Abstract: A method comprises determining a plurality of available parking spaces within a threshold distance of a destination; determining an initial utility score for each of the available parking spaces based on a distance between each of the available parking spaces and the destination; determining a safety score associated with each of the available parking spaces based on crime data and a distance between each of the parking spaces and a major street, an entrance to a parking facility in which each available parking space is located, or an entrance to a point of interest; determining an adjusted utility score for each of the available parking spaces based on the initial utility score and the safety score associated with each of the available parking spaces; and selecting a parking space among the plurality of available parking spaces based on the adjusted utility score for the parking space.

Abstract: A method for performing correlation-based parking availability estimation is provided. The method includes retrieving a plurality of parking availability correlation models, wherein each of the parking availability correlation models estimates parking availability for a target parking block based on parking availability for one or more other parking blocks; selecting a subset of the parking availability correlation models, wherein the subset comprises parking availability correlation models that estimate parking availability of the target parking block based on parking availability for one or more other parking blocks for which current parking availability is known; selecting a parking availability correlation model from the subset based on a model quality indicator associated with each parking availability correlation model; and estimating parking availability for the target parking block based on the selected parking availability correlation model.

Abstract: A parking seeker detection system and method for updating an availability of one or more parking spots of a parking spot database is provided. The method includes determining whether a target vehicle is a registered member vehicle, and in response to determining that the target vehicle is not a registered member vehicle, identifying a target parking spot in which the target vehicle is intending to park and updating an availability of a parking spot of a parking spot database corresponding to the target parking spot.

Abstract: A method for optimizing a parking spot database is provided including determining a utility score of a candidate route to a candidate parking spot of a plurality of parking spots of a parking spot database based on an availability status of a non-candidate parking spot along the candidate route, and selecting the candidate route from a plurality of candidate routes based on the utility score.

Abstract: A system for guiding an object using a robot is provided. The system includes a server and a robot. The server is configured to receive a request and a location of an object from a device of the object, instruct a vehicle to move to a first location based on the location of the object in response to receiving the request, obtain information about a second location related to a robot, and transmit information about the second location to the device of the object. The robot is configured to identify the object at the second location, and guide the object from the second location to the first location in response to identifying the object at the second location.

Abstract: A method receives a first image set depicting a merging zone, the first image set including first image(s) associated with a first timestamp; determines, using a trained first machine learning logic, a first state describing a traffic condition of the merging zone at the first timestamp using the first image set; determines, from a sequence of states describing the traffic condition of the merging zone at a sequence of timestamps, using a trained second machine learning logic, second state(s) associated with second timestamp(s) prior to the first timestamp of the first state using a trained backward time distance; computes, using a trained third machine learning logic, impact metric(s) for merging action(s) using the first state, the second state(s), and the merging action(s); selects, from the merging action(s), a first merging action based on the impact metric(s); and provides a merging instruction including the first merging action to a merging vehicle.

Abstract: A vehicular micro cloud includes a set of connected vehicles that are operable to provide computational services to the set of connected vehicles. The disclosure includes embodiments for mobility-oriented data replication in the vehicular micro cloud. In some embodiments, a method includes, for each data set stored by the set of connected vehicles, determining a number of replicas to generate based on one or more mobility-based criteria. The method includes generating instances of replica data that describe the replicas. The method includes, for individual instances of replica data, determining which of the connected vehicles included in the set to use as storage locations for the individual instances of replica data based on the one or more mobility-based criteria. The method includes causing the individual instances of replica data to be stored in the storage locations. For example, the individual instances of replica data are transmitted to the storage locations.

Abstract: A method of improving efficiency of a vehicle behavior controller using a traffic state estimation network is described. The method includes feeding an input of a feature extraction network of the vehicle behavior controller with a sequence of images. The sequence of images include a highway section and corresponding traffic data. The method also includes disentangling an estimated behavior of a controlled ego vehicle. by the traffic state estimation network. The traffic state estimate network disentangles the estimated of the controlled ego vehicle from extracted traffic state features of the input provided by the feature extraction network. The method further includes selecting an action to adjust an autonomous behavior of the controlled ego vehicle according to the estimated behavior of the controlled ego vehicle.

Abstract: The disclosure includes embodiments for managing vehicles under anomalous driving behavior. In some embodiments, a method includes determining, by the processor, an ego behavior associated with the ego vehicle and a remote behavior associated with a remote vehicle. The method includes calculating a variance between the ego behavior and the remote behavior. The method includes determining a presence of an anomaly based on the variance satisfying a threshold, wherein the satisfying the threshold indicates the ego behavior is incompatible with the remote behavior. The method includes generating a driving management plan which is configured to mitigate the anomaly and is consistent with the ego behavior. The method includes implementing the driving management plan so that the threshold is no longer satisfied.

Abstract: The disclosure includes embodiments for transforming anomalous behavior into model behavior to mitigate the anomalous behavior. In some embodiments, a method includes analyzing sensor data to identify an occurrence of anomalous behavior in a roadway environment. The method includes determining a model behavior which is non-anomalous. The method includes determining a number of vehicles that is optimal to transform anomalous behavior into the model behavior. The method includes forming a vehicular micro cloud which complies with the number of vehicles that is optimal. The formation of the vehicular micro cloud is triggered by the identification of the anomalous behavior. The method includes providing individualized control messages on a vehicle-by-vehicle basis to members of the vehicular micro cloud. The control messages include digital data which instructs the members on how to behave in order to transform the anomalous behavior into the model behavior.

