78 results about "Software agent" patented technology

A distributed indexing and resolution architecture is disclosed for decentralized systems requiring modular, trust-scoped mutation control and dynamic alias governance. The system comprises a plurality of index entries arranged in a parent-child hierarchy, each associated with a structured alias and governed by one or more anchor nodes. Anchors perform localized resolution, mutation validation, and restructuring operations under deterministic policy constraints, enabling semantic scope enforcement and entropy-sensitive adaptation without requiring global consensus or centralized control. The architecture supports scoped alias traversal, asynchronous mutation proposals, and elastic anchor registration based on system state metrics. The indexing substrate may be integrated into heterogeneous infrastructures, including systems comprising distributed software agents, semantic execution platforms, or pseudonymous identity frameworks. Anchors coordinate within defined trust domains to ensure lineage continuity, dynamic rekeying, and semantic integrity across independently governed segments of a decentralized namespace.

A multi-agent and human project management and collaboration tool can provide the ability to create and manage projects and software agents in a cohesive framework. The collaboration tool can provide a multi-agent, multi-player channel where agents and humans work together to achieve an outcome. The collaboration tool can initiate a new project, where the objective is defined, and a team is dynamically built to achieve this objective. The team roles are parsed, and an agent is created for each role. Each agent is equipped with a personalized skillset and a working memory, allowing them to effectively collaborate with humans and each other to progress autonomously through the project stages. The agents can complete tasks based on the project plan and the results are stored in the agent's working memory. The project status is updated, and a report can be generated after each stage, providing a record of the project's progress.

A system is disclosed herein for attributing user affective responses to discrete elements within an experience. Software agents extract token instances from sensor data, while a response decomposition module allocates portions of measured affective response to tokens based on user attention. A token library stores token-response associations across users and domains, enabling prediction of responses in new contexts. An orchestration module coordinates agents and library updates, with privacy managers ensuring local processing of raw data. The system distinguishes tokens of interest from background elements, supporting cross-domain learning and overcoming limitations of static, siloed affective measurement systems.

A system and method for privilege assurance protection of computer networks that remedies the deficiencies of the current directory service structure. The system uses a software agent to collect and store snapshots of all network resources on a computer network by identifying network domains, searching the directory service of each domain for network resources, and periodically querying the network resources for changes. The software agent communicates with a backend server which provides searching, querying, storage, administrative and other functionality to the agent via remote procedure calls.

The governed interaction interface maintains an identity-preserving semantic-state for a knowledge entity and updates it only through validated semantic evidence generated autonomously by execution mechanisms. Semantic intentions describe semantic needs and are transformed into capability-requests containing no operational commands. Execution mechanisms—including environmental devices, perception models, symbolic analyzers, virtual actors, and software agents—interpret capability-requests independently through tool-side semantic interpreters and may act or decline to act. The system does not observe or evaluate tool behavior and receives only semantic evidence describing semantic meaning of any resulting effect. The governance engine evaluates each evidence fragment independently under identity, coherence, lineage, evidentiary sufficiency, deviation constraints, and contextual compatibility. Semantic-state transitions occur only when validated evidence satisfies governance constraints. The system performs no prediction, optimization, control computation, multimodal fusion, or supervisory coordination, remaining fully separated from device-level behavior.

A user of an identity management system may configure a software agent with a set of parameters within one or more user interfaces of an agent management service. The set of parameters may include an identifier of an application programming interface (API) endpoint of a first service that the software agent is authorized to query and one or more permissions associated with queries to the API endpoint by the software agent. Further, the user may receive an authentication token for the software agent from the agent management service. The software agent may use the authentication token for accessing resources of the service via the API endpoint in accordance with the one or more permissions. Thus, the user may configure the first software agent with the authentication token to enable the software agent to perform the queries to the API endpoint of the first service.

To assist users in artificial intelligence driven software development, techniques for agentic software development assistance leveraging application behavioral models are disclosed. A software agent receives a codebase issue and orchestrates an iteration of a refining process. To orchestrate the process, the software agent feeds the codebase issue and instructions to generate a plan into a development assistant, which automatically returns a plan to the agent. The agent feeds the plan and instructions to generate a solution to the codebase issue back to the development assistant, which automatically returns a solution to the agent. The agent feeds the solution back to the development assistant, which tests the solution in order to verify the solution's quality in responding to the codebase issue. The solution is awarded a quality score, and the quality score is returned to the agent. Based on the quality score, another iteration of the refining process can be performed.

