Abstract: A method of generating a tessellated output file comprising receiving a computer-aided design (CAD) model file defining a CAD model including a plurality of vertices, a plurality of curves, a plurality of surfaces, and at least one volume; generating base polygon data defining the CAD model for additive manufacturing, the base polygon data including a plurality of connected polygons, each polygon including a plurality of nodes, a plurality of edges, and a face; generating data for a vertex metadata container including a listing of CAD model vertices and one polygon node associated with each CAD model vertex; generating data for a curve metadata container including a listing of CAD model curves and at least one polygon edge associated with each CAD model curve; and generating data for a surface metadata container including a listing of CAD model surfaces and at least one polygon face associated with each CAD model surface.

Abstract: A neural inference chip is provided, including at least one neural inference core. The at least one neural inference core is adapted to apply a plurality of synaptic weights to a plurality of input activations to produce a plurality of intermediate outputs. The at least one neural inference core comprises a plurality of activation units configured to receive the plurality of intermediate outputs and produce a plurality of activations. Each of the plurality of activation units is configured to apply a configurable activation function to its input. The configurable activation function has at least a re-ranging term and a scaling term, the re-ranging term determining the range of the activations and the scaling term determining the scale of the activations. Each of the plurality of activations units is configured to obtain the re-ranging term and the scaling term from one or more look up tables.

Abstract: A processor-implemented method includes receiving, at a processor, from a transceiver and without human intervention, location data associated with a target. The processor receives, from the transceiver, product data for a resupply request referencing the target, the product data including at least one of a product type or a product quantity. The processor generates, without human intervention, unmanned autonomous vehicle (UAV) mission data based on the location data and the product data, the UAV mission data including a representation of at least one UAV and flight path data for the at least one UAV. The UAV mission data is caused to transmit to at least one UAV controller to cause the at least one UAV controller to initiate navigation of the at least one UAV according to the UAV mission data.

Abstract: Modular neural network computing apparatus are provided with distributed neural network storage. In various embodiments, a neural inference processor comprises a plurality of neural inference cores, at least one model network interconnecting the plurality of neural inference cores, and at least one activation network interconnecting the plurality of neural inference cores. Each of the plurality of neural inference cores comprises memory adapted to store input activations, output activations, and a neural network model. The neural network model comprises synaptic weights, neuron parameters, and neural network instructions. The at least one model network is configured to distribute the neural network model among the plurality of neural inference cores. Each of the plurality of neural inference cores is configured to apply the synaptic weights to input activations from its memory to produce a plurality of output activations to its memory.

Abstract: Systems are provided that can produce symbolic and numeric representations of the neural network outputs, such that these outputs can be used to validate correctness of the implementation of the neural network. In various embodiments, a description of an artificial neural network containing no data-dependent branching is read. Based on the description of the artificial neural network, a symbolic representation is constructed of an output of the artificial neural network, the symbolic representation comprising at least one variable. The symbolic representation is compared to a ground truth symbolic representation, thereby validating the neural network system.

Abstract: Neural network accelerator output ranking is provided. In various embodiments, a system comprises a data memory; a memory controller configured to access the data memory; a plurality of comparators configured in a tree; a register; and a two-way comparator. The memory controller is configured to provide a first plurality of values from the data memory to the comparator tree. The comparator tree is configured to perform a plurality of concurrent pairwise comparisons of the first plurality of values to arrive at a first greatest value of the first plurality of values. The two-way comparator is configured to output the greater of the greatest value from the comparator tree and a stored value from the register. The register is configured to store the output of the two-way comparator.

Abstract: Conflict-free, stall-free, broadcast networks on neural inference chips are provided. In various embodiments, a neural inference chip comprises a plurality of network nodes and a network on chip interconnecting the plurality of network nodes. The network comprises at least one pair of directional paths. The paths of each pair have opposite directions and a common end. The network is configured to accept data at any of the plurality of nodes. The network is configured to propagate data along a first of the pair of directional paths from a source node to the common end of the pair of directional paths and along a second of the pair of directional paths from the common end of the pair of directional paths to one or more destination node.

Abstract: Unmanned autonomous vehicle (UAV) selection information identifying a UAV, flight path information, and a relative elevation difference value that is a difference between a first elevation value of a start location and a second elevation value of a drop location are received. Navigation of the UAV is initiated using the UAV selection information and the flight path information. A relative altitude difference value is obtained using a first altitude value of the UAV associated with the start location and a second altitude value of the UAV associated with the drop location. In response to a difference between the relative elevation difference value and the relative altitude difference value being outside a predefined threshold, UAV is caused to adjust the second altitude value such that an updated difference between the relative elevation difference value and an updated relative altitude difference value is within the predefined threshold.

Abstract: A battery boosting device adapted to be affixed internally to a vehicle, the battery boosting device comprising two or more sets of cables adapted to jump start an internal batter and/or an external battery.

