70 results about "Cyberphysical systems" patented technology

Computer-implemented methods and systems for training an adversarial neural network to simulate a stealthy blind false data injection attack on a cyber physical system are provided. Supervised learning is used to generate an initial attack vector, by an adversarial attack generation model, based on inferred grid topology and historical measurements. A final attack vector is generated, by an adversarial verification model, based on a filtered subset of the initial attack vector utilizing a substitute bad data detection threshold, wherein the final attack vector enables creation of a counter measure.

This disclosure provides a method for the standardized production, financing and delivery of housing using distributed ledger technology and a decentralized artificially intelligent cyber-physical system (CPS or invention) for compiling data and executing human and deep learning sequences for the purposes of standardizing the planning and execution of real estate development, regulatory compliance, financing, supply chain procurement, industrial manufacturing processes, logistical processes as well as project management tasks, and other various construction, manufacturing and delivery tasks and procedures to standardize development, production, and delivery of standardizes building modules and standardized building components in a standardized smart factory that are used to assemble structures.

A system to protect an asset includes a plurality of monitoring nodes each generating a data stream of current monitoring node values in time-domain. An anomalous space data source stores sets of anomalous feature vectors for each of the plurality of monitoring nodes generated by an anomaly detection model. A current system function processor, coupled to the plurality of monitoring nodes over a network, receives receive data streams from each of the plurality of monitoring nodes, and generates a set of current feature vectors. An anomaly detection computer coupled to the plurality of monitoring nodes receives the data streams from the plurality of monitoring nodes and the set of anomalous feature vectors, and outputs at least one decision boundary based on processing, using the anomaly detection model, of the current feature vectors relative to the sets of anomalous feature vectors.

Methods and systems for parsimoniously updating cyber-system models based on then-current conditions. A wide variety of cyber-physical system models are deployed for ensuring that a system operates within its constraints while nevertheless achieving an assigned mission. Updating models is resource intensive, and thus should be done only when a particular degree of accuracy of a model is needed. Modern cyber-physical systems comprise sensors, transducers, and actuators that interoperate to carry out specific cyber-physical system behaviors. A control invariant barrier function (CBF) implements safety constraints that relate environmental parameters to safe behaviors of the cyber-physical system. Responsive to detected changes in the environment, a model update activity corresponding detected changes in the environment is initiated, but only when the result of such a model update impacts safe and / or mission-critical operation of the cyber-physical system. Invocation of model updates that would not impact safe operation of the cyber-physical system are suppressed.

A cyber-physical system for conducting a telehealth session. In embodiments, the system includes a hardware control box that enables patients (e.g., elderly or cognitively-impaired patients) to easily participate in the telehealth session. In some embodiments, the system analyzes sensor data (e.g., thermal images, eye tracking data, etc.) and calculates state variables that form a mathematical representation (a “digital twin”) indicative of the physical, emotive, cognitive, or social state of the patient. In some of those embodiments, the system calculates Myasthenia Gravis core examination metrics by performing computer vision analysis on patient video data and / or audio analysis on patient audio data. In some embodiments, the system controls a pan-tile-zoom camera to zoom in on each region of interest that is relevant to the examination. In some embodiments, the digital twin is used as an input of a heuristic computer reasoning system, which uses artificial intelligence to support clinical diagnosis and decision-making.

Disclosed herein are devices, systems, and methods for the intelligent fuzzing of cyber-physical systems, including the use of artificial intelligence (AI) for such fuzzing. In at least one embodiment, a method is disclosed for intelligent fuzzing, which includes the steps of (1) characterizing “normal” behavior of the cyber-physical system being tested, which may include, for example, utilization of a causal model to analyze causal relationships between various messages sent to and / or from the system being tested, (2) identifying dependency relationships between these various messages and / or generating a dependency graph for the messages, and (3) fuzzing an incoming message from the system under test. The fuzzing process may include querying the causal model to determine if the message is contained within the model. Fuzzing may further include altering the message's contents by, for instance, randomly manipulating particular message fields and / or replacing letters with other letters.

