road segment; identifying one or more objects of the environment from the sensor data; determining a coarse location of the vehicle; comparing the identified one or more objects with corresponding ground truth objects; calculating an observation in response to the comparison of the identified one or more objects with the corresponding ground truth objects; determining a location of the vehicle with a higher precision than the coarse location based, at least in part, on the observation; and allocating resources of the apparatus based, at least in part, on the observation.

Abstract: An approach is provided for identifying and characterizing mobile work zones (e.g., roadway striping, pothole filling, tree trimming, etc.). The approach, for example, involves processing probe data at a lane level to determine at least one forward forming shockwave associated with a congestion front of a mobile roadwork zone on at least one lane. The approach also involves providing data indicating the mobile roadwork zone as an output. The approach further involves determining a propagation rate of the mobile roadwork zone. The approach further involves determining a congestion recovery speed, where the congestion is caused by the mobile roadwork zone.

Abstract: A method, apparatus and computer program product are provided for establishing correspondences between images using through generation of a translation between the different perspectives. Methods may include: receiving first sensor data from a first image sensor, where the first sensor data includes a first image of an environment captured from a first perspective; receiving second sensor data from a second image sensor of a second image of the environment captured from a second perspective; identifying image correspondence points between the first sensor data and the second sensor data; computing pairwise vectors between corresponding pairs of first projected points of the first sensor data and second projected points of the second sensor data; clustering the pairwise vectors according to magnitude and orientation; and generating a translation vector from clusters of pairwise vectors, where the translation vector represents a shift of the second image sensor data to correspond to ground truth data.

Abstract: A system, method, etc. for providing a geopolitical region discovery service comprising obtaining an indication of at least a first location an end user device; wherein the indication of a first location is obtained at a predefined time interval, obtaining geopolitical compliance data for one or more geopolitical areas proximate to the first location of the end user device, determining a geopolitical interaction indicator based, at least in part, on the obtained indication of first location and geopolitical compliance data for one or more geopolitical areas proximate to the first location; and, generating at least one feedback for an end user device or software client.