Abstract: A system and method for generating a computer-based united cellular lattice structure includes dividing a part volume into a number of adjacent subvolumes each having a side combination corresponding to a number and orientation of adjoining sides and non-adjoining sides. A modified lattice cell may be generated from a base lattice cell for each side combination of the subvolumes such that each modified lattice cell has face surfaces on faces thereof corresponding to non-adjoining sides and does not have face surfaces on faces thereof corresponding to adjoining sides. Copies of the modified lattice cells may then be generated and inserted into corresponding subvolumes such that the faces of the modified lattice cell copies having face surfaces are positioned along non-adjoining sides of the subvolumes and faces of the modified lattice cell copies not having face surfaces are positioned along adjoining sides of the subvolumes.

Abstract: A battery boosting device adapted to be affixed internally to a vehicle, the battery boosting device comprising two or more sets of cables adapted to jump start an internal batter and/or an external battery.

Abstract: A neural network processor system is provided comprising at least one neural network processing core, an activation memory, an instruction memory, and at least one control register, the neural network processing core adapted to implement neural network computation, control and communication primitives. A memory map is included which comprises regions corresponding to each of the activation memory, instruction memory, and at least one control register. Additionally, an interface operatively connected to the neural network processor system is included, with the interface being adapted to communicate with a host and to expose the memory map.

Abstract: Described is a low-profile mouth guard formed of layered materials with variable hardness and different molding temperatures joined together to form a composite structure. The mouth guard is formed from a unique blend of materials including PCL (Polycaprolactate), EVA (Polyethylene Polyvinylacetate) and TPU (Thermoplastic Urethane). The unique blend allows a low-profile design that delivers both comfort and protection.

Abstract: A system and method of creating a shape-conforming lattice structure for a part formed via additive manufacturing. The method includes receiving a computer model of the part and generating a finite element mesh. A lattice structure including a number of lattice cellular components may also be generated. Some of the mesh elements of the finite element mesh may be deformed so that the finite element mesh conforms to the overall shape of the part. The lattice structure may then be deformed so that the lattice structure has a cellular periodicity corresponding to the finite elements of the finite element mesh. In this way, the part retains the benefits of its overall shape and the benefits of lattice features without introducing structural weak points, directional stresses, and other structural deficiencies.

Abstract: Described is a multi-purpose instrument and music stand. The instrument and music stand includes a base assembly having a plurality of legs. A rod projects up from the base assembly. At least one attachment tang with a pair of side recesses is pivotally attached with the rod. A tang receptor is included that selectively attaches with the attachment tang. The tang receptor is formed with a button and geared detent parts. Depressing the button causes the pair of geared detent parts to simultaneously retract from the pair of side recesses of the attachment tang, thereby allowing a user to selectively attach and detach a music holder or other item from the tang and stand.

Abstract: The present invention relates to a system for bypassing stomach and creating a barrier for the proximal small bowel to include duodenum and jejunum so that food delivered is directly in the jejunum that results in weight loss. The technique involves placement and fixation of a gastric bypass device within the stomach to cover the small bowel. The present invention also provides a device and method which enables the suturing of the gastric bypass device. The duodeno-jejunal barrier remains unchanged and helps in bypassing the absorptive surface of the proximal small bowel.

Abstract: A system and method of creating a shape-conforming lattice structure for a part formed via additive manufacturing. The method includes receiving a computer model of the part and generating a finite element mesh. A lattice structure including a number of lattice cellular components may also be generated. Some of the mesh elements of the finite element mesh may be deformed so that the finite element mesh conforms to the overall shape of the part. The lattice structure may then be deformed so that the lattice structure has a cellular periodicity corresponding to the finite elements of the finite element mesh. In this way, the part retains the benefits of its overall shape and the benefits of lattice features without introducing structural weak points, directional stresses, and other structural deficiencies.

Abstract: A system and method of creating a lattice structure for a part formed via additive manufacturing. The method includes generating a finite element mesh including a first finite element, generating a first natural space data structure such that the first finite element includes the first natural space data structure, generating a first unit cell corresponding to the first finite element, and generating a first design space data structure such that the first unit cell includes the first design space data structure and the first natural space data structure. The method further includes modifying the first unit cell via the first design space data structure and modifying the first unit cell via the first natural space data structure to correspond to the modifications of the first unit cell in the first design space data structure.

Abstract: A method for coordinating multiplexed communication of data among multiple piloted assets and multiple requestors via a common software application includes receiving a mission request via the common software application, defining a common mission based on the mission request via the common software application, and coordinating the multiplexed communication using multiple communication channels. The coordinating includes exchanging data with multiple piloted aerial drone assets relaying a communication from one of the piloted assets to at least one other piloted asset, via the software application. When a new piloted asset joins the common mission, communication is facilitated via the software application between the further piloted asset and the multiple piloted assets. Upon detection that a piloted asset from the multiple piloted assets is no longer associated with the common mission, communication via the software application between that piloted asset and the remaining piloted assets is prevented.

Abstract: A system for locating, segmenting, annotating and retrieving multimedia files, provides a database of metadata relating to multimedia files, a database manager, and a database client for accessing the data contained within the database. The database client, together with the metadata database and database manager provide a variety of different functionalities, namely a deep linking functionality, a segmentation functionality, a metadata annotation functionality, a retrieval functionality, and an access functionality. The user, through database client annotates the multimedia file or segment of multimedia file with metadata which is saved in the database. When the user desires to locate a multimedia file, the metadata is searched or browsed to locate the database entry associated with the multimedia file in question.