This disclosure provides a method for the standardized production, financing and delivery of housing using distributed ledger technology and a decentralized artificially intelligent cyber-physical system (CPS or invention) for compiling data and executing human and deep learning sequences for the purposes of standardizing the planning and execution of real estate development, regulatory compliance, financing, supply chain procurement, industrial manufacturing processes, logistical processes as well as project management tasks, and other various construction, manufacturing and delivery tasks and procedures to standardize development, production, and delivery of standardizes building modules and standardized building components in a standardized smart factory that are used to assemble structures.

This document presents one or more advantageous approaches for Reinforcement Learning (RL) powered management of one or more transmission parameters, such as transmit power and diversity, for maximizing the application-layer reliability and availability of a Cyber-Physical System (CPS) with a minimized level of radio / power resource consumption. Example mathematical models are also disclosed and are useful for transforming high-level “intents” (e.g., KPIs that are applicable to industrial automation and control systems) into low-level orchestration objectives that drive the RL-based control. These objectives are subsequently employed in the definition of an RL-powered “orchestrator,” which may comprise an appropriately configured network node or other computing platform associated with the wireless communication network used to provide inter-device communications for a CPS comprising a population of devices. Further, the disclosure details example communication—e.g., observations and corresponding control signaling—between the orchestrator and the environment being managed.

A system for performing a three-dimensional computer vision task using a neural radiance field grid representation of a scene produced from two-dimensional images of at least a portion of the scene can include a processor and a memory. A neural radiance field grid network module can produce, from the two-dimensional images, three-dimensional patches of the neural radiance field grid representation of the scene. A three-dimensional shifted window visual transformer module can produce, from the three-dimensional patches, a feature map. A first decoder module can produce, from the feature map, the neural radiance field grid representation of the scene to train the system. A second decoder module can produce, from the feature map, the neural radiance field grid representation of the scene to perform the three-dimensional computer vision task for a cyber-physical system.

Methods for securing cyber-physical systems include attack generation systems, methods, and devices configured to assess vulnerabilities of cyber-physical systems include an example method of training an attack generative model using an attack policy to generate an attack dataset, training discriminators using a random attack dataset and generated attack dataset and training a generator using the trained discriminators.

The invention provides an elastic control method and system for denial of service attack of a network physical system, relates to the field of security control of the network physical system, and aims to solve the problem that an aperiodic and boundless dual-channel DOS attack problem cannot be solved. Sending of the triggering data is triggered, the data are transmitted under the condition that a transmission rule is met, the observer is switched to update state estimation of the observer, and an estimation value is sent to the controller; the controller sends a confirmation signal to the switching observation device, calculates a control signal according to the estimated value, and sends the control signal to the actuator; and the actuator feeds back the confirmation signal and executes the calculated control signal. According to the invention, based on an event triggering mechanism, through designing a safe transmission strategy, switching an observer and an elastic control method, state estimation and control of a network physical system under a dual-channel independent DoS attack are realized, the stability of the system is ensured, and communication resources are optimized.

A system and method for modelling a cyber-physical system to act as a honeypot for cyberattacks. The method including: building a virtual instance of the cyber-physical system including a physical layer and a cyber layer that controls the physical layer; generating a safety set defined by control barrier functions, the safety set delineates the bounds within which the cyber-physical system can operate safely; receiving a cyberattack payload from an attacking device; simulating physical dynamics of the physical layer and operation of the cyber layer; projecting whether the cyberattack payload can force the cyber-physical system to exit the safety set based on the simulated physical dynamics; and performing a safety action on the cyber-physical system when the physical system is projected to exit the safety set due to the cyberattack payload.

