30 results about "Human operator" patented technology

An energy-conscious workflow planning system based on LLM with dynamic resource allocation, consisting of: a workflow input interface configured to receive workflow-directed acyclic graphs (DAGs), energy budget constraints, system performance constraints, and natural language requests from human operators; a large language model (LLM) logic module connected to the workflow input interface and configured to analyze the workflow specifications and system constraints in natural language, generate energy-conscious planning recommendations based on the analyzed workflow specifications, and provide explainable planning rationales in natural language; a reinforcement learning-based scheduling unit connected to the LLM reasoning module and configured to: receive scheduling recommendations from the LLM reasoning agent, fine-tune task-resource assignments by dynamically adapting to runtime variations, and perform online resource redistribution under runtime variability; an energy monitoring unit configured to: continuously monitor CPU and GPU utilization in heterogeneous clusters, track power consumption and thermal limits per node, and generate energy profiles for system components; a multi-objective optimization engine configured to: perform a Pareto-optimal scheduling analysis that balances energy consumption, lead time and reliability, apply statistical and AI-supported trade-off analyses and ensure optimal resource allocation based on Pareto frontier analysis; a dynamic resource allocation unit configured to: use predictive models that incorporate LLM inferences and feedback from reinforcement learning, reassign tasks between nodes and clusters while minimizing energy consumption and improving system throughput based on the predictive models; a performance optimization module configured to optimize scheduling decisions using multi-criteria optimization analysis; and a user interface that allows human operators to override and refine planning strategies in real time based on verifiable planning reasons.

The invention discloses a multi-objective reinforcement learning man-machine cooperative assembly task allocation method based on neighborhood parameter migration, and the method comprises the steps: building a mathematical model which aims at minimizing the physiological fatigue accumulated value of a human operator and minimizing the maximum completion time for a multi-objective optimization problem of task allocation in a man-machine cooperative assembly system; a multi-target problem is decomposed into N standard sub-problems by adopting a weighting and decomposition strategy, and training is accelerated through a neighborhood parameter migration strategy; each sub-problem is solved based on a near-end strategy optimization algorithm of an Actor-Critic framework, Gaussian noise is added to an Actor network to simulate environment uncertainty, an action mask mechanism is introduced to process priority constraints of assembly tasks, and it is ensured that a generated task allocation scheme is always feasible. According to the method, the convergence speed and diversity of the Pareto solution set can be remarkably improved, and the assembly efficiency and operator fatigue are effectively balanced.

An object of a system according to an embodiment is to improve service experience by responding to an inquiry from a customer immediately and on a 24-hour basis.SOLUTION: A system according to an embodiment includes a natural language processing unit, a learning unit, a transfer unit, a translation unit, an aggregation unit, and a feedback analysis unit. The natural language processor uses natural language processing to understand the intent of the customer's question and generate an appropriate answer. The learner evolves to provide optimal answers from past dialogues. The forwarder automatically forwards the complex query to a human operator. The translation unit supports multiple languages, allows a customer to ask a question in his / her native language, and automatically translates the question to provide an answer. The aggregation unit aggregates inquiry leads of a plurality of services into one application. The feedback analyzer automatically analyzes the feedback from the customer and sends it immediately to the product.SELECTED DRAWING: Figure 1

The present invention relates to a method, a computer program with instructions, and devices for controlling a means of transportation. In a first step, a situation requiring assistance is detected (10). Subsequently, a data set for the situation is determined (11). This data set includes at least environmental data that was acquired before the situation occurred. Optionally, based on the data set, it is checked (12) whether an AI model implemented in the means of transportation is capable of resolving the situation. If this is not the case, the data set is sent to a control center (13). The control center receives (14) the data set and checks (15) whether an AI model of the control center is capable of resolving the situation. Instructions for resolving the situation are then generated (16), either by the appropriate AI model or by a human operator. The instructions are sent to the means of transportation (17).The means of transport receives (18) the instructions from the control center and executes them. Optionally, feedback can be sent to the control center (19) indicating whether the instructions were successful in resolving the situation. The invention further relates to a means of transport in which a method or device according to the invention is used, as well as a computer model for use in a method according to the invention.

