Abstract:
A method includes receiving a plurality of images from an imaging device carried by a movable object. The method further includes determining whether an environment type of the movable object is a first type or a second type according to sensing data received from one or more sensors of the movable object, identifying a target based on the plurality of images using a first method in response to determining that the environment type is the first type and using a second method different from the first method in response to determining that the environment type is the second type, and determining state information of the movable object or the target based on the plurality of images.
Abstract:
A method for generating a first person view (FPV) of an environment includes, with aid of one or more processors individually or collectively, analyzing stereoscopic video data of the environment to determine environmental information and generating augmented stereoscopic video data of the environment by fusing the stereoscopic video data and the environmental information.
Abstract:
The present disclosure relates to systems, methods, and computer-readable storage devices for the coordination of actions between movable objects. For example, a method may coordinate actions between at least a first and a second movable object. The method may detect, by the first movable object, the position of a target. The method may control the position of a first movable object based on the position of the target. A command may be received for the first movable object to perform an action in coordination with the second movable object. In some embodiments, the command may be to take images of the target to create a bullet time effect.
Abstract:
A method for supporting target tracking includes obtaining a feature model for a target that represents imagery characteristics of the target, extracting one or more features from one or more images captured by an imaging device carried by a movable object, and applying the feature model on the one or more features to determine similarity between the one or more features and the feature model.
Abstract:
A method for controlling a movable object includes receiving a user input indicative of a command to adjust a perception of a target while tracking the target, determining a subsequent perception of the target based on the user input, and generating one or more control signals to move the movable object based on the subsequent perception of the target.
Abstract:
A delivering method using an unmanned aerial vehicle includes configuring a pre-set condition, locking a cargo, acquiring verification information about a recipient, comparing the verification information with the pre-set condition, and determining that verification is passed to unlock the cargo when the verification information is consistent with the pre-set condition.
Abstract:
A system for establishing a baseline of a stereoscopic imaging device having a microlens array and methods for making and using the same. The system acquires an object distance between the microlens array and an object of interest and selects first and second lenses from the microlens array based upon the acquired object distance. The system likewise can perform simultaneous localization and mapping (SLAM) with the imaging device. In one embodiment, the system can acquire first and second stereoscopic frames with the microlens array. The system thereby can measure rotations of the second stereoscopic frame with an Inertial Measurement Unit (IMU) and match the first and second stereoscopic frames by combining the rotation data with the first and second stereoscopic frames. The system thereby can enable SLAM systems to perform more accurately and more practically in various indoor and/or outdoor environments.
Abstract:
A method for controlling a movable object includes determining, based on current location information of an obstacle, whether the obstacle is located within a certain distance range from the movable object, and in response to determining that the obstacle is located within the certain distance range from the movable object: determining, based on the current location information of the obstacle, whether the obstacle meets a first distance criterion or a second distance criterion, to obtain a determination result, and performing a first adjustment or a second adjustment on one or more movement characteristics of the movable object based on the determination result. The first adjustment is different from the second adjustment. One of the first adjustment and the second adjustment includes changing a movement direction of the movable object from a first direction to a second direction. The first direction is opposite to the second direction.
Abstract:
Systems, methods, and computer-readable storage devices for obstacle detection may include an exemplary method of obstacle detection by a computing device. The method includes receiving a first dataset indicating a first surrounding with a first perspective; receiving a second dataset indicating a second surrounding with a second perspective; and generating a composite dataset from the first and second datasets. The method additionally includes identifying an obstacle using the composite dataset.
Abstract:
A movable object for detecting an obstacle includes a first passive infrared sensor having a first detection range and a first field of view, and one or more second passive infrared sensors each having a second detection range and a second field of view. The second detection range is longer than the first detection range and the second field of view is smaller than the first field of view. The movable object further includes one or more processors configured to calculate a distance from the movable object to the obstacle based on data from at least one of the first passive infrared sensor or the one or more second passive infrared sensors, and determine whether to effect a collision avoidance maneuver for the movable object to avoid the obstacle based on the distance.