Disclosed herein are systems and methods for reducing or mitigation alert fatigue from real-time alerts in cyber-physical systems or other types of edge computing systems are provided. In one or more examples, the edge computing system monitor can look for one or more patterns within received data that can indicate malicious activity or other conditions that may warrant a real-time or near-real time response from the operator. In one or more examples, a detection of any of the specified patterns in the streaming data can trigger an alert to the operator of the edge computing system. In one or more examples, the alerts can be suppressed until the number of alerts associated with a particular pattern crosses a pre-determined threshold. Additionally or alternatively, alerts can be suppressed based on a duration that the alerts have been generated. The suppression of alerts can be configured to reduce operator alert fatigue.

A system and method for enhancing reliability and trustworthiness in cyber-physical systems using artificial intelligence are disclosed. The system comprises a plurality of sensors for collecting real-time data, a data processing unit that analyzes the data using machine learning models to identify patterns and anomalies, and a communication interface for transmitting recommendations based on the analysis. The machine learning models include an adversarial random forest (ARF) model and a probabilistic circuit model, and the analysis results are cross-validated across the collected data to verify accuracy and reliability. The method involves collecting data, analyzing it using the machine learning models, cross-validating the results, generating recommendations, and displaying them to an end-user. The invention provides a novel approach for enhancing the reliability and transparency of AI-driven cyber-physical systems.

Computer-implemented methods and systems for training an adversarial neural network to simulate a stealthy blind false data injection attack on a cyber physical system are provided. Supervised learning is used to generate an initial attack vector, by an adversarial attack generation model, based on inferred grid topology and historical measurements. A final attack vector is generated, by an adversarial verification model, based on a filtered subset of the initial attack vector utilizing a substitute bad data detection threshold, wherein the final attack vector enables creation of a counter measure.

An integrated cyber-physical system and method for high-fidelity distress monitoring and automated alert error mitigation. The system comprises a wearable multi-parametric sensing suite (100) for primary anomaly detection and an autonomous mobile robotic agent (200) for secondary contextual verification. Upon detection of a potential distress event (S110), the robotic agent (200) executes a priority undocking sequence (S120) from a charging station (500) to perform multi¬ modal acquisition (S130), including 3D skeletal pose estimation and acoustic sentiment analysis. A Cloud Processing Engine (300) synthesizes these data streams — including optional spatial occupancy data from an environmental sensor network (400) — to calculate a Validation Confidence Score (VCS) (S140) using a dynamic weighted fusion algorithm. This score enables the system to autonomously perform alert de-escalation (S150) for verified safe-states or execute prioritized escalation protocols (S160) to a caregiver interface (600) and an emergency services gateway (700). The invention effectively reduces alarm fatigue and improves emergency response reliability by providing contextual validation of biometric triggers.

The present invention relates to a cyber-physical system for optimizing a simulation model for chemical mechanical polishing based on actual measurement data of chemical mechanical polishing. The chemical mechanical polishing system includes a polishing apparatus (1) for polishing the workpiece (W) and an arithmetic system (47). The arithmetic system (47) includes a simulation model including at least a physical model configured to output an estimated polishing physical quantity including an estimated polishing rate of the workpiece (W). The arithmetic system (47) is configured to: input polishing conditions for the workpiece (W) into the simulation model; output the estimated polishing physical quantity of the workpiece (W) from the simulation model; and determine model parameters of the simulation model that bring the estimated polishing physical quantity closer to a measured polishing physical quantity of the workpiece (W).

We provide security measures that integrate and analyze information from cyberspace and physical space using generative AI. [Solution] A security support device 10 is provided, comprising: a cyber information collection means 150 for collecting anomaly information in cyberspace; a physical information collection means 160 for collecting anomaly information in physical space; a cyber-physical correspondence means 170 for associating anomaly information in cyberspace with anomaly information in physical space; a generation AI input / output means 110 for inputting the associated anomaly information in cyberspace and anomaly information in physical space as prompts to a generation AI 20 and obtaining security measures as a response from the generation AI 20; and a CPS integrated security measure determination means 120 for assigning priority and reliability to the security measures, which are the response from the generation AI 20, and transmitting them to a CPS (cyber-physical system) 30.