Human operators (201A) perform activities within a facility in interactions with technical equipment (301A, 301B, 301C). A series (700) of images (701, 702) visualizes activities that are performed by either the same human operator (201A) or by different human operators. In the images, a computer separates areas (701x, 702x) that comprise biometric image data from areas (701y, 702y) with data that indicates the interactions. The computer obtains identifiers (ALPHA, ALPHA) of the operators (201A) but removes the biometric image. The computer then combines images areas and provides an audit record (500) by storing the modified first and second images (701′, 701′) and the identifiers.

Techniques for providing decision rationales for machine-learning guided processes are described herein. In some embodiments, the techniques described herein include processing queries for an explanation of an outcome of a set of one or more decisions guided by one or more machine-learning processes with supervision by at least one human operator. Responsive to receiving the query, a system determines, based on a set of one or more rationale data structures, whether the outcome was caused by human operator error or the one or more machine-learning processes. The system then generates a query response indicating whether the outcome was caused by the human operator error or the one or more machine-learning processes.

Systems and methods for controlling a performance of a digital character portrayed at a display device are disclosed. According to at least one embodiment, a method for controlling a performance of a digital character portrayed at a display device includes determining a presence of a person located in a physical environment, and in response to determining the presence of the person, facilitating control of the performance of the digital character portrayed at the display device by a human operator, by an artificial intelligence (AI) game engine, or by a combination thereof.

We provide a contact center system that can smoothly connect users to human operators if their requests cannot be resolved by automated responses such as chatbots. [Solution] In a contact center system equipped with a processor, when the processor receives a user request from a user terminal, it performs the following processes: extracting a standard response sentence corresponding to the request from a storage unit that stores standard response sentences and presenting it to the user terminal; and, if the request is not resolved by the standard response sentence, presenting an interactive element to the user terminal along with the standard response sentence for the user to start verbal communication with an operator. When the interactive element is operated by the user, it provides the user terminal with a connection to a pre-registered destination associated with the standard response sentence presented together with the interactive element.

A human operator (201A) performs an activity within a facility that interacts with the technical equipment (301A, 301B, 301C). A series (700) of images (701, 702) visualizes activities performed by the same human operator (201A) or different human operators. In the image, the computer separates a region (701x, 702x) including biometric image data from a region (701y, 702y) having data indicative of the interaction. The computer obtains an identifier (ALPHA, ALPHA) of the operator (201A) but removes the biometric image. The computer then combines the image regions and provides a review record (500) by storing the modified first and second images (701 ', 701') and the identifier.

The present invention relates to systems and methods for ingesting of raw client data into a supply chain optimization system. The client enterprise data system (EDS) provides raw data, typically in tabular format across many files, to a data management module of the optimization system. This data is then profiled. and a set of mapping AI models is applied to the profiled data. Each of the plurality of mapping AI models corresponds to a single input feature of the optimization model. As the mapping models are applied, a live preview is generated. This live preview is reviewable by a human operator, and human input can be provided. When there is input, the individual mapping model is updated (without impacting the other mapping models), and a new live preview for that given feature is generated. The resulting standardized feature set may be consumed by the supply chain optimization model.

We provide the system. [Solution] Means for receiving inquiry data, A natural language processing means for analyzing the content of the aforementioned query data, A means for classifying the content of inquiries based on the aforementioned analysis results, A means of generating responses to inquiries using generative artificial intelligence, Means for transmitting the generated response, A means of transferring the inquiry to a human operator based on the aforementioned analysis results, A system that includes this.

