Provided are systems and methods for training a robot. The method commences with collecting, by the robot, sensor data from a plurality of sensors of the robot. The sensor data may be related to a task being performed by the robot based on an artificial intelligence (AI) model. The method may further include determining, based on the sensor data and the AI model, that a probability of completing the task is below a threshold. The method may continue with sending a request for operator assistance to a remote computing device and receiving, in response to sending the request, teleoperation data from the remote computing device. The method may further include causing the robot to execute the task based on the teleoperation data. The method may continue with generating training data based on the sensor data and results of execution of the task for updating the AI model.

There is provided a control apparatus including: a model output acquisition unit configured to acquire an operation model output which is output according to inputting state data indicating a state of a facility, to an operation model trained by machine learning to output an action in accordance with the state of the facility by using simulation data from a simulator that simulates an operation in the facility; an index acquisition unit configured to acquire an index which is calculated by using real data from the facility and which is for monitoring a difference between a behavior of the simulator and an actual operation in the facility; a correction unit configured to correct the operation model output based on the index; and a control unit configured to output a manipulated variable for controlling a control target provided in the facility, according to the corrected operation model output.

A robotic arm having one or more hybrid (soft-rigid) joints includes a first link, a second link, and a joint interconnecting the first link and the second link, such that the first link is movable relative to the second link along an axis of motion. The joint includes: a socket component coupled to a distal end portion of the second link and a ball component coupled to a proximal end portion of first link, the ball component is configured to rotationally fit within the socket component. The joint also includes a flexible membrane encasing the socket component and the ball component. The robotic arm is controllable using a reinforcement learning algorithm training using a simulation of the robotic arm and optionally, further training in a physical world.

Mitigating the reality gap through utilization of technique(s) that enable compliant robotic control and/or compliant robotic contact to be simulated effectively by a robotic simulator. The technique(s) can include, for example: (1) utilizing a compliant end effector model in simulated episodes of the robotic simulator; (2) using, during the simulated episodes, a soft constraint for a contact constraint of a simulated contact model of the robotic simulator; and/or (3) using proportional derivative (PD) control in generating joint control forces, for simulated joints of the simulated robot, during the simulated episodes. Implementations additionally or alternatively relate to determining parameter(s), for use in one or more of the techniques that enable effective simulation of compliant robotic control and/or compliant robotic contact.

A robot control device includes: a trained model built by being trained on work data; a control data acquisition section which acquires control data of the robot based on data from the trained model; base trained models built for each of a plurality of simple operations by being trained on work data; an operation label storage section which stores operation labels corresponding to the base trained models; a base trained model combination information acquisition section which acquires combination information when the trained model is represented by a combination of a plurality of the base trained models, by acquiring a similarity between the trained model and the respective base trained models; and an information output section which outputs the operation label corresponding to each of the base trained models which represent the trained model.

Various aspects of techniques, systems, and use cases include robot safety. A device in a network may include processing circuitry and memory including instructions, which when executed by the processing circuitry, cause the processing circuitry to perform operations. The operations may include collecting telemetry data for a robot, the robot operating according to a path control plan generated using reinforcement learning with a safety factor as a reward function, and detecting that a safety event, involving a robot action, has occurred with the robot and an object. The operations may include simulating a recreation of the safety event to determine whether a simulated action matches the robot action.

An adjustment support device includes: a storage unit for storing, with force state data and position data in an operation when performing force control of the industrial robot as a state variable and with data indicating a result of determining whether a result of the force control is success or failure based on predetermined criteria as determination data, a learning model generated by machine learning; an analysis unit for analyzing the learning model to analyze, for a control parameter used when the force control of the industrial robot has failed, an adjustment method of the control parameter for improving a degree of success of the force control; and an adjustment determination unit for determining, based on a result of the analysis by the analysis unit, an adjustment method of the control parameter in the force control used when the force control has failed and outputting the adjustment method.

Techniques are disclosed to perform task allocation for autonomous systems by implementing machine-learning to perform task allocation to Autonomous Mobile Robots (AMRs) in an environment. The disclosed techniques also provide for enhanced path planning and the identification of AMR health and failure prediction to further improve upon task allocation and system efficiency.

Various aspects of methods, systems, and use cases include techniques for training or using a model to control a robot. A method may include identifying a set of action primitives applicable to a set of robots, receiving information corresponding to a task (e.g., a collaborative task), and determining at least one action primitive based on the received information. The method may include training a model to control operations of at least one robot of the set of robots using the received information and the at least one action primitive.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an industrial plant controller that controls operation of an industrial plant. In one aspect, a method comprises generating training data using an industrial plant simulation model that simulates operation of the industrial plant. The industrial plant controller is trained by a reinforcement learning technique using the training data. The industrial plant controller is configured to process an input comprising a state vector characterizing a state of the industrial plant in accordance with a plurality of industrial plant controller parameters to generate an action selection policy output that defines a control action to be performed to control the operation of the industrial plant.