Methods, apparatus, and computer-readable media for determining and utilizing corrections to robot actions. Some implementations are directed to updating a local features model of a robot in response to determining a human correction of an action performed by the robot. The local features model is used to determine, based on an embedding generated over a corresponding neural network model, one or more features that are most similar to the generated embedding. Updating the local features model in response to a human correction can include updating a feature embedding, of the local features model, that corresponds to the human correction. Adjustment(s) to the features model can immediately improve robot performance without necessitating retraining of the corresponding neural network model.

A corpus curation method, system, and non-transitory computer readable medium, include mapping a kinematic motion of a robot to a granular feature of an item in the corpus and answering a user question using the mapped kinematic motion embedded in an answer by the robot.

Apparatus and methods for training and controlling of, for instance, robotic devices. In one implementation, a robot may be trained by a user using supervised learning. The user may be unable to control all degrees of freedom of the robot simultaneously. The user may interface to the robot via a control apparatus configured to select and operate a subset of the robot's complement of actuators. The robot may comprise an adaptive controller comprising a neuron network. The adaptive controller may be configured to generate actuator control commands based on the user input and output of the learning process. Training of the adaptive controller may comprise partial set training. The user may train the adaptive controller to operate first actuator subset. Subsequent to learning to operate the first subset, the adaptive controller may be trained to operate another subset of degrees of freedom based on user input via the control apparatus.

Apparatus and methods for training and controlling of e.g., robotic devices. In one implementation, a robot may be utilized to perform a target task characterized by a target trajectory. The robot may be trained by a user using supervised learning. The user may interface to the robot, such as via a control apparatus configured to provide a teaching signal to the robot. The robot may comprise an adaptive controller comprising a neuron network, which may be configured to generate actuator control commands based on the user input and output of the learning process. During one or more learning trials, the controller may be trained to navigate a portion of the target trajectory. Individual trajectory portions may be trained during separate training trials. Some portions may be associated with robot executing complex actions and may require additional training trials and/or more dense training input compared to simpler trajectory actions.

A robot system including a master device configured to receive a manipulating instruction from an operator and transmit the received manipulating instruction as a manipulating input signal, a plurality of slave robots configured to operate according to the manipulating input signal transmitted from the master device, a management control device configured to manage operations of the plurality of slave robots, respectively, and an output device configured to output information transmitted from the management control device. The management control device determines a priority of transmitting the manipulating input signal from the master device to the slave robot among the plurality of slave robots that are in a standby state of the manipulating input signal, and transmits information related to the determined priority to the output device. Thus, the operator is able to efficiently transmit the manipulating input signal to the plurality of slave robots through the master device.

Plurality of robot main bodies a remote control device including contactless action detecting part configured to detect contactless action including at least one given operation instructing action by operator, and control device communicably connected to remote control device and configured to control operations of plurality of robot main bodies, are provided. Control device includes memory part configured to store operational instruction content data defining operation mode of robot main body corresponding to the at least one operation instructing action, operational instruction content identifying module configured to identify operation mode of robot main body corresponding to one of operation instructing action detected by contactless action detecting part based on operational instruction content data, and motion controlling module configured to control operation of at least one given robot main body among plurality of robot main bodies based on operation mode identified by operational instruction content identifying module.

A robot system which is capable of reducing an operator's workload and easily correcting preset operation of a robot. The robot system includes a robot main body having a plurality of joints, a control device configured to control operation of the robot main body and an operating device including a teaching device configured to teach the control device one of positional information on the robot main body and angular information on the plurality of joints so as to execute an automatic operation of the robot main body and a manipulator configured to receive a manipulating instruction from an operator to manually operate the robot main body or correct the operation of the robot main body under the automatic operation.

Methods, apparatus, and computer-readable media for determining and utilizing corrections to robot actions. Some implementations are directed to updating a local features model of a robot in response to determining a human correction of an action performed by the robot. The local features model is used to determine, based on an embedding generated over a corresponding neural network model, one or more features that are most similar to the generated embedding. Updating the local features model in response to a human correction can include updating a feature embedding, of the local features model, that corresponds to the human correction. Adjustment(s) to the features model can immediately improve robot performance without necessitating retraining of the corresponding neural network model.

A robot system which includes a manipulator configured to receive a manipulating instruction from an operator, a slave arm having a plurality of joints, and a control device configured to control operation of the slave arm. The control device is configured, while the slave arm is operating at a speed equal to or higher than a first given the threshold, even when an operational instruction value for correcting the operation of the slave arm is inputted from the manipulator during an automatic operation of the slave arm, to prevent the correction of the operation of the slave arm.

Grasping of an object, by an end effector of a robot, based on a grasp strategy that is selected using one or more machine learning models. The grasp strategy utilized for a given grasp is one of a plurality of candidate grasp strategies. Each candidate grasp strategy defines a different group of one or more values that influence performance of a grasp attempt in a manner that is unique relative to the other grasp strategies. For example, value(s) of a grasp strategy can define a grasp direction for grasping the object (e.g., “top”, “side”), a grasp type for grasping the object (e.g., “pinch”, “power”), grasp force applied in grasping the object, pre-grasp manipulations to be performed on the object, and/or post-grasp manipulations to be performed on the object.