Abstract: In an aspect, a processing apparatus receives detection signaling indicating a field incursion in a detection field of a LIDAR device mounted to a marine vessel and oriented to detect objects within a free space alongside the marine vessel. The processing apparatus applies plausibility testing to the field incursion, including determining whether parameters derived from the detection signaling are characteristic of a human body falling from the marine vessel through the detection field. The processing apparatus outputs signaling indicating a person overboard event to an alarm or control system onboard the marine vessel, in response to the field incursion passing the plausibility testing.

Abstract: In an aspect, a processing apparatus receives detection signaling indicating a field incursion in a detection field of a LIDAR device mounted to a marine vessel and oriented to detect objects within a free space alongside the marine vessel. The processing apparatus applies plausibility testing to the field incursion, including determining whether parameters derived from the detection signaling are characteristic of a human body falling from the marine vessel through the detection field. The processing apparatus outputs signaling indicating a person overboard event to an alarm or control system onboard the marine vessel, in response to the field incursion passing the plausibility testing.

Abstract: According to methods, apparatuses, and computer program products described by way of non-limiting examples presented herein, a computer system operates as an intelligent node in a distributed network. In one or more embodiments, the installation and execution of a node operating computer program product (“node software”) on a computer system configures that computer system to operate as an intelligent node in a distributed network. The computer system may reside at the edge of the network and thus may include a number of physical world sensors and/or actuators that represent valuable but hard-to-reach data sources/sinks from the perspective of the network at large. The node software operates the computer system as an edge application server and enterprise service bus, intelligently processing data from sensors, actuators, and other edge systems and devices. Other applications, such as server-based data processing and storing, are also easily implemented via the node software.

Abstract: A reconfigurable, hierarchical component-based architecture and framework for rapidly developing sensor device enabling software applications can have a rule-based data structuring system, a series table system, and a system of acquiring data from a data provider. User defined relationships can be established to enable exchange of data between the rule-based data structuring system, the series table system, the system of acquiring data, another rule-based data structuring system, an object oriented wrapping system, a data filtering system, a data structure grouping and merging system, a data device writer system, and a network communications system.

Abstract: A system for deploying component-based software systems for the implementation of desired sensor data process flows makes use of a toolset of software components specially tailored for building sensor data process flows. In one or more embodiments, the system comprises a front-end application operative to generate system specifications from user input, each system specification defining a desired component-based software system of one or more sensor components and their corresponding configurations in accordance with sensor component class definitions (e.g., the toolset) for a predefined set of sensor components tailored for building desired sensor data process flows.

Abstract: The teachings herein disclose methods and apparatus that provide software components (objects) as building blocks for rapidly developing sensor-enabled software applications. In at least one embodiment, a component-based software toolset provides data source components for obtaining sensor data, data series components for automatically collecting sensor data incoming from data source components, data field components for controlling that collection, and data destination components for automatically receiving data from the data series components. By providing methods and properties within these and other components in the toolset, the programmer builds sophisticated, hierarchical data flow processing with little or no manual programming.

Abstract: A system for deploying component-based software systems for the implementation of desired sensor data process flows makes use of a toolset of software components specially tailored for building sensor data process flows. In one or more embodiments, the system comprises a front-end application operative to generate system specifications from user input, each system specification defining a desired component-based software system of one or more sensor components and their corresponding configurations in accordance with sensor component class definitions (e.g., the toolset) for a predefined set of sensor components tailored for building desired sensor data process flows.

Abstract: A reconfigurable, hierarchical component-based architecture and framework for rapidly developing sensor device enabling software applications can have a rule-based data structuring system, a series table system, and a system of acquiring data from a data provider. User defined relationships can be established to enable exchange of data between the rule-based data structuring system, the series table system, the system of acquiring data, another rule-based data structuring system, an object oriented wrapping system, a data filtering system, a data structure grouping and merging system, a data device writer system, and a network communications system.

Abstract: Feed forward neural network models for associative content addressable memory utilize a first level matrix of resistor connections to store words and compare addressing cues with the stored words represented by connections of unit resistive value, and a winner-take-all circuit for producing a unary output signal corresponding to the word most closely matched in the first matrix. The unary output signal is converted to a binary output code, such as by a suitable matrix. Cues are coded for the address input as binary 1=+V, binary 0=-V, and unknown =0V. Two input amplifiers are employed with two input conductors for each input bit position, one noninverting and the other inverting, so that the winner-take-all circuit at the output of the first matrix may be organized to select the highest number of matches with stored words as the unary output signal.