Abstract: A method for controlling a drone includes receiving a natural language request for information about a spatial location, parsing the natural language request into data requests, configuring a flight plan and controlling one or more drones to fly over the spatial location to obtain data types based on the data requests, and extracting and analyzing data to answer the request. The method can include extracting data points from the data types, obtaining labels from a user for one or more of the data points, predicting labels for unlabeled data points from a learning algorithm using the labels obtained from the user, determining the predicted labels are true labels for the unlabeled data points and combining the extracted data, the user labeled data points and the true labeled data points to answer the request for information. The learning algorithm may be active learning using a support vector machine.

Abstract: Aspects of the invention relate to a UAV (drone) data marketplace where requests for UAV services are matched with registered UAVs. Drone operators register in the marketplace drones in their fleet with their capabilities and requesters of drone services make their request in the marketplace. The requests can be a set of data to be collected (e.g., optical images, NIR data, temperatures, etc.) and/or actions to be performed (e.g., deploying spare machinery parts to a tractor in the field), a location from which that data is collected and/or location to which a delivery is to be made, a pipeline of analytics to be performed on the data (e.g., optical recognition, NDVI computation, fertilizer application recommendation), a timeframe in which to collect the data (e.g., “by next week”, “by the end of today”), and a market value for how much the Requester is willing to pay for that data or operation.

Abstract: Systems and methods for storing in a first database a user personal profile, storing in a second database per-restaurant profiles for a plurality of restaurants, enabling the user to connect to a cognitive computer, enabling the user to interact with the cognitive computer for generating a personalized recipe based on user culinary selections and the user profile in the first database, the personalized recipe including a first list of ingredients, determining by the cognitive computer whether there are one or more first type candidate restaurants for preparing the personalized recipe based on the per-restaurant profiles in the second database, the first type candidate restaurant being determined to be able to prepare the personalized recipe with the first list of ingredients, receiving a selection of a selected restaurant from the first type candidate restaurant and contracting out the preparation of the personalized recipe to the selected restaurant.

Abstract: A database comprises historical information of a user's response to previous notifications. The database is accessed to determine a time at which to provide a (new) notification to the user, utilizing at least: a) current user activity status (e.g., determined from measurement information collected from one or more personal devices and/or user calendar events; b) time/day; and c) context information about the notification (e.g., geo-location, indoors/outdoors) including notification type (e.g., calendar entry, email, IM). The user gets the notification via a portable device at the determined time. A machine learning model can select the determined time by discriminating features of the previous notifications for which the user immediately attended versus those that were deferred and/or ignored. Content of the notification can also be altered in view of such discriminating features so as to increase a likelihood the user will immediately attend to the provided notification.

Abstract: Controlling drones and vehicles in package delivery, in one aspect, may include routing a delivery vehicle loaded with packages to a dropoff location based on executing on a hardware processor a spatial clustering of package destinations. A set of drones may be dispatched. A drone-to-package assignment is determined for the drones and the packages in the delivery vehicle. The drone is controlled to travel from the vehicle's dropoff location to transport the assigned package to a destination point and return to the dropoff location to meet the vehicle. The delivery vehicle may be alerted to speed up or slow down to meet the drone at the return location, for example, without the delivery vehicle having to stop and wait at the dropoff location while the drone is making its delivery.

Abstract: A method for applying GPS UAV attitude estimation to accelerate computer vision. The UAV has a plurality of GPS receivers mounted at fixed locations on the UAV. The method includes receiving raw GPS measurements from each GPS satellite in view of the UAV, the raw GPS measurements comprising pseudo-range and carrier phase data representing the distance between each GPS receiver and each GPS satellite. Carrier phase and pseudo-range measurements are determined for each GPS receiver based on the pseudo-range and carrier phase data. The GPS carrier phase and pseudo-range measurements are compared pair-wise for each pair of GPS receiver and satellite. An attitude of the UAV is determined based on the relative distance measurements. A 3D camera pose rotation matrix is determined based on the attitude of the UAV. Computer vision image search computations are performed for analyzing the image data received from the UAV in real time using the 3D camera pose rotation matrix.

Abstract: Filtering a set of social messages received in real time to yield a sub-set likely to relate to some first substantive content subject (such as a video available online). For each message in the sub-set, a respective social message sender, and a corresponding social influence value for each respective social message sender is determined. A prediction of the extent of the spread of the substantive content subject is made based on the social influence values of the social message senders.

Abstract: A method and system for determining a user's location in a pre-mapped environment from one or more images of the user's vicinity is provided. The method includes providing a probabilistic data structure derived from a second data structure, querying the second data structure for corresponding feature characteristics stored in the second data structure that respectively correspond to each feature characteristic from a plurality of feature characteristics, which may be a subset of an initial plurality of feature characteristics extracted from the one or more images from the user's vicinity, and identifying the user's location from the corresponding feature characteristics. The plurality of feature characteristics is determined by querying the probabilistic data structure.