The present disclosure provides a control system for a humanoid robot comprising a bipedal action model (BAM) with hierarchical architecture including a beta model executing cognitive tasks at lower frequency, ingesting multimodal sensory inputs including visual data and natural language instructions, and an alpha model executing reactive tasks at higher frequency, communicatively coupled to the beta model. The BAM is trained on retargeted robot training data derived from robot-free training data. At runtime, the BAM outputs continuous control commands as parallel-generated action chunks controlling at least 18 degrees of freedom. The system includes a wearable collection apparatus capturing movement data from a human operator without physical connection to the robot, and a retargeting module translating robot-free training data into robot training data by solving embodiment mismatches between human and robot kinematic structures.

A system for robotic skill learning using vision-force servoing control. A robot performs an assembly operation and a camera, mounted in the workspace or on the robot arm, provides images of the operation. An image encoder neural network is trained to output a vision signal comprising pose features indicating distance from target location, where the vision signal is greatly reduced in size compared to the images. The vision signal is provided to a second neural network along with force / torque and tool center point velocity data. The second neural network operates as a vision-force servo controller, and is co-trained using offline reinforcement learning pre-training and online human operator correction. A pseudo-random sampling technique is used to improve the training of the vision-force control neural network. The vision-force servo controller outputs a next robot motion which is used by a compliance controller to control motion of the robot during the assembly operation.

Systems and methods presented herein facilitate operation of well-related tools. In certain embodiments, a variety of data (e.g., downhole data and / or surface data) may be collected to enable optimization of operations related to the well-related tools. In certain embodiments, the collected data may be provided as advisory data (e.g., presented to human operators of the well to inform control actions performed by the human operators) and / or used to facilitate automation of downhole processes and / or surface processes (e.g., which may be automatically performed by a computer implemented surface processing system (e.g., a well control system), without intervention from human operators). In certain embodiments, the systems and methods described herein may enhance downhole operations (e.g., milling operations) by improving the efficiency and utilization of data to enable performance optimization and improved resource controls of the downhole operations.

The invention relates to the technical field of robot control, in particular to a ground search and rescue robot sharing control method based on multi-level man-machine instruction fusion, which comprises the following steps: after a task-level instruction sent by a human is received, a robot performs global optimal path search; after a behavior-level instruction sent by a human is received, the robot performs risk assessment on the local environment in the map, and if a risk value is greater than or equal to a threshold value, the robot does not respond to the behavior-level instruction; otherwise, executing the behavior-level instruction, and searching the nearest path point in the global path to recover the task after execution. And after an action level instruction sent by a human operator is received, according to the calculated risk value, the man-machine authority is dynamically allocated, and an optimal control instruction is generated. According to the method, autonomous planning and manual intervention can be dynamically balanced, the adaptability to complex scenes is improved, the burden of operators is reduced, and the efficiency loss caused by high-frequency micro-operation is avoided.

The application relates to the technical field of autonomous value keeping based on a rule tree, and discloses an autonomous value keeping method and system based on a rule tree, which comprises the following steps: the autonomous value keeping method and system based on the rule tree provided by the application introduce a rule tree and a behavior logic engine, realize unmanned autonomous operation of a training station, break the dependence of traditional training on human operators, effectively reduce the idle time of the training station, and fully utilize existing resources; each operation task can be automatically executed according to a preset rule tree logic, the continuity and consistency of a training process are ensured, training interruption caused by the intermittence of manual operation is avoided, and the coherence of a training task is ensured; a dynamic rule reasoning mechanism is adopted, an operation decision path is adjusted through real-time state feedback, the judgment logic and the behavior mode of a real operator can be simulated, the system has a certain adaptability, and the intelligent level of simulation training is improved.

The invention provides a telephone outbound man-machine cooperation judgment method and system based on AI. The method comprises the steps of obtaining multi-modal data such as texts, audio emotion and customer portraits; taking a text as a query, taking an emotion and a portrait as key values, and carrying out lightweight cross attention fusion and pooling to obtain fusion features; constructing a non-cyclic state pool, aggregating historical states through time decay, and splicing the historical states with the current features to form state enhanced representation; and inputting the representation into a lightweight scoring module, jointly predicting a turning probability and a dynamic threshold value, and judging whether turning to manpower or not according to a comparison result of the two. The method is low in time delay and high in judgment accuracy, continuous optimization can be carried out along with service feedback, and the service experience and the resource utilization rate are improved.

