26 results about "Autonomous agent" patented technology

Systems and methods disclosed herein perform privacy-preserving evaluations of artificial intelligence (AI) agents. A first AI agent associated with a first entity obtains a machine-readable data structure defining one or more operative boundaries for a second AI agent associated with a second entity. The system generates a unique fixed reference value representing the machine-readable data structure by applying a first transformation operation set, and transmits the unique fixed reference value to a multi-agent storage to store the value. The system receives, via the multi-agent storage, a verification artifact from the second AI agent that indicates an observed value based on internal operational data of the second AI agent corresponding to the operative boundaries. The first AI agent determines a verification status of the verification artifact by comparing the unique fixed reference value with the observed value, and autonomously generates a verification record including a representation of the verification status.

Systems and methods disclosed herein perform privacy-preserving evaluations of artificial intelligence (AI) agents. The system identifies an auditing AI agent from a set of auditing AI agents for assessing target AI agent sets. The system obtains a data structure that defines operative boundaries for a target AI agent set and generates a reference value by applying a first transformation operation set on the data structure. The system transmits the reference value to a multi-agent storage and receives, via the multi-agent storage, a verification artifact from the target AI agent set that indicates an observed value generated by applying a second transformation operation set on an artifact set generated by the target AI agent set. The system determines, via the auditing AI agent, a verification status and responsive to a particular artifact failing to satisfy one or more assessment metrics, generates an action set to modify the target AI agent set.

This invention belongs to the field of artificial intelligence communication technology and discloses an AI autonomous intelligent agent execution framework architecture system based on layered decoupling. The invention integrates the core capabilities of ReAct execution loop, intelligent context compression, and autonomous error correction in the decision layer to achieve accurate parsing of instruction intent, dynamic iteration of decision logic, and autonomous error correction. The transmission layer establishes a multi-platform message intelligent routing and protocol unification mechanism to achieve efficient data interaction and instruction transmission across platforms and protocols. The execution layer integrates hybrid memory management and sandbox-isolated practical execution capabilities to achieve multi-dimensional memory of execution data, secure isolation of execution behavior, and accurate implementation of practical tasks. The system's layers are clearly defined, decoupled, independent, and collaborative, possessing high scalability, cross-platform adaptability, execution security, and decision autonomy.

Systems, methods, and devices for monitoring and controlling communications between autonomous agents are disclosed. The system monitors real-time communications between autonomous agents, intercepting and recording each communication. Communications are translated to a standardized language and processed through communication protocol filters that evaluate compliance with predefined operational policies. The system parses each translated communication to identify policy violations. When violations are detected, the system modifies communications to ensure compliance with operational policies. All communications and modifications are recorded via distributed ledger technology for audit and accountability purposes. This approach enables comprehensive oversight of autonomous agent interactions while maintaining tamper-proof records of all monitoring and control activities.

A method for determining a human activity in a real environment including at least one usage sensor for at least one device of the environment. The method includes learning by an artificial intelligence and an inference by the artificial intelligence to determine the human activity in the environment. For the learning by the artificial intelligence, the method includes generating artificial intelligence training data, based on a simulation of an activity of at least one autonomous agent operating within a simulation of the environment generated from a digital twin of the real environment and using at least one digital replica of the device.

Imitation learning, or artificial intelligence-based learning from demonstration, aims to acquire an agent policy by observing and mimicking the behavior demonstrated in expert demonstrations. Imitation learning can be used to generate reliable and robust learned policies in a variety of tasks involving sequential decision-making, such as autonomous driving and robotics tasks. However, existing methods that use next-token-prediction (NTP) models, where the policy reduces to a classifier over a discrete set of trajectory tokens, suffer from covariate shift due to their open-loop training a closed-loop execution. The present disclosure provides closed-loop fine tuning of autonomous agent policies in a manner that can mitigate covariate shift.

Methods, systems, and devices are described for orchestrating server management in a modern IT network. The described techniques may be implemented to manage any number of networked severs, whether local, remote, or both. Server orchestration may leverage a central, cloud-based management system and / or one or more autonomous agents installed on servers with the network. The autonomous agents may each be registered with the supervisory server and may have awareness of one another.

