Abstract: Advanced cybersecurity systems and methods may provide enhanced security of critical infrastructure without the risks associated with traditional penetration testing. The systems and methods may model network vulnerabilities and simulate attack scenarios to identify potential security breaches. This supports the integration of generic rules and common properties, allowing for dynamic and scalable vulnerability analysis across complex network systems. Additionally, the systems and methods incorporate a multi-purpose fuzzer and an AI-driven Cyber Reasoning System (CRS) that autonomously detects and remedies software vulnerabilities based on predefined logic and pattern recognition. This comprehensive approach identifies existing vulnerabilities and predicts potential future threats by analyzing changes in network behavior and data flow.

Abstract: A system for autonomous cybersecurity probing includes a scanning module adapted to convert a target computing device or network scan to machine readable form. The scanning module includes an ingest module which processes the scan to create nodes representing the target ports, port status, and vulnerabilities. A command module includes a plurality of nodes representing facts, rules, actions, and verifiers associated with one or more vulnerabilities identified by the scanning module. The command module is configured to determine whether to launch an attack. An attack module is configured to, on receipt of instructions from the command module, assign an attack based on a one of the one or more vulnerabilities. A verifier module is configured to determine success or failure of the assigned attack and to return an indicator of the determined success or failure to the command module. Methods for cybersecurity probing using the described system are provided.

Abstract: Defensible AI systems and methods may provide technical solutions for technical problems facing typical AI systems. An expert system may be used to address problems facing AI that systems operate without providing visibility into their internal decision-making processes. The expert system may be developed with meaning-assigned fact nodes. A gradient descent style training process may be used to improve the performance of expert system networks. In an example, a gradient descent training process identifies the contributions of rules and fact values to the outcome fact values, then distributes a portion (e.g., velocity value determined portion) of the error to each rule input weighting based on its proportion of overall contribution. These expert systems may use various approaches to training, such as various selected inputs used to calculate the difference value (e.g., error value), various network designs, various error and augmentation levels, and various different training levels.

Abstract: The systems and methods described herein include attitude determination and control system (ADCS) and associated methods. Systems for determining attitude may be used by various vehicle types, such as to determine the vehicle's attitude relative to an external point of reference. The ADCS may be used for passive or active stabilization of spin on multiple axes. The ADCS uses an incorporated autonomous control algorithm to characterize the effects of actuation of the system components and simultaneously trains its response to attitude actuators. This characterization generates and updates a movement model, where the movement model is used to indicate or predict the effect of one or more attitude actuators given vehicle state information.

Abstract: The systems and methods described herein include an approach to performing quality assessment for 3-D printed objects during the printing process, for collecting data regarding 3-D printed objects, and for capturing data to make a digital model of an object. This approach uses sensor data (e.g., digital imagery) to characterize printing progress or to detect 3-D printing defects that would otherwise result in printing incomplete objects, such as premature printing job termination, dry printing, over/under application, movement of the filament, and other defects. Sensor data capturing can also be used as part of a destructive scanning process to perform post-printing object assessment or to collect data on a real-world object to facilitate creation of a digital model. These systems and methods may leverage the discrete nature of a pixel provided through digital imagery to be assessed with limited computational resources in a non-recursive manner.

Abstract: The systems and methods described herein include attitude determination and control system (ADCS) and associated methods. Systems for determining attitude may be used by various vehicle types, such as to determine the vehicle's attitude relative to an external point of reference. The ADCS may be used for passive or active stabilization of spin on multiple axes. The ADCS uses an incorporated autonomous control algorithm to characterize the effects of actuation of the system components and simultaneously trains its response to attitude actuators. This characterization generates and updates a movement model, where the movement model is used to indicate or predict the effect of one or more attitude actuators given vehicle state information.

Abstract: The systems and methods described herein include an approach to performing quality assessment for 3-D printed objects during the printing process, for collecting data regarding 3-D printed objects, and for capturing data to make a digital model of an object. This approach uses sensor data (e.g., digital imagery) to characterize printing progress or to detect 3-D printing defects that would otherwise result in printing incomplete objects, such as premature printing job termination, dry printing, over/under application, movement of the filament, and other defects. Sensor data capturing can also be used as part of a destructive scanning process to perform post-printing object assessment or to collect data on a real-world object to facilitate creation of a digital model. These systems and methods may leverage the discrete nature of a pixel provided through digital imagery to be assessed with limited computational resources in a non-recursive manner.