The Universal Transportation Operating System (UniTOS) is an autonomous cyber-physical operating system that coordinates all movable Assets—machines, containers, vehicles, and software agents—across every Domain of human and machine activity. UniTOS integrates three core services: the Process Layout Asset Navigation Tool (PLANT) for PLAN authoring and systemic learning, the PLAN Execution Tool (PLANET) for Asset-level execution and variance reporting, and the Monitor-Manage-Maintain (M3) service for Domain-level coordination and variance reconciliation.

A vehicle artifact (e.g., software, firmware, content, etc.) distribution system uses direct vehicle-to-vehicle communications to extend a range of physical infrastructure and to improve speed and cost of distributing artifacts to vehicles. An artifact to be distributed is assigned a vehicle-to-vehicle distribution policy that defines gateway vehicle selection, wherein gateway vehicles are used to distribute the artifact to other recipient vehicles. A local software agent executing on the vehicles distributes the artifact in accordance with the vehicle-to-vehicle distribution policy using locally installed antennas of the vehicles.

This operation mode selection device selects an operation mode of a ship power system including one or a plurality of generators and batteries, the operation mode is determined through at least the number of generators in operation, the selection of a generator to be operated, a load sharing ratio of the generator and a battery, the selection of charging and discharging of the battery, and the device comprises a reinforcement learning unit which selects the operation mode; and performing reinforcement learning on a software agent.

The disclosed systems, methods and computer readable media relate to managing Non-Volatile Memory Express (NVMe) over Transmission Control Protocol (TCP) (NVMeOTCP) connections between a smart network interface card (smartNIC) and a block storage data plane (BSDP) of a cloud computing environment. A software agent (“agent”) executing at the smartNIC may manage a number of network paths (active and, in some cases, passive network paths). The agent may monitor the network traffic (e.g., input / output operations (IOPS)) through the paths (e.g., using established NVMeOTCP connections corresponding to the paths). If a condition is met relating to a performance threshold associated with the monitored paths, the agent may increase or decrease the number established NVMeOTCP connections to match real time network conditions.

A computing system may be used to support techniques to automate tasks using large language models (LLMs) and software agents. A user device may provide a data set to the computing system, and the computing system may process the data set to identify the one or more tasks using an LLM. The computing system may select one or more software agents to execute each of the identified tasks. For example, the computing system may identify a respective type for each task, and the computing system may select a respective software agent of a set of supported software agents configured to execute the respective type of task. Each software agent may execute a respective task to produce an output, such as by generating a summary of a transcript, transmitting one or more communications to users associated with the organization, or providing responses to inquiries, among other examples.

Systems and methods described herein provide a computer-implemented method for process simulation using a multi-agent system. A software agent is trained using historical process data associated with a process. Each software agent represents a single task in the process. Based on the candidate process model, the process is simulated by executing the multi-agent system to generate simulation results. The software agent operates autonomously within the multi-agent system based on the trained behavior. The process is also performed using a runtime configuration based on the candidate process model. Execution of the process generates runtime results for comparison with the simulation results. In some examples, the candidate process model is adjusted based on the comparison, and simulation and execution of the process are repeated using the adjusted multi-agent system and the adjusted runtime configuration, respectively. A deployment process model is generated based on the adjusted candidate process model.

A system and method for bi-directional quality testing of web real-time communications (WebRTC) sessions. In an embodiment, the system and method comprise an operator network, cloud contact center, cloud contact center agent application, and a synthetic software agent comprised of agent automation software, injected API shim code, virtual audio devices, audio processing applications, and media servers, capable of performing automated and to end communication testing. In order to provide end to end testing, especially with respect to voice quality, the synthetic agent software may control and monitor the audio channels (both send and receive) of the browser communication session.

The present disclosure is directed to systems and methods directed to improving the functions of a vehicle. Systems and methods are provided that provide a custom tool that autogenerates a set of software agents that allows a system to separate processing, transmission and receiving of messages to achieve better synchronization. The disclosure herein also provides a simplified method of key provisioning by designating one client as a server and assigning a symmetric key to every other client permanently provisioned between that client and the server. Systems and method are further provided that predict faults in a vehicle. Systems and methods are also provided that preserve data in the event of a system crash. Systems and methods are also provided in which an operating system of a vehicle detects the presence of a new peripheral and pulls the related interface file for that new peripheral. Further, a data synchronization solution is provided herein which provides optimized levels of synchronization.