Disclosed herein are system, method, and computer program product embodiments for increasing computer, network, and data security via user generated tokens including zero-knowledge proofs within an agentic architecture. A client device may utilize an autonomous agent to communicate with an autonomous agent at a merchant device to identify whether the merchant device is selling a desired product. A client device may generate a token including the product, a merchant identifier, purchase amount, timestamp, and zero-knowledge proof. The client device may sign the token using an encryption key. The agent at the client device may transmit the signed token to the agent at the merchant device identified via the merchant identifier. The client device may receive a result indicating whether the token and zero-knowledge proof were verified by the merchant device.

Scene reconstruction is an image processing technique that creates a model of a scene from a given input, typically involving the creation of a three-dimensional (3D) scene model from one or more input two-dimensional (2D) images of the scene. High-quality scene reconstruction and rendering are useful for various applications, such as autonomous agent applications and scene editing applications. Existing scene reconstruction methods reach their limits with dynamic scenes where moving objects are inconsistent across different views at different times. These methods generally lack the motion cues essential for effective object-environment decomposition, which, moreover, leads to incomplete environment reconstruction of new views due to persistent occlusion of environmental structures.The present disclosure integrates spatial memories or a spatial memory from previous iterations of a scene when learning a representation of a temporally changing scene, thereby providing observations of obscured areas and contextual information for road users to enable more comprehensive and efficient learning of the scene representation.

A computing services environment may include application servers providing computing services including access to a database system, a unified metadata framework including autonomous agent definitions referencing action definitions defining a plurality of actions capable of being performed within the computing services environment, an agent service configured to instantiate an autonomous agent instance based on an autonomous agent definition, and an orchestration layer configured to determine an orchestration plan based on novel planning text generated by a generative language model. The orchestration plan may include a subset of the plurality of actions identified in the novel planning text. The computing services environment may execute the subset of the plurality of actions within the computing services environment.

A method for generating a motion plan for an autonomous agent involves capturing scene data for the agent's environment, including map data and past state data for one or more objects. A transformer-based encoder generates a set of scene embedding tokens from the scene data to represent a fixed environmental context. A transformer-based decoder autoregressively generates a sequence of action tokens representing a future motion trajectory. The generation of each subsequent action token is based on the scene embedding tokens and previously generated tokens. Each action token is selected from a discrete action space of unique Verlet actions representing accelerations. A future motion trajectory is determined from the sequence of action tokens and supplied to a motion planning module of the autonomous agent.

The disclosed concepts relate to employing agent behavior models to control agent behavior in applications, such as video games or simulations. For example, in some implementations, agent behavior models with relatively greater resource utilization, such as generative language models, are assigned to agents at higher levels of an agent hierarchy. Agent behavior models with relatively less resource utilization, such as reinforcement learning or hard-coded models, are assigned to agents at lower levels of the agent hierarchy.

Scene reconstruction is a computer vision process that creates a model of a scene from a given input, usually including creating a three-dimensional (3D) scene model from one or more input two-dimensional (2D) images of the scene. High-quality scene reconstruction and rendering is useful for various applications, such as autonomous agent applications and scene editing applications. Existing scene reconstruction methods encounter limitations in dynamic scenes where moving objects are not consistent across views from different times, and these methods generally lack the motion cues essential for effective object-environment decomposition and further lead to incomplete environment reconstruction of novel views from persistent occlusion of environmental structures. The present disclosure integrates spatial memory from prior traversals of a scene when learning a representation of a temporally changing scene, which can provide observations of occluded areas and contextual information for traffic participants for more comprehensive and efficient scene representation learning.

The method for dynamically updating an environmental representation of an autonomous agent can include: receiving a set of inputs S210; generating an environmental representation S220; and updating the environmental representation S230. Additionally or alternatively, the method S200 can include providing the environmental representation to a planning module S240 and / or any other suitable processes. The method S200 functions to generate and / or dynamically update an environmental representation to facilitate control of an autonomous agent.

This disclosure relates to a multi-agent computer control system that employs an iterative optimization process to optimize coordination among autonomous agents. The system initializes system states and relational constraints, receives state intentions from multiple agents, and computes a set of permissible state combinations. The optimization process transforms the system states into Base_Equilibrium_Factor_Northogonal (BEFN) states, where equilibrium geometry, factor geometry, and normal orthogonal geometry define structured search directions. By searching along the equilibrium and normal orthogonal geometries and updating the base points, the system iteratively refines the states to determine the optimal states. Unlike traditional gradient descent methods that rely on gradient-based minimization, this system employs multi-directional search mechanisms that improve constraint-aware optimization, avoid local minima, and enhance agent coordination. The optimal states are used to dynamically control the agents, thereby ensuring efficient, scalable, and adaptive decision-making in real-time multi-agent environments. This disclosure is applicable to autonomous traffic coordination, robotic systems, distributed computing, multi-objective optimization, and other multi-agent systems requiring dynamic optimization.

