An example computing system is configured to: (i) receive, from one or more sensors of a vehicle in motion over a body of water, a set of sensor data, (ii) based on the set of sensor data, determine (a) an instantaneous distance between the vehicle and a surface of the body of water and (b) an instantaneous slope of the surface of the body of water, (iii) based on at least one of the instantaneous distance or the instantaneous slope, determine a statistical representation of the surface of the body of water, and (iv) based on the determined statistical representation of the surface of the body of water, adjust one or more control surfaces of the vehicle to change one or more of a speed, altitude, heading, or attitude of the vehicle.

A method for controlling an autonomous unmanned aerial vehicle for delivery of a medication package includes determining a thermal control period for the medication package. The method also includes identifying a delivery location corresponding to the medication package. The method also includes identifying at least one environmental characteristic of an environment that includes a delivery three-dimensional flight path between a starting location and the delivery location, wherein the at least one environmental characteristic indicates an actual weather value at the delivery location. The method also includes determining whether to deliver the medication package based on the thermal control period and the at least one environmental characteristic, using the unmanned aerial vehicle.

Disclosed are systems and methods to determine preferred delivery points within a parcel that are available to receive an aerial delivery of an item. For each delivery point a plurality of criteria scores may be determined for various criteria based on a processing of parcel data, image data, and/or sensor data corresponding to the parcel. The criteria may be aerial navigation related criteria or user preference criteria. The criteria scores may then be used to determine a suitability score for each delivery point. In some implementations, a user may specify a preferred delivery point and/or indicate one user criteria as more important than another criteria.

An unmanned aerial vehicle configured to be operated relative to a land vehicle. The unmanned aerial vehicle includes a processing circuitry configured to operate the unmanned aerial vehicle in a self-propelled mode when the land vehicle is stationary or moving with a speed below a threshold speed or operate the unmanned aerial vehicle in a towed mode, in which the unmanned aerial vehicle is towed by the land vehicle, when the land vehicle is moving with a speed above the threshold speed.

An autonomous vehicle that is equipped with image capture devices can use information gathered from the image capture devices to plan a future three-dimensional (3D) trajectory through a physical environment. To this end, a technique is described for image-space based motion planning. In an embodiment, a planned 3D trajectory is projected into an image-space of an image captured by the autonomous vehicle. The planned 3D trajectory is then optimized according to a cost function derived from information (e.g., depth estimates) in the captured image. The cost function associates higher cost values with identified regions of the captured image that are associated with areas of the physical environment into which travel is risky or otherwise undesirable. The autonomous vehicle is thereby encouraged to avoid these areas while satisfying other motion planning objectives.

The processor supplies flight commands to the flight control system by selectively blending pilot input with control signals from the autopilot. The processor generates a projected recovery trajectory through successive iterations, each beginning at the current aircraft location and using a recovery constraint selectable by the processor to influences a degree of flight aggressiveness. A detection system that identifies and invokes a state of threat existence if a threat exists along the projected recovery trajectory. The processor during threat existence in a first iteration commands an initial soft recovery, with permitted blended pilot input. If the threat exists on subsequent iteration, the processor commands a more aggressive recovery while attenuating blended pilot input.

A management system comprises a movement management unit that communicates via a communication device with a plurality of moving bodies including an autonomous moving body provided with an autonomous control unit for moving autonomously, and that manages the movement of the plurality of moving bodies. The movement management unit comprises a priority/subordination determination unit that determines the degree of priority/subordination relating to the respective movements of the plurality of moving bodies on the basis of individual information of the moving bodies. The autonomous moving body comprises: a priority/subordination comparison unit that compares another priority/subordination degree, which is a degree of priority/subordination determined by the priority/subordination determination unit for another moving body that is a moving body, from among the plurality of moving bodies, different from the autonomous moving body, and the host priority/subordination degree, which is a degree of priority/subordination determined by the priority/subordination determination unit for the autonomous moving body; or a priority/subordination reception unit that receives the comparison results of the host priority/subordination degree and the other priority/subordination degree obtained by comparisons by the moving body management unit.

A technique relates to autonomous obstacle avoidance. A vehicle is controlled on a route to a destination, the route being a three-dimensional route. An obstruction is determined on the global route. A local route including alternate points to avoid the obstruction is autonomously determined. The local route is autonomously converged back to the global route in response to avoiding the obstruction.

The present disclosure relates to a method for determining an action for collision avoidance in a craft. The method (100) comprises obtaining (110) object data comprising three-dimensional object data points (420); obtaining (120) state data of the craft (260); determining (140) at least one set of manoeuvre paths (410a,b,c) for the craft (260) based on the obtained craft state data; determining (150) a set of distance thresholds (421) for the three-dimensional object data points (420) based on the object data; comparing (160) each set of manoeuvre paths (410a,b,c) with the object data and the set of distance thresholds (421), wherein the set of manoeuvre paths (410a,b,c) is identified as a colliding set of manoeuvre paths (410a,b,c) when each path of the set of manoeuvre paths (410a,b,c) is at least partially within the corresponding distance threshold (421) of at least one three-dimensional object data point (420); and determining (170) an action upon identification of at least one colliding set of manoeuvre paths (410a,b,c).

A technique for controlling an unmanned aerial vehicle (UAV) includes monitoring a sensed airspeed of the UAV, obtaining a commanded speed for the UAV, wherein the commanded speed representing a command to fly the UAV at a given speed relative to an airmass or to Earth, and when the commanded speed is greater than the sensed airspeed, using the commanded speed in lieu of the sensed airspeed to inform flight control decisions of the UAV.