Disclosed herein is an agent-first application architecture that facilitates intelligent, metadata-driven interactions across distributed computing environments. The system comprises a Metadata Repository, which defines communication policies, interaction rules, and contextual parameters. An Agent Orchestration Layer governs autonomous software agents, dynamically managing their lifecycle, task delegation, and execution logic. A Self-Learning Engine continuously refines agent decision-making models through real-time metadata updates and adaptive learning mechanisms. The system further incorporates an Event-Driven Middleware, which enables real-time, event-triggered execution of agent actions, ensuring adaptive, scalable, and autonomous decision-making across complex multi-system environments. By eliminating rigid API dependencies and leveraging metadata-centric intelligence, the architecture enhances cross-system interoperability, resilience, and adaptability. The invention is particularly suited for dynamic, multi-agent ecosystems requiring real-time adaptability, self-optimization, and intelligent orchestration of system interactions.

Systems and methods are provided for facilitating digital asset transfer via natural language commands using a conversation or chat interface associated with a digital wallet application. The digital wallet owner may exchange commands with an assistant user. The assistant user may comprise an automated software agent. The digital wallet owner may transfer digital assets to a recipient's digital wallet identified by command to the assistant user. The transfer command may comprise the recipient's name preceded with a “@” notation.

This invention discloses a training-free method for improving GUI localization capabilities based on an operation chain, relating to the fields of human-computer interaction and intelligent software agent technology. The method includes: acquiring a target interface image and natural language commands; obtaining initial candidate regions through a basic localization model; performing local cropping followed by secondary prediction and mapping back to the original image to update the region; applying a dynamic mask to the image and repeatedly predicting to generate spatially mutually exclusive candidate boxes; constructing a labeled comparison image; combining the input error correction model with commands and rule-based prompts; and selecting the optimal candidate box as the final localization result. This invention significantly reduces accuracy and ambiguity biases in localization by performing a series of interpretable inference time-series operations on the input screenshot and natural language commands, thereby improving the executable localization capabilities and stability of desktop intelligent agents on edge devices such as computers and mobile phones.

Embodiments of the invention are directed to systems, methods, and computer program products for provisioning a virtual lab environment for software development. In some embodiments, the method includes defining a set of features in the virtual lab environment; receiving, via a user input on a user interface, an unstructured dataset; generating a structured dataset using a generative artificial intelligence (AI) engine, where the unstructured dataset is an input of the generative AI engine; identifying, based on the structured dataset, a preferred configuration, where the preferred configuration comprises a target set of features; initiating a code delivery process, where the code delivery process includes configuring the set of features in the virtual lab environment to match the target set of features. The method may also include identifying at least one available software agent and displaying the at least one available software agent on the user interface.

A computer-implemented method of training a software agent for protecting a computer system includes receiving performance monitoring counter (PMC) data, including data from one or more hardware performance monitoring counters (PMCs) of the computer system, and receiving an output from an anomaly detector that is arranged to detect anomalies in the computer system. State data, including at least the PMC data, is provided to the software agent which selects a security policy from a set of security policies in dependence upon the state data, and applies the selected security policy to the computer system. A reward function is evaluated to determine a reward value which depends on at least the output of the anomaly detector. The reward value is used to train the software agent using reinforcement learning for selecting security policies from the set of security policies.

A processor circuit is configured to receive, while a current software agent is executing in a current execution mode, an indication of an execution mode transition that attempts to change an execution mode index from a current value indicative of the current execution mode to a new value indicative of a new execution mode. The processor circuit is configured to access, based on the current privilege level, particular execution mode transition information in a storage circuit to determine whether the execution mode transition is allowed. The processor circuit is configured to determine, based on particular execution mode transition information, whether to update the current value indicative of the current execution mode. The processor is further configured to update the current value indicative of the current execution mode based on a determination that the execution mode transition is permitted.

A method for the automated configuration of software systems involves processing a configuration request from multiple software agents, each of which has a large language model. Each agent contributes to fulfilling the request based on its role and expertise (domain expert, application expert, configuration expert). The result is a new or adapted configuration of the software system.

In some embodiments, the present disclosure provides an exemplary method that may include steps of identifying at least one computing specification image within a plurality of computing specification images; monitoring each data agent within the plurality of preinstalled data agents for a predetermined period of time to establish a data agent usage baseline associated with each data agent within the plurality of preinstalled data agents; utilizing a chaos engineering algorithm to dynamically perturb each data agent; calculating a usage test score for each data agent within the plurality of preinstalled data agents; calculating an overall data agent-specific usage score associated with each data agent within the plurality of preinstalled data agents based on the plurality of data agent-specific usage test scores; and rejecting at least one data agent within the plurality of preinstalled data agents from being utilized to launch the instance of the software application.