A method for determination of anomalies in a cyber-physical system (CPS) includes generating one or more diagnostic rules configured to calculate at least one auxiliary CPS variable. One or more values of the at least one auxiliary CPS variable are calculated for a predefined output interval of time based on collected values of a group of primary CPS variables for a predefined input interval of time based on the generated diagnostic rule. An anomaly is determined based on the collected values of the group of primary CPS variables and the one or more calculated values of the at least one auxiliary CPS variable.

A system for determining light conditions in a space of interest and to control a device thereon includes a light sensor arranged to detect light entering a sub-volume of the space of interest to provide measurement data corresponding to at least an intensity and direction distribution of light entering the sub-volume: a data processor configured to communicate with the light sensor to receive the measurement data, the data processor configured with a computational model to provide a calculated distribution of light based on the measurement date from the sub-volume of the space of interest; and a control system configured to communicate with the data processor to receive the calculated distribution of light and provide control signals based on the calculated distribution of light.

Accurate estrus detection of sows is critical to achieving a high farrowing rate and maintaining good reproductive performance. However, the conventional method of estrus detection uses a back pressure test by farmers, which is time-consuming and labor-intensive with a significant degree of error. This disclosure is of an automated estrus detection method by monitoring the change in vulva swelling around the estrus using a three-dimensional measurement device, e.g., LiDAR camera, which includes an RGB camera and a depth camera. This sow estrus detection improves accuracy and efficiency, reduces labor and cost, and improves the sustainability of swine production using a data-driven decision-making system based on a robotic cyber-physical system (CPS) that can utilize deep learning detection based on a deep learning model.

The present invention relates to a method for error detection and handling in a DevOps-pipeline for a cyber physical system and related system. The method comprising the steps of: writing and committing code destined to be used for monitoring and / or controlling operations of components of the CPS; compiling the code and packaging it into deployable units; deploying the code per deployable unit into one or more component of the CPS; connecting one or more domain-specific language, DSL, elements to respective one or more CPS components; gathering data from the respective one or more CPS components by the one or more DSL elements; processing by the one or more DSL elements the data gathered in a format of the respective one or more CPS component; comparing by the one or more DSL element the processed data to a predefined failure condition to detect a failure.

One embodiment provides a system controller circuitry for mitigating a neurophysiological disorder. The system controller circuitry includes an optimization module, a feedback control module, and a model update module. The optimization module is configured to predict a sequence of control inputs based, at least in part, on a fractional order model of a neurophysiological system. A duration of the predicted sequence of control inputs corresponds to a prediction horizon. The feedback control module is configured to provide at least a portion of the sequence of control inputs to the neurophysiological system based, at least in part, on a current state of the neurophysiological system. A duration of the at least a portion corresponds to a control horizon. The model update module is configured to update one or more of a model parameter and / or an objective function parameter at an update interval, based, at least in part, on a recent state of the neurophysiological system.

A server can include an interface configured to receive sensor data relating to a cyber-physical system (CPS), as well as a processor. The processor can be configured to determine a performance parameter of the CPS from the sensor data; select a resource allocation strategy based on the performance parameter; execute an algorithmic process on at least a portion of the sensor data according to the resource allocation strategy; and control a transmitter to send data, representing an output of the executed artificial neural network, to the robot.

The application provides a resilient control method and system for denial of service attacks on networked physical systems, and relates to the field of security control of networked physical systems, and aims to solve the problem of non-periodic and boundaryless double-channel DOS attack. An event trigger generates an event when measurement data meets a specific trigger condition, transmits the trigger data, transmits the data under the condition of meeting the transmission rule, switches the observer to update the state estimation, and sends the estimation value to the controller; the controller sends an acknowledgement signal to the switching observer, calculates the control signal according to the estimation value, and sends the control signal to the actuator; the actuator feeds back the acknowledgement signal and executes the calculated control signal. Based on the event trigger mechanism, by designing a safe transmission strategy, a switching observer and a resilient control method, the state estimation and control of the networked physical system when suffering from double-channel independent DoS attack are realized, and the system stability is ensured and the communication resources are optimized.