Abstract: System, methods, and other embodiments described herein relate to detecting obstacles from unknown objects during automated driving by a vehicle. In one embodiment, a method includes generating, from an image that includes an unknown object, a depth map from a depth estimation component and processed data from a semantic segmentation component in parallel by using a neural network model. The method also includes detecting that the unknown object is an obstacle when the unknown object satisfies criteria using an optical model according to the depth map and a segmentation map. The method also includes determining a height of the obstacle and a distance to the obstacle according to the optical model and the criteria. The method also includes adapting a vehicle plan of the automated driving according to the height.

Abstract: Anomalous events in a driving environment can be detected and/or validated by leveraging inherent human behavior. A notification of the anomalous event can be received from an initial connected entity in the driving environment. In response, an inquiry can be sent to connected entities spatiotemporally related to the initial connected entity as to whether any person associated with the connected entities is exhibiting an inherent human behavior. When a threshold level of responses to the inquiry are received from the connected entities indicating that an inherent human behavior has been detected, the one or more connected entities can be caused to upload driving scene data for at least a time when the inherent human behavior by the occupant of the connected vehicle was detected.

Abstract: The disclosure includes embodiments re-identification and revocation for misbehaving vehicle detection. A misbehaving vehicle is an example of an anomaly. A method includes receiving from a vehicular micro cloud, by a connected vehicle that is a member of the vehicular micro cloud, vehicular micro cloud data describing an anomaly and a first state of the anomaly in a geographic area. The method includes causing a sensor to record sensor data describing the geographic area. The method includes determining, based on the sensor data and the vehicular micro cloud data, whether the anomaly is still present and in the first state. The method includes generating evidence data describing a second state of the anomaly. The method includes transmitting the evidence data to the vehicular micro cloud. The method includes modifying an operation of the connected vehicle based on the evidence data.

Abstract: In an example, a method determines a first controllable vehicle traveling along a mitigation road segment of a roadway and determines a control lane in the mitigation road segment. The control lane includes the first controllable vehicle and is impedible by the first controllable vehicle. The method determines a first open lane in the mitigation road segment and applies a target mitigation speed to the first controllable vehicle in the control lane. The first open lane is adjacent to the control lane in the mitigation road segment and the target mitigation speed is based on a traffic state of the first open lane. The target mitigation speed adjusts a traffic stream that flows through the first open lane to mitigate traffic congestion located downstream of the mitigation road segment.

Abstract: Anomalous events in a driving environment can be detected and/or validated by leveraging inherent human behavior. A notification of the anomalous event can be received from an initial connected entity in the driving environment. In response, an inquiry can be sent to connected entities spatiotemporally related to the initial connected entity as to whether any person associated with the connected entities is exhibiting an inherent human behavior. When a threshold level of responses to the inquiry are received from the connected entities indicating that an inherent human behavior has been detected, the one or more connected entities can be caused to upload driving scene data for at least a time when the inherent human behavior by the occupant of the connected vehicle was detected.

Abstract: An ego vehicle can be assisted when parking in a parking area. One or more sensors can be configured to acquire environment data of at least a portion of an external environment of the ego vehicle. The environment data can include at least a portion of the parking area. Parking availability data can be acquired for at least a portion of the parking area. The parking availability data can include a plurality of parking spaces and one or more vehicles parked in one or more of the parking spaces. The parking availability data can further include one or more attributes of the vehicles parked in the parking spaces. A current location of the ego vehicle within the parking area can be determined based on at least the parking availability data and the environment data.

Abstract: A method for generating a view for an unmanned aerial vehicle is provided. The method includes obtaining an origin and a destination of the unmanned aerial vehicle, determining a group of imaging devices based on a route between the origin and the destination of the unmanned aerial vehicle, and obtaining a view of the unmanned aerial vehicle following the route based on images of the unmanned aerial vehicle captured by the group of imaging devices. The group of imaging devices form one or more collaborative camera sensor networks. The collaborative camera sensor networks identify an unmanned aerial vehicle and create human operator's view of the unmanned aerial vehicle. The route of the unmanned aerial vehicle is determined based on the availability of imaging devices of the camera sensor network. The route of the unmanned aerial vehicle may be changed based on the availability of imaging devices.

Abstract: A method of creating a parking facility map includes obtaining one or more images of a parking facility, generating an estimate map of parking spaces in the parking facility based at least on the one or more images (the estimate map of parking spaces includes a plurality of estimate parking space shapes), identifying a parking block that includes a cluster of estimate parking spaces among the plurality of estimate parking space shapes, determining a common geometric pattern among the cluster of estimate parking spaces, correcting geometrical features of one or more estimate parking spaces in the cluster of estimate parking spaces based at least in part on the common geometric pattern, identifying one or more gaps among the cluster of estimate parking space shapes, and determining a number of inserted parking spaces to fill the one or more gaps based at least in part on the common geometric pattern.

Abstract: A sensor control system for a vehicle includes one or more processors and a memory communicably coupled to the one or more processors. The memory stores a zone of interest determination module including computer-readable instructions that when executed by the one or more processors cause the one or more processors to determine if at least one localized zone of interest resides within a preliminary zone of interest of the vehicle. The memory also stores a sensor control module including computer-readable instructions that when executed by the one or more processors cause the one or more processors to control one or more operational parameters of at least one sensor so as to, if at least one localized zone of interest resides within the preliminary zone of interest, bring at least a portion of the at least one localized zone of interest into a of field view of the at least one sensor.

Abstract: A system and method for controlling one or more vehicles with one or more controlled vehicles may include one or more processors and a memory in communication with the one or more processors. The memory may include one or more modules that cause the one or more modules to obtain a state of an environment having a universe of vehicles operating therein, identify one or more anomaly vehicles from the universe of vehicles operating in the environment, select one more actions to control a plurality of controlled vehicles to control the operation of one or more anomaly vehicles and direct the plurality of controlled vehicles execute the one or more actions. The selecting of one or more actions may be performed by utilizing a reinforcement-learning trained algorithm.