A system for data-driven, modular decision making and trajectory generation includes a computing system. A method for data-driven, modular decision making and trajectory generation includes: receiving a set of inputs; selecting a learning module such as a deep decision network and / or a deep trajectory network from a set of learning modules; producing an output based on the learning module; repeating any or all of the above processes; and / or any other suitable processes. Additionally or alternatively, the method can include training any or all of the learning modules; validating one or more outputs; and / or any other suitable processes and / or combination of processes.

One example method includes obtaining, using one or more autonomous agents, user information; generating, using the one or more autonomous agents and based on the user information, a plurality of prompts; generating, using the one or more autonomous agents, as a plurality of candidate digital components, one or more candidate digital components for each prompt of the plurality of prompts; determining, using the one or more autonomous agents, digital component scores of the plurality of candidate digital components; determining, using the one or more autonomous agents and based on the digital component scores of the plurality of candidate digital components, at least one digital component of the plurality of candidate digital components; and generating, using the one or more autonomous agents, an output digital component including the at least one digital component.

Scalable methods and systems include coordinating in real-time, the motion in a mapped environment of large groups of autonomous agents. The agents and machines may include autonomous mobile robots (AMRs) used in logistics and manufacturing, autonomous vehicles, drones, humanoids robots, or characters in a computer simulation or game. A coordinator server may periodically determine, during a polling period, a current location, velocity, and target destination of agents. The coordinator server may issue a sequence of travel commands for the agents to execute until the next polling period. In some implementations, the coordinator server is configured with an iterative pathfinding process to generate paths for the agents sequentially. The generated path of an agent at an iteration respects reservations in space-time made by previous agents in the same iteration and by other agents in the previous iteration. The computation of such paths can be distributed among one or more processors for scaling to large fleets of agents.

Task priority scheduling and resource allocation of autonomous agents are described. The scheduling may utilize sensor data characteristics to facilitate scheduling decisions. Also described is the refinement of task priority scheduling utilizing fleet management information-based scene and environment information, such as by using information available to the fleet management controller.

A method is proposed for determining human activity in an environment comprising at least one sensor for monitoring the use of at least one piece of environmental equipment. The method includes artificial intelligence (AI) training and AI inference to determine human activity in said environment. It is characterized in that the method comprises, for the AI ​​training: - the generation of AI training data from a simulation of the activity of at least one autonomous agent evolving in an environmental simulation generated from a digital twin of the real environment and using at least one digital replica of said equipment. Abstract: Figure 3

A method for conditional operation of an autonomous agent includes: collecting a set of inputs; processing the set of inputs; determining a set of policies for the agent; evaluating the set of policies; and operating the ego agent. A system for conditional operation of an autonomous agent includes a set of computing subsystems (equivalently referred to herein as a set of computers) and / or processing subsystems (equivalently referred to herein as a set of processors), which function to implement any or all of the processes of the method.

A multi-agent control system coordinates autonomous agents to efficiently search two-dimensional and three-dimensional surfaces for targets. The system employs a multi-phase decentralized control strategy including deployment, circle formation, and spiral search phases. During circle formation, agents use a distributed consensus-based auction algorithm with wind-aware or current-aware cost calculations to assign themselves to evenly-spaced angular positions around a mission center. Agents then execute search trajectories with adjacent passes spaced at twice the sensor footprint radius to ensure complete coverage without gaps or overlaps. For two-dimensional surfaces, trajectories follow Archimedean spirals proven optimal for targets with radially-symmetric probability distributions. For three-dimensional surfaces, trajectories follow geodesic coils that conform to surface geometry using geodesic distance fields and level-set projection. The system includes obstacle avoidance capabilities using best-gap selection. Applications include search and rescue, reconnaissance, environmental monitoring, and surveillance.

A system for addressing failure in an autonomous agent includes a driving subsystem, a control subsystem, a central computing subsystem, and an autonomous vehicle (AV) sensor subsystem. The system can optionally additionally include a power subsystem, a vehicle chassis subsystem, a communication subsystem, a distributed computing and / or processing subsystem, a supplementary sensor subsystem, and / or any other components. A method for addressing failure can include any or all of: detecting and responding to a failure; and operating the vehicle. Additionally or alternatively, the method 200 can include any other processes.