System and methods for computer modeling in medicine. A sort of period table of medical models is described for personalized diagnostics, prognostics and therapeutics, including at least 80 major categories of medical models. Generative artificial intelligence and geometric deep learning techniques, and algorithms including 2D and 3D graph machine learning and GenAI algorithms, are described, tailored and applied to diagnostic disease description, prognostic prediction and therapeutic development and management, including generation of novel synthetic drugs. The AI and machine learning techniques and algorithms are applied to understand each individual's genetic, RNA and protein anomalies that represent the source of many unique patient diseases. AI-enabled software agents assist physicians and researchers in building patient medical models. Several personalized medicine applications of individualized medical modeling include cardiovascular

Disclosed herein are embodiments for managing a virtual reality (VR) training exercise via a management server. The management server outputs a graphical dashboard including one or more skill nodes, and selects one or more software agents associated with the skill nodes. The management server provides the software agents to at least one host computing system communicatively coupled to a near-to-eye display device. The near-to-eye display device is configured to display a virtual three dimensional (3D) training environment including a plurality of interactive 3D virtual objects. The software agents are configured to collect VR observables data while the trainee performs actions within the virtual 3D training environment. Based on the VR observables data collected, the management server determines that one or more skills have been demonstrated during the training exercise, and updates the one or more skill nodes to graphically indicate the one or more skills demonstrated by the trainee.

A software agent is installed on each endpoint device within a plurality of endpoint devices. Each software agent manages a unique endpoint device. Each software agent is communicatively coupled to an orchestration software component. The software agents monitor security activities on each endpoint device and send security information details on each endpoint device to the orchestration software. The orchestration software generates security categories and a compliance score in each category for each endpoint device. The orchestration software creates a proactive security score (PSS) for each endpoint device, based on the security information compliance scores. The orchestration software generates a global proactive security score (GPSS) based on the average PSS for the plurality of endpoint devices. The orchestration software reports the PSS and GPSS for the plurality of endpoint devices.

Access a digital image frame containing an object of interest and receive object selection instructions for pixels associated with the object of interest. Assign a software agent to each pixel, whereby each software agent applies an iterative deep reinforcement learning process to its pixel to estimate the probability that the pixel is one of the pixels associated with the object of interest, and classifies whether the pixel is one of the pixels associated with the object of interest based on the estimated probability. Using multiple software agents and other software agents associated with regions of the object of interest to communicate or collaborate enables the software agents to compute more accurate estimated probabilities. For example, probability estimation errors can be more effectively avoided in cases where pixel values ​​are occluded (e.g., the boundary region between skin and hair).

Methods and control devices for providing functionality in a vehicle. The present invention relates to a method for providing functionality in a vehicle, particularly a software-defined vehicle (SDV), wherein at least two software modules, particularly two software agents, are executed on a control device, characterized in that the at least two software modules are distinct from each other, and each of the at least two software modules provides the functionality through a separate communication interface.

The control application generates a log file that provides information relating to the management of the plurality of devices being responsible. The information in the log file depends on a message received from the device or a command issued to the device. For example, in the case of a smart meter, the generated log file includes meter readings, firmware changes, firmware updates, indications as to whether the meter has been opened, indications of overheating at the meter. The software agent extracts information from the log file, which can be analyzed to determine any anomalies that may be associated with attacks on the system. The analysis performed may include pattern matching with information obtained from the log file, performing AI analysis on information obtained from the log file, identifying signatures, and the like.

Method for the automated coordination of a maintenance process for a motor vehicle, characterized by the following features: - Electronic reception of technical diagnostic data representing the condition of vehicle components via a data interface, - Semantic analysis of the received diagnostic data using a software agent based on a large language model, whereby the language model automatically derives a specific maintenance requirement and a list of spare parts required for maintenance as a machine-readable instruction from the technical diagnostic data, - Automated sending of electronic queries to external computer systems of workshops and spare parts suppliers using the derived instructions for determining workshop capacities and spare parts availability in real time, - Calculating an optimized maintenance appointment by algorithmically comparing the derived maintenance requirements with the determined workshop capacities and spare parts availability.