Abstract: A method for controlling a drone includes receiving a natural language request for information about a spatial location, parsing the natural language request into data requests, configuring a flight plan and controlling one or more drones to fly over the spatial location to obtain data types based on the data requests, and extracting and analyzing data to answer the request. The method can include extracting data points from the data types, obtaining labels from a user for one or more of the data points, predicting labels for unlabeled data points from a learning algorithm using the labels obtained from the user, determining the predicted labels are true labels for the unlabeled data points and combining the extracted data, the user labeled data points and the true labeled data points to answer the request for information. The learning algorithm may be active learning using a support vector machine.

Abstract: A method includes defining a two-dimensional geographic region by two-dimensional geographic coordinates to define the bounds of the region, converting each of the two-dimensional coordinates to three dimensional coordinates by way of a lookup stored in a computer readable medium, generating a three-dimensional grid of points, each spaced in an arrangement to encompass coverage of a predetermined ground area, and applying heuristics for a shortest path planning, relative to the three-dimensional grid of points.

Abstract: In an approach to cooperative evidence gathering, a first computing device receives a request for data corresponding to an event, where a second computing device detecting the event initiates the request for data. The first computing device aggregates data from one or more sensors, the one or more sensors associated with one or more first computing devices within a proximity of a location of the event. The first computing device determines whether at least a portion of the aggregated data is applicable to the event.

Abstract: In an approach to cooperative evidence gathering, a first computing device receives a request for data corresponding to an event, where a second computing device detecting the event initiates the request for data. The first computing device aggregates data from one or more sensors, the one or more sensors associated with one or more first computing devices within a proximity of a location of the event. The first computing device determines whether at least a portion of the aggregated data is applicable to the event.

Abstract: A method and system for determining a user's location in a pre-mapped environment from one or more images of the user's vicinity is provided. The method includes providing a probabilistic data structure derived from a second data structure, querying the second data structure for corresponding feature characteristics stored in the second data structure that respectively correspond to each feature characteristic from a plurality of feature characteristics, which may be a subset of an initial plurality of feature characteristics extracted from the one or more images from the user's vicinity, and identifying the user's location from the corresponding feature characteristics. The plurality of feature characteristics is determined by querying the probabilistic data structure.

Abstract: A method for clustering images includes acquiring initial image data including a scene of interest. A 3D model is constructed of the scene of interest based on the acquired initial image data. Additional image data including the scene of interest is acquired. The additional image data is fitted to the 3D model. A line-of-sight of the additional image data is estimated based on the fitting to the 3D model. The additional image data is clustered according to the estimated line-of-sight.

Abstract: A method for clustering images includes acquiring initial image data including a scene of interest. A 3D model is constructed of the scene of interest based on the acquired initial image data. Additional image data including the scene of interest is acquired. The additional image data is fitted to the 3D model. A line-of-sight of the additional image data is estimated based on the fitting to the 3D model. The additional image data is clustered according to the estimated line-of-sight.

Abstract: A method for controlling a drone includes receiving a natural language request for information about a spatial location, parsing the natural language request into data requests, configuring a flight plan and controlling one or more drones to fly over the spatial location to obtain data types based on the data requests, and extracting and analyzing data to answer the request. The method can include extracting data points from the data types, obtaining labels from a user for one or more of the data points, predicting labels for unlabeled data points from a learning algorithm using the labels obtained from the user, determining the predicted labels are true labels for the unlabeled data points and combining the extracted data, the user labeled data points and the true labeled data points to answer the request for information. The learning algorithm may be active learning using a support vector machine.

Abstract: Filtering a set of social messages received in real time to yield a sub-set likely to relate to some first substantive content subject (such as a video available online). For each message in the sub-set, a respective social message sender, and a corresponding social influence value for each respective social message sender is determined. A prediction of the extent of the spread of the substantive content subject is made based on the social influence values of the social message senders.

Abstract: A method for clustering images includes acquiring initial image data including a scene of interest. A 3D model is constructed of the scene of interest based on the acquired initial image data. Additional image data including the scene of interest is acquired. The additional image data is fitted to the 3D model. A line-of-sight of the additional image data is estimated based on the fitting to the 3D model. The additional image data is clustered according to the estimated line-of-sight.

Abstract: A method for clustering images includes acquiring initial image data including a scene of interest. A 3D model is constructed of the scene of interest based on the acquired initial image data. Additional image data including the scene of interest is acquired. The additional image data is fitted to the 3D model. A line-of-sight of the additional image data is estimated based on the fitting to the 3D model. The additional image data is clustered according to the estimated line-of-sight.