Abstract: Systems and methods are disclosed for tracking objects in a physical environment using visual sensors onboard an autonomous unmanned aerial vehicle (UAV). In certain embodiments, images of the physical environment captured by the onboard visual sensors are processed to extract semantic information about detected objects. Processing of the captured images may involve applying machine learning techniques such as a deep convolutional neural network to extract semantic cues regarding objects detected in the images. The object tracking can be utilized, for example, to facilitate autonomous navigation by the UAV or to generate and display augmentative information regarding tracked objects to users.

Abstract: Systems and methods are disclosed for tracking objects in a physical environment using visual sensors onboard an autonomous unmanned aerial vehicle (UAV). In certain embodiments, images of the physical environment captured by the onboard visual sensors are processed to extract semantic information about detected objects. Processing of the captured images may involve applying machine learning techniques such as a deep convolutional neural network to extract semantic cues regarding objects detected in the images. The object tracking can be utilized, for example, to facilitate autonomous navigation by the UAV or to generate and display augmentative information regarding tracked objects to users.

Abstract: Systems and methods are disclosed for tracking objects in a physical environment using visual sensors onboard an autonomous unmanned aerial vehicle (UAV). In certain embodiments, images of the physical environment captured by the onboard visual sensors are processed to extract semantic information about detected objects. Processing of the captured images may involve applying machine learning techniques such as a deep convolutional neural network to extract semantic cues regarding objects detected in the images. The object tracking can be utilized, for example, to facilitate autonomous navigation by the UAV or to generate and display augmentative information regarding tracked objects to users.

Abstract: Systems and methods are disclosed for tracking objects in a physical environment using visual sensors onboard an autonomous unmanned aerial vehicle (UAV). In certain embodiments, images of the physical environment captured by the onboard visual sensors are processed to extract semantic information about detected objects. Processing of the captured images may involve applying machine learning techniques such as a deep convolutional neural network to extract semantic cues regarding objects detected in the images. The object tracking can be utilized, for example, to facilitate autonomous navigation by the UAV or to generate and display augmentative information regarding tracked objects to users.

Abstract: Systems and methods are disclosed for tracking objects in a physical environment using visual sensors onboard an autonomous unmanned aerial vehicle (UAV). In certain embodiments, images of the physical environment captured by the onboard visual sensors are processed to extract semantic information about detected objects. Processing of the captured images may involve applying machine learning techniques such as a deep convolutional neural network to extract semantic cues regarding objects detected in the images. The object tracking can be utilized, for example, to facilitate autonomous navigation by the UAV or to generate and display augmentative information regarding tracked objects to users.

Abstract: Extrusion of feed materials of a high temperature formable non-ferrous metal susceptible to embrittlement during air processing uses an extrusion system having a rotatable wheel and shoe covering part of the length of a groove around the periphery of the wheel to form an arcuate passageway, the shoe having an abutment which substantially closes a second end of the passageway and an extrusion die spaced from the abutment by a die chamber. The process includes pre-heating the feed material to not less than about 390° C. in a chamber defined by a feeder device, maintaining a protective atmosphere in the chamber of the feeder device while the feed material is heated. The pre-heated feed material then is passed to an inlet end of the passageway, and drawn along the passageway, to be forced by the abutment into the die chamber and through an extrusion orifice of the die to provide extruded product.

Abstract: Titanium flat product is produced by passing a titanium powder green flat material through a pre-heating station and heated under a protective atmosphere to a temperature at least sufficient for hot rolling. The pre-heated flat material then is passed through a rolling station while still under a protective atmosphere and hot rolled to produce a hot rolled flat product of a required level of hot densification. The hot rolled flat product is passed through a cooling station while still under a protective atmosphere, and cooled to a temperature at which it can be passed out of a protective atmosphere. In the process, the hot rolling provides the predominant hot densification mechanism involved.

Abstract: Extrusion of feed materials of a high temperature formable non-ferrous metal susceptible to embrittlement during air processing uses an extrusion system having a rotatable wheel and shoe covering part of the length of a groove around the periphery of the wheel to form an arcuate passageway, the shoe having an abutment which substantially closes a second end of the passageway and an extrusion die spaced from the abutment by a die chamber. The process includes pre-heating the feed material to not less than about 390° C. in a chamber defined by a feeder device, maintaining a protective atmosphere in the chamber of the feeder device while the feed material is heated. The pre-heated feed material then is passed to an inlet end of the passageway, and drawn along the passageway, to be forced by the abutment into the die chamber and through an extrusion orifice of the die to provide extruded product.

Abstract: Titanium flat product is produced by passing a titanium powder green flat material through a pre-heating station and heated under a protective atmosphere to a temperature at least sufficient for hot rolling. The pre-heated flat material then is passed through a rolling station while still under a protective atmosphere and hot rolled to produce a hot rolled flat product of a required level of hot densification. The hot rolled flat product is passed through a cooling station while still under a protective atmosphere, and cooled to a temperature at which it can be passed out of a protective atmosphere. In the process, the hot rolling provides the predominant hot densification mechanism involved.