Abstract: An aerial vehicle platform includes an aerial vehicle, a gimbal coupled to the aerial vehicle, and a camera mounted to the gimbal. An attitude sensing system includes an inertial measurement unit to sense attitude and an attitude adjustment module to generate an attitude adjustment for adjusting the sensed attitude to compensate for drift error.

Abstract: The disclosure describes systems and methods for a stabilization mechanism. The stabilization mechanism may be used in conjunction with an imaging device. The method may be performed by a control system of the stabilization mechanism and includes obtaining a device setting from an imaging device. The method may also include obtaining a configuration of the stabilization mechanism. The method includes determining a soft stop based on the device setting, the configuration, or both. The soft stop may be a virtual hard stop that indicates to the stabilization mechanism to reduce speed as a field of view of the imaging device approaches the soft stop. The method may also include setting an image stabilization mechanism parameter based on the determined soft stop.

Abstract: This disclosure relates to systems and methods for vehicle guidance. Stereo images may be obtained at different times using a stereo image sensor. A depth image may be determined based on an earlier obtained pair of stereo images. The depth image may be refined based on predictions of an earlier stereo image and a later obtained stereo image. Depth information for an environment around a vehicle may be obtained. The depth information may characterize distances between the vehicle and the environment around the vehicle. A spherical depth map may be generated from the depth information. Maneuver controls for the vehicle may be provided based on the spherical depth map.

Abstract: Controlling an unmanned aerial vehicle to traverse a portion of an operational environment of the unmanned aerial vehicle may include obtaining an object detection type, obtaining object detection input data, obtaining relative object orientation data based on the object detection type and the object detection input data, and performing a collision avoidance operation based on the relative object orientation data. The object detection type may be monocular object detection, which may include obtaining the relative object orientation data by obtaining motion data indicating a change of spatial location for the unmanned aerial vehicle between obtaining the first image and obtaining the second image based on searching along epipolar lines to obtain optical flow data.

Abstract: The disclosure describes systems and methods for stabilizing an imaging device with an image stabilization device. The image stabilization device includes sensors, one or more arms, one or more motors, and a control unit. The sensors provide sensor data including orientation data, angular velocity data, and acceleration data. Each of the motors is associated with a respective arm. The control unit is configured to set a setpoint of the imaging device to a default setpoint, receive sensor data from the sensors, determine whether a flip condition exists, in response to determining that the flip condition exists, fix the setpoint of the imaging device, and in response to determining that the flip condition does not exists, maintain the default setpoint of the imaging device by moving at least one of respective arms with one of the respective motors.

Abstract: Systems and methods are disclosed for image signal processing. For example, methods may include determining an orientation setpoint for an image sensor; based on a sequence of orientation estimates for the image sensor and the orientation setpoint, invoking a mechanical stabilization system to adjust an orientation of the image sensor toward the orientation setpoint; receiving an image from the image sensor; determining an orientation error between the orientation of the image sensor and the orientation setpoint during capture of the image; based on the orientation error, invoking an electronic image stabilization module to correct the image for a rotation corresponding to the orientation error to obtain a stabilized image; and storing, displaying, or transmitting an output image based on the stabilized image.

Abstract: This disclosure relates to systems and methods for vehicle guidance. Stereo images may be obtained at different times using a stereo image sensor. A depth image may be determined based on an earlier obtained pair of stereo images. The depth image may be refined based on predictions of an earlier stereo image and a later obtained stereo image. Depth information for an environment around a vehicle may be obtained. The depth information may characterize distances between the vehicle and the environment around the vehicle. A spherical depth map may be generated from the depth information. Maneuver controls for the vehicle may be provided based on the spherical depth map.

Abstract: The disclosure describes systems and methods for calibrating an image stabilization mechanism. One method includes a control system sending a command to thermally condition one or more sensors to a predetermined temperature. During thermal conditioning to the predetermined temperature, the control system sends a command to drive one or more motors of the image stabilization mechanism to cause movement of an imaging device coupled to the image stabilization mechanism. After thermal conditioning to the predetermined temperature, the control system sends a command to stop driving the one or more motors of the image stabilization mechanism to stop movement of the imaging device coupled to the image stabilization mechanism. After stopping the driving of the one or more motors, the control system sends a command to calibrate the one or more sensors.

Abstract: This disclosure describes systems and methods for vision-based navigation of an aerial vehicle. A method includes operations of acquiring a first image, and identifying features in a first image pyramid of the first image. The method includes acquiring a second image after the first image, and identifying the features in a second image pyramid of the second image. The method includes determining current navigation information of the aerial vehicle according to changes in position of the respective features between the first image pyramid and the second image pyramid. The method also includes predicting whether a sufficient number of the features will be identifiable at a third time that is in the future and after the second image was taken, and limiting flight as compared to the user flight instructions if an insufficient number of the features are predicted to be identifiable.

Abstract: An aerial vehicle platform includes an aerial vehicle, a gimbal coupled to the aerial vehicle, and a camera mounted to the gimbal. An attitude sensing system includes an inertial measurement unit to sense attitude and an attitude adjustment module to generate an attitude adjustment for adjusting the sensed attitude to compensate for drift error.

Abstract: An electronic gimbal is attached to a mounting platform and enables a mounted device such as a camera to rotate about three axes of rotation. The electronic gimbal may be configured with “soft stop” positions, past which the gimbal adjusts the target rotation to slow the rate of rotation as the motor approach mechanical stops.

Abstract: This disclosure relates to systems and methods for determining predicted risk for a flight path of an unmanned aerial vehicle. A previously stored three-dimensional representation of a user-selected location may be obtained. The three-dimensional representation may be derived from depth maps of the user-selected location generated during previous unmanned aerial vehicle flights. The three-dimensional representation may reflect a presence of objects and object existence accuracies for the individual objects. The object existence accuracies for the individual objects may provide information about accuracy of existence of the individual objects within the user-selected location. A user-created flight path may be obtained for a future unmanned aerial flight within the three-dimensional representation of the user-selected location. Predicted risk may be determined for individual portions of the user-created flight path based upon the three-dimensional representation of the user-selected location.

Abstract: This disclosure relates to unmanned aerial vehicles with parallax disparity detection offset from horizontal. The unmanned aerial vehicles use two optical elements, which are arranged to be separated by a horizontal distance and a vertical distance when the unmanned aerial vehicles operate leveled with respect to ground, to determine parallax disparity of an object.

Abstract: This disclosure relates to providing flight control for an unmanned aerial vehicle based on tilted optical elements. The UAV may include a housing, a motor, a first image sensor, a second image sensor, a first optical element having a first field of view, a second optical element having a second field of view, and one or more processors. The first optical element and the second optical element may be carried by the housing such that the vertical fields of view above the midline plane of the housing are greater than the vertical fields of view below the midline plane of the housing when the UAV is tilted during flight, and such that portions of the fields of view overlap. Flight control for the UAV may be provided based on parallax disparity of an object within the overlapping fields of view.