The “Industrial Digestive System” encapsulates a novel paradigm in waste management and resource recovery, bridging the gap between sophisticated computational intelligence and industrial automation. The hardware component replicates the natural digestive system's efficiency in processing a diverse range of inputs, including municipal and industrial waste, transforming them into valuable outputs like biofuels, chemicals, and advanced composite / conglomerate materials. This process is achieved through a series of mechanized operations. Complementing the hardware, the software facet of the invention is rooted in advanced machine learning and optimization algorithms that optimize multivariable functions for the most efficient conversion of input to resources, orchestrating the transformation from varied waste streams, elements, and molecules into solid, liquid, and gaseous outputs. The result is an automated, intelligent manufacturing machine that enhances product quality and operational efficacy. “The Industrial Digestive System” can be seen as a new branch of cybernetics, a cyber-physical system inspired by nature.

The invention discloses an information and environment security risk identification method based on multi-modal data analysis, and aims to solve the problem that the'coupling risk 'of information security and environment security in a network-physical system is difficult to jointly identify. The method comprises the following steps: acquiring and aligning network side data and environment side data with timestamps and asset identifiers; outputting a network anomaly probability, an affected control object and a control instruction parameter by using a Transform network side security detection model; constructing two groups of boundary conditions of an abnormal control instruction and a baseline control instruction based on the affected control object, and performing conditional deduction in the physical information neural network model to obtain two environmental state prediction sequences; calculating an anti-fact deduction residual score according to the residual difference between the observation data and the two prediction sequences; and inputting the network side evidence and the environment side evidence into the micro evidence fusion model to obtain a coupling risk score and grade, and outputting a corresponding influenced control object, thereby realizing the technical effects of interpretable joint identification and grading early warning of the information security and environment security coupling risk.

An adaptive control method for cyber-physical systems (CPS) subjected to injection and deception attacks includes the following steps: S1: Constructing a control system model under a cyber-physical framework; S2: Defining actuator attacks when the system's sensors and actuators are subjected to adversary injection and deception attacks; S3: Defining sensor attacks when the system is subjected to sensor attacks; S4: Rewriting the system model under cyberattacks; S5: Based on the rewritten system model, obtaining the adaptive control law u using the BackStepping method; S6: Further deriving that all signals in the closed-loop system are globally bounded based on the adaptive control law u in S5, and designing a controller by introducing Nussbaum even functions and their variable derivatives. When the system is subjected to injection and deception attacks, the control parameters are adjusted to make the adjustment error arbitrarily small. This invention can ensure that the adjustment error can be arbitrarily small under injection and deception attacks by adjusting the control parameters.

The invention provides an unsupervised multivariate time series data anomaly detection method based on a generative adversarial network. The implementation process of the method comprises the steps of collecting and preprocessing data, constructing and training a proposed method model, calculating an anomaly score by using a pre-trained model to carry out anomaly detection, and evaluating the detection performance of the model so as to carry out optimization. The proposed method uses a long short-term memory recurrent neural network (LSTM-RNN) as a base model (i.e., a generator and a discriminator) to capture temporal correlation of time sequence distribution in a GAN framework. In addition, the multivariate time series data anomaly detection method provided by the invention does not independently process each data stream, but considers the whole variable set at the same time to capture potential interaction between variables. Meanwhile, a generator and a discriminator of GAN training are fully utilized, and a novel anomaly score is used to detect anomalies by combining the discriminator and a reconstruction error. By implementing the method provided by the invention, the abnormal behavior in the network physical system can be effectively detected, and timely alarm and response are provided for a system operator, so that the safety and reliability of the system are ensured.

A server may include an interface that is configured to receive sensor data related to a cyber-physical system (CPS); and a processor. The processor may be configured to determine a performance parameter of the CPS from the sensor data; select a resource allocation strategy based on the performance parameter; execute an algorithmic process on at least a portion of the sensor data according to the resource allocation strategy; and control a transmitter to send data representing an output of the executed artificial neural network to the robot.