Abstract: Methods and systems are provided for fusing responses of a classifier that provides a model uncertainty measure, while accounting for viewpoint-dependent variations in object appearance and correlations in classifier responses, and accounting for localization uncertainty.

Abstract: Methods and systems are provided for belief space planning for multiple robots at an initial belief state, for reducing belief space uncertainty, including receiving multiple, candidate, multi-robot paths, each including one or more pose constraints; generating a candidate factor graph for each candidate multi-robot path, and generating from each candidate factor graph a topology graph; calculating a signature metric from each topology graph, wherein the signature metric is one of a Von Neumann (VN) metric and a spanning tree (ST) metric; selecting a multi-robot path having the greatest signature metric from among the candidate, multi-robot paths; and instructing the multiple robots to proceed according to the selected multi-robot path.

Abstract: A system and methods are provided for decision making under uncertainty, for selecting an optimal action from among multiple candidate actions in belief space planning (BSP), including, for a new propagated belief: selecting from stored propagated beliefs a closest propagated belief; selecting, from among the stored measurement samples associated with the closest propagated belief, re-use measurement samples that provide a representative set of a measurement likelihood distribution corresponding to the new propagated belief; determining an information gap between a stored posterior belief associated with a re-use measurement sample and a new posterior belief that would be inferred by applying the re-use measurement sample to the new propagated belief; updating the stored posterior belief to account for the information gap; and calculating objective values for the multiple candidate actions, the objective values being weighted summations including immediate scores of the updated posterior beliefs.

Abstract: Methods and systems are provided for classifying an object appearing in multiple sequential images.

Abstract: Methods and systems are provided for fusing responses of a classifier that provides a model uncertainty measure, while accounting for viewpoint-dependent variations in object appearance and correlations in classifier responses, and accounting for localization uncertainty.

Abstract: A method for estimating a plurality of camera, comprising using at least one processor executing a code for: extracting a plurality of image features of a plurality of landmarks from a plurality of images captured by at least one camera from at least one pose, the plurality of landmarks calibrated with respect to a certain coordinate system; identifying among the plurality of image features at least one image feature documented in at least some of the images; producing scale values of at least one common image feature by analyzing the at least some of the images; determining a plurality of estimated poses of the at least one camera with respect to the certain coordinate system by using the scale values in calculating a minimal re-projection error between the plurality of image features and a plurality of predicted image features; and outputting the plurality of estimated poses.

Abstract: Methods and systems are provided for controlling a robot to navigate a perceptually aliased environment, the robot being configured to infer a first Gaussian mixture model (GMM) of a first pose of the robot, and to plan an optimal control action by: determining a second GMM of a simulated pose of the robot by simulating applying the first control action to the first GMM; determining from the second GMM a range of possible observations; calculating scene probabilities for associating each possible observation with each aliased scene of a set of aliased scenes; calculating from the scene probabilities an objective function to determine an estimated cost of the first control action; and comparing the estimated cost of the first control action with estimated costs calculated for one or more additional control actions, to determine an optimal control action.

Abstract: An autonomous system comprising: at least one hardware processor; a sensors module; and a non-transitory computer-readable storage medium having stored thereon program instructions, the program instructions executable by the at least one hardware processor to: receive, from said sensors module, a set of measurements associated with a joint state of said autonomous system, infer, based, at least in part, on said set of measurements, a current belief regarding said joint state of said autonomous system, determine a control action based on said inference, wherein said determining comprises calculating a future belief regarding a future joint state of said autonomous system, wherein said future joint state is as a result of said control action, execute said control action, and generate a new inference based, at least in part, on said future belief, wherein said future belief is updated based on a new set of measurements from said sensors module.

Abstract: A system and methods are provided for decision making under uncertainty, for selecting an optimal action from among multiple candidate actions in belief space planning (BSP), including, for a new propagated belief: selecting from stored propagated beliefs a closest propagated belief; selecting, from among the stored measurement samples associated with the closest propagated belief, re-use measurement samples that provide a representative set of a measurement likelihood distribution corresponding to the new propagated belief; determining an information gap between a stored posterior belief associated with a re-use measurement sample and a new posterior belief that would be inferred by applying the re-use measurement sample to the new propagated belief; updating the stored posterior belief to account for the information gap; and calculating objective values for the multiple candidate actions, the objective values being weighted summations including immediate scores of the updated posterior beliefs.

Abstract: Methods and systems are provided for belief space planning for multiple robots at an initial belief state, for reducing belief space uncertainty, including receiving multiple, candidate, multi-robot paths, each including one or more pose constraints; generating a candidate factor graph for each candidate multi-robot path, and generating from each candidate factor graph a topology graph; calculating a signature metric from each topology graph, wherein the signature metric is one of a Von Neumann (VN) metric and a spanning tree (ST) metric; selecting a multi-robot path having the greatest signature metric from among the candidate, multi-robot paths; and instructing the multiple robots to proceed according to the selected multi-robot path.

Abstract: A method for estimating a plurality of camera, comprising using at least one processor executing a code for: extracting a plurality of image features of a plurality of landmarks from a plurality of images captured by at least one camera from at least one pose, the plurality of landmarks calibrated with respect to a certain coordinate system; identifying among the plurality of image features at least one image feature documented in at least some of the images; producing scale values of at least one common image feature by analyzing the at least some of the images; determining a plurality of estimated poses of the at least one camera with respect to the certain coordinate system by using the scale values in calculating a minimal re-projection error between the plurality of image features and a plurality of predicted image features; and outputting the plurality of estimated poses.

Abstract: A method for estimating a plurality of camera, comprising using at least one processor executing a code for: extracting a plurality of image features of a plurality of landmarks from a plurality of images captured by at least one camera from at least one pose, the plurality of landmarks calibrated with respect to a certain coordinate system; identifying among the plurality of image features at least one image feature documented in at least some of the images; producing scale values of at least one common image feature by analyzing the at least some of the images; determining a plurality of estimated poses of the at least one camera with respect to the certain coordinate system by using the scale values in calculating a minimal re-projection error between the plurality of image features and a plurality of predicted image features; and outputting the plurality of estimated poses.

Abstract: Methods and systems are provided for controlling a robot to navigate a perceptually aliased environment, the robot being configured to infer a first Gaussian mixture model (GMM) of a first pose of the robot, and to plan an optimal control action by: determining a second GMM of a simulated pose of the robot by simulating applying the first control action to the first GMM; determining from the second GMM a range of possible observations; calculating scene probabilities for associating each possible observation with each aliased scene of a set of aliased scenes; calculating from the scene probabilities an objective function to determine an estimated cost of the first control action; and comparing the estimated cost of the first control action with estimated costs calculated for one or more additional control actions, to determine an optimal control action.

Abstract: A method for estimating a plurality of camera, comprising using at least one processor executing a code for: extracting a plurality of image features of a plurality of landmarks from a plurality of images captured by at least one camera from at least one pose, the plurality of landmarks calibrated with respect to a certain coordinate system; identifying among the plurality of image features at least one image feature documented in at least some of the images; producing scale values of at least one common image feature by analyzing the at least some of the images; determining a plurality of estimated poses of the at least one camera with respect to the certain coordinate system by using the scale values in calculating a minimal re-projection error between the plurality of image features and a plurality of predicted image features; and outputting the plurality of estimated poses.