Abstract: Structure inspections are performed with a high resolution by first performing a modeling flight path at a relatively high altitude over the structure. Images are gathered during the modeling flight path and are used to generate a three dimensional model of the structure. From the three dimensional model a lower altitude closer inspection flight path is defined and executed.

Abstract: Provided herein are systems and methods for providing reliable control of an unmanned aerial vehicle (UAV). A system for providing reliable control of the UAV can include a computing device that can execute reliable and unreliable programs. The unreliable programs can be isolated from the reliable programs by virtue of executing one or more of the programs in a virtual machine client. The UAV can initiate a recovery action when one or more of the unreliable programs fail. The recovery action can be performed without input from one or more of the unreliable programs.

Abstract: An unmanned aerial vehicle includes a camera, one or more sensors, memory storing first instructions that define an overall mission, and memory storing one or more mission cues. The vehicle further includes one or more processors configured to execute a first part of the first instructions to perform a first part of the overall mission. The processors are configured to process at least one of the image data and the sensor data to detect a presence of at least one of the mission cues. The processors are configured to, in response to detecting a mission cue, interrupting execution of the first instructions and executing second instructions to control the unmanned aerial vehicle to perform a first sub-mission of the overall mission. The processors are configured to after executing the second instructions, performing a second part of the overall mission by executing a second part of the first instructions.

Abstract: Provided herein are systems and method for autonomously or semi-autonomously landing an unmanned aerial vehicle (UAV) on a landing pad. The landing pad can include features configured to correct misalignment of the UAV on the landing pad. The landing pad can additionally include one or more markers than can be identified by the UAV to aid the UAV in locating the landing pad and determining the location of the UAV relative to the landing pad.

Abstract: An unmanned aerial vehicle responds to mission cues during a mission. The mission cues are characteristics of image and/or sensor data. The unmanned aerial vehicle may change data gathering operations or may perform sub-missions within a mission in response to the mission cues.

Abstract: An unmanned aerial vehicle includes a closely integrated emergency recovery and operation systems for an unmanned aerial vehicle with built-in levels of redundancy and independence to maximize the likelihood of a controlled velocity landing. The unmanned aerial vehicle may include multiple processors and multiple state estimating modules such as inertial measurement units to independently determine the operational and error status of the unmanned aerial vehicle. Base on predictive or projected computations, the emergency recovery system may determine a suitable time for a recovery action, such as parachute deployment, and execute the recovery action.

Abstract: Topology based adaptive data gathering is disclosed herein. Payload data gathering by an unmanned aerial vehicle can be adjusted based on topological or topographical characteristics of the area of flight by the unmanned aerial vehicle. The unmanned aerial vehicle collects payload data over an area and may scale up the rate of payload data gathering or slow down the flight as the unmanned aerial vehicle flies over a high or complex structure. Conversely, the unmanned aerial vehicle may advantageously scale down the rate of payload data gathering or speed up the flight as the unmanned aerial vehicle flies over a simple structure or an empty area.

Abstract: Provided herein are systems and methods for providing reliable control of an unmanned aerial vehicle (UAV). A system for providing reliable control of the UAV can include a computing device that can execute reliable and unreliable programs. The unreliable programs can be isolated from the reliable programs by virtue of executing one or more of the programs in a virtual machine client. The UAV can initiate a recovery action when one or more of the unreliable programs fail. The recovery action can be performed without input from one or more of the unreliable programs.

Abstract: A data communication system for unmanned aerial vehicles includes communication links comprising a low-throughput capacity communication link and a high-throughput capacity communication link. The data communication system can also include a base station, to which the unmanned aerial vehicles send aerial data, and from which the unmanned aerial vehicles receive command signals. As the unmanned aerial vehicles perform missions in an open, distant airspace, the unmanned aerial vehicles can gather large volume data such as aerial images or videos. The data communication system allows opportunistic transfer of the gathered aerial data from the unmanned aerial vehicles to the base station when a high-throughput communication link is established. The data communication system allows constant communication between the base station and the unmanned aerial vehicles to send and receive low volume, operation-critical data, such as commands or on-going flight path changes, using a low-throughput communication link.

Abstract: An enhanced distance detection system for an autonomous or semi-autonomous vehicle is described here. The distance detection system includes a distance detector, which may have a limited scope of distance detection, and a directional controller, which allows extending the dimension or scope of the distance detector as the vehicle travels and performs missions. The directional controller can change the detection direction of the distance detector with a motorized gimbal or functionally similar system, and the change in the detection direction can be integrated with the status of and other instructions executed by the vehicle.