A twin roll casting system comprises counter-rotating casting rolls having a nip between the casting rolls and capable of delivering cast strip downwardly from the nip, a casting roll controller configured to adjust at least one process control setpoint between the casting rolls in response to control signals, a cast strip sensor capable of measuring at least one parameter of the cast strip, and a controller coupled to the cast strip sensor to receive cast strip measurement signals from the cast strip sensor and coupled to the casting roll controller to provide control signals to the casting roll controller, the controller comprising a reinforcement learning (RL) Agent. The RL Agent further comprises a model-free actor-critic agent having a value function and a policy function, the RL Agent having been trained on a plurality of casting system operation datasets composed of casting runs executed by a plurality of different human operators.

The application relates to the technical field of robots, and provides a multi-degree-of-freedom bionic manipulator control method. The method first establishes a kinematics model and a dynamics relationship of the bionic manipulator, and on this basis, a multi-modal training and real-time control method for a 20-degree-of-freedom bionic manipulator is provided. The method comprehensively utilizes three heterogeneous sensing modalities of inertial measurement, computer vision and surface electromyography to synchronously perceive the hand movement intention of a human operator, and realizes fusion regression of multi-source information based on a deep neural network to output target control instructions of joints of the manipulator. In the training stage, an inertial measurement system is used to provide high-precision joint angle true values; in the deployment stage, vision and surface electromyography are mainly used as network inputs, and real-time closed-loop control is realized in combination with feedback at the manipulator end, so that high-precision control of the multi-degree-of-freedom bionic manipulator is realized.

A system uses a machine-learning model to solve micro-escalation events. The system receives, from an artificial intelligence (Al) agent, a stream of input data from a user during a real-time conversation between the Al agent and the user. The system applies the machine-learning model to the stream of input data to identify micro-escalation events. The system creates a communication channel between the agent and one or more human operators. The system assigns each micro-escalation event to a human operator via the communication channel. The system receives a result generated by a respective human operator for each micro-escalation event via the communication channel. The system dynamically integrates the results to form an overall response to the stream of input data from the agent into the real-time conversation.

An entity (14), method and systems for automatically assigning an interaction between a respective user device (10-1, 10-2, … 10-N) and a user assistance centre (15) to an operator device (12-1) are described. The entity comprises a monitoring unit (141) configured for carrying out a monitoring of a first interaction between a first user (10-1) and the user assistance centre (15), wherein the first interaction is assigned to a first operator device (12-1). A prediction unit (142) is present to provide, while a second interaction is in progress and on the basis of said monitoring, a prediction of switching said first interaction to an artificial or human operator device, wherein the second interaction is an interaction between a second user (10-2) and the user assistance centre. An assignment unit (143) assigns, on the basis of said prediction, the second interaction (10-2) to one among said first operator device and another one among one or more operator devices.

A system includes one or more robot arms configured to pick and / or place items into totes. The system is designed to reduce overall downtime. In the system, if the robot arm is unable to pick and / or place items into the tote, the system routes the tote to a human operator who rearranges the items in the tote in such a way that the robot arm is more likely to be able to pick the items from the tote. The system further includes an artificial intelligence (AI) system that determines whether totes are able to be picked by the robot arm. The AI system further monitors the operation of the robot arms and human operators for training purposes. The data from picking and rearranging operations are used to train the AI system.

Techniques for task error correction for robots, such as collaborative robots (cobots), are disclosed. A robot controller can include an error detector for detecting errors during the execution of a collaborative human-robot task and an error corrector for correcting the detected error. The error corrector can include a correction planner and a moderator. The correction planner can determine an error correction plan based on the detected error. The error correction plan can include corrective subtasks to control the cobot to correct the detected error. The moderator can determine a moderation plan based on the determined error correction plan. The moderation plan includes an assistance subtask configured to control the cobot to assist a human operator in correcting the detected error.The error corrector can generate a control signal to control the cobot based on the correction plan and the moderation plan.

