Methods, systems, and apparatus, including computer programs encoded on a computer storage medium that automatically calibrates robots and sensors assigned to perform a task in an operating environment. One of the methods includes obtaining a representation of a robotic operating environment. A user selection of a plurality of components to be configured to operate in the robotic operating environment is received. A mapping is obtained between pairs of components to be calibrated and one or more respective calibration processes to perform to calibrate each pair of components. From the mapping, one or more calibration processes to be performed on pairs of components based on the user selection of the plurality of components is computed. Calibration instruction data describing how to perform the one or more calibration processes to be performed on the pairs of components of the user selection is determined and presented.

Provided is a method of providing food to a user, the method including determining to provide first food among the food to the user; moving a first gripper to a container that contains the first food, determining whether the first gripper reciprocates in the container, calculating a weight difference value indicating an amount of change in a total weight of the food before and after the reciprocating in response to a determination that the first gripper reciprocates in the container, and determining that the first food is provided to the user based on the weight difference value. In addition, an apparatus for providing food to a user to perform the food providing method is provided. Also, a non-transitory computer-readable storage medium storing programs to perform the food providing method is provided.

A robot system can include a main body; a manipulator installed on the main body; a sensor configured to detect an object approaching a restricted region including the manipulator; a camera configured to monitor the restricted region and the object approaching the restricted region; a storage configured to store a material for an operation of the manipulator, the storage including an inlet for receiving the material; a remaining amount sensor configured to detect an amount of the material remaining in the storage; and a controller configured to change the restricted region based on at least one of a result of detection of the remaining amount sensor and image information of the camera, and in response to the sensor detecting that the object is within the restricted region, stop manipulation of the manipulator.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for providing user feedback for robotic demonstration learning. One of the methods includes initiating a local demonstration learning process to collect respective local demonstration data for each of one or more demonstration subtasks defined by a skill template to be executed by a robot. Local demonstration data is repeatedly collected for each of the one or more demonstration subtasks of the skill template while a user manipulates a robot to perform each of the one or more demonstration subtasks defined by the skill template. A respective progress value for each of the one or more demonstration subtasks defined by the skill template is maintained. A user interface presentation is generated that presents a suggested demonstration to be performed by the user based on a respective progress value for each demonstration subtask.

A method includes determining motion imitation information for causing a system to imitate a physical task using a first machine learning model that is trained using motion information that represents a performance of the physical task, determining a predicted correction based on the motion information and a current state from the system using a second machine learning model that is trained using the motion information, determining an action to be performed by the system based on the motion imitation information and the predicted correction; and controlling motion of the system in accordance with the action.

A control device includes: first circuitry that generates a command to cause a robot to autonomously grind a grinding target portion; second circuitry that generates a command to cause the robot to grind a grinding target portion according to manipulation information from an operation device; third circuitry that controls operation of the robot according to the command; storage that stores image data of a grinding target portion and operation data of the robot corresponding to the command; and forth circuitry that performs machine learning by using image data of a grinding target portion and the operation data for the grinding target portion, receives the image data as input data, and outputs an operation correspondence command corresponding to the operation data as output data. The first circuitry generates the command, based on the operation correspondence command.

A computer-implemented method of training a machine learnable model for controlling and/or monitoring a computer-controlled system. The machine learnable model is configured to make inferences based on a probability distribution of sensor data of the computer-controlled system. The machine learnable model is configured to account for symmetries in the probability distribution imposed by the system and/or its environment. The training involves sampling multiple samples of the sensor data according to the probability distribution. Initial values are sampled from a source probability distribution invariant to the one or more symmetries. The samples are iteratively evolved according to a kernel function equivariant to the one or more symmetries. The evolution uses an attraction term and a repulsion term that are defined for a selected sample in terms of gradient directions of the probability distribution and of the kernel function for the multiple samples.

A computer-implemented method executed by data processing hardware of a robot causes the data processing hardware to perform operations. The operations include receiving a sensor pointing command that commands the robot to use a sensor to capture sensor data of a location in an environment of the robot. The sensor is disposed on the robot. The operations include determining, based on an orientation of the sensor relative to the location, a direction for pointing the sensor toward the location, and an alignment pose of the robot to cause the sensor to point in the direction toward the location. The operations include commanding the robot to move from a current pose to the alignment pose. After the robot moves to the alignment pose and the sensor is pointing in the direction toward the location, the operations include commanding the sensor to capture the sensor data of the location in the environment.

A method for modularizing high dimensional neural networks into neural networks of lower input dimensions. The method is suited to generating full-DOF robot grasping actions based on images of parts to be picked. In one example, a first network encodes grasp positional dimensions and a second network encodes rotational dimensions. The first network is trained to predict a position at which a grasp quality is maximized for any value of the grasp rotations. The second network is trained to identify the maximum grasp quality while searching only at the position from the first network. Thus, the two networks collectively identify an optimal grasp, while each network's searching space is reduced. Many grasp positions and rotations can be evaluated in a search quantity of the sum of the evaluated positions and rotations, rather than the product. Dimensions may be separated in any suitable fashion, including three neural networks in some applications.

A robot teaching device capable of simplifying a work involved in teaching a robot. The robot teaching device includes a position data storing section configured to store target position data for a robot; an operation instruction storing section configured to store an operation instruction for arranging the robot at a target position, the operation instruction not including the target position data; and a position data writing section configured to acquire current position data of the robot when the operation instruction is input, the position data writing section further configured to write, to the position data storing section, the current position data as the target position data together with a unique identifier, in which the identifier of the target position data is automatically given to the operation instruction being input to teach an operation to the robot.