A computer system and method for dynamic interaction between human operators and persistent cognitive machines is disclosed. The invention enables adaptive collaboration in operational environments by integrating multimodal translation, cognitive processing, load balancing, trust calibration, operational learning, and team coordination. Human inputs such as voice, gestures, biometric signals, and contextual data are converted into prompts for a cognitive core that processes reasoning through multi-stage language models and thought caching. Operator cognitive load is quantified by combining physiological and behavioral indicators, and tasks are dynamically allocated between human and machine based on load, task complexity, and trust. Operational modes transition between advisory, collaborative, autonomous, and override states with safeguards to ensure stability and human primacy. Outputs are adapted in detail, modality, and timing according to operator state. Continuous learning captures interaction patterns and team dynamics, providing personalized adaptations, distributed knowledge sharing, and resilience to component failures.

One example method is performed at a far edge device and includes collecting data with one or more IoT (Internet of Things) devices, feeding the data to a feedback loop that includes multiple stages, running the feedback loop, providing learning information, comprising output from one or more of the stages of the feedback loop, to a central manager by way of a learn interface, accessing learning information generated by one or more other far edge devices, and updating the feedback loop using the learning information generated by the one or more other far edge devices.

A computer-implemented method for assisting a human operator in manually adjusting a vehicle's speed. The method includes receiving live data related to the vehicle's current speed, applying logic to input data including the live data, and sending a signal. The signal is sent to the human operator based on the logic. The signal is at least one of quantifying the vehicle's future speed and indicating that the future speed exceeds a threshold.

A humanoid robot control method, system, storage medium and program product, in the method, the pressure distribution data is weighted and calculated to obtain the pressure barycenter coordinates of the cooperative contact point; a time window sequence is constructed based on the pressure barycenter coordinates, and the instantaneous velocity and acceleration of the pressure barycenter are obtained through difference operation; the size and direction of the resultant force exerted by the human operator at the cooperative contact point are calculated based on the instantaneous velocity and acceleration; the size and direction of the resultant force are transmitted along the joint chain of the humanoid robot to be converted to obtain the instantaneous force state of each joint of the humanoid robot; the corresponding joint output torque is calculated according to the instantaneous force state of each joint; and the joints of the humanoid robot are controlled to move cooperatively according to the joint output torque. The present application is used for accurately identifying and predicting the human cooperation intention, realizing accurate force matching and natural action synchronization in the human-machine cooperation process, so as to improve the safety and efficiency of the cooperation task.

Systems and methods for controlling performance of a digital character depicted at a display device are disclosed. According to at least one embodiment, a method for controlling performance of a digital character depicted at a display device includes: determining a presence of a person located in a physical environment; and responsive to determining the presence of the person, facilitating control of performance of the digital character depicted at the display device by a human operator, by an artificial intelligence (AI) game engine, or by a combination thereof.

A system uses a machine-learning model to solve micro-escalation events. The system receives, from an artificial intelligence (AI) agent, a stream of input data from a user during a real-time conversation between the AI agent and the user. The system applies the machine-learning model to the stream of input data to identify micro-escalation events. The system creates a communication channel between the agent and one or more human operators. The system assigns each micro-escalation event to a human operator via the communication channel. The system receives a result generated by a respective human operator for each micro-escalation event via the communication channel. The system dynamically integrates the results to form an overall response to the stream of input data from the agent into the real-time conversation.

The system according to the embodiment aims to automate the identification of a fault location and the presentation of a recovery method in server operation in an on-premise environment. [Solution] A system according to an embodiment includes a learning unit, an investigation unit, a presentation unit, and an execution unit. The learning unit studies a server configuration diagram or design document. The investigation unit investigates the location of a failure in the server based on the information learned by the learning unit. The presentation unit presents a recovery method based on the location of the failure investigated by the investigation unit. The execution unit allows a human operator to review the recovery method presented by the presentation unit and automatically executes approved operations.