A system comprises a database; at least one hardware processor coupled with the database; and one or more software modules that, when executed by the at least one hardware processor, receive at least one of sensory data from a robot and images from a camera, identify and build models of objects in an environment, wherein the model encompasses immutable properties of identified objects including mass and geometry, and wherein the geometry is assumed not to change, estimate the state including position, orientation, and velocity, of the identified objects, determine based on the state and model, potential configurations, or pre-grasp poses, for grasping the identified objects and return multiple grasping configurations per identified object, determine an object to be picked based on a quality metric, translate the pre-grasp poses into behaviors that define motor forces and torques, communicate the motor forces and torques to the robot in order to allow the robot to perform a complex behavior generated from the behaviors.

A first method comprising: predicting a scene of an environment using a model of the environment and based on a first scene of the environment obtained from sensors observing scenes of the environment; comparing the predicted scene with an observed scene from the sensors; and performing an action based on differences determined between the predicted scene and the observed scene. A second method comprising applying a vibration stimuli on an object via a computer-controlled component; obtaining a plurality of images depicting the object from a same viewpoint, captured during the application of the vibration stimuli. The second method further comprising comparing the plurality of images to detect changes occurring in response to the application of the vibration stimuli, which changes are attributed to a change of a location of a boundary of the object; and determining the boundary of the object based on the comparison.

A grasping robot includes: a grasping mechanism configured to grasp a target object; an image-pickup unit configured to shoot a surrounding environment; an extraction unit configured to extract a graspable part that can be grasped by the grasping mechanism in the surrounding environment by using a learned model that uses an image acquired by the image-pickup unit as an input image; a position detection unit configured to detect a position of the graspable part; a recognition unit configured to recognize a state of the graspable part by referring to a lookup table that associates the position of the graspable part with a movable state thereof; and a grasping control unit configured to control the grasping mechanism so as to displace the graspable part in accordance with the state of the graspable part recognized by the recognition unit.

Apparatuses, systems, and techniques to grasp objects with a robot. In at least one embodiment, a neural network is trained to determine a grasp pose of an object within a cluttered scene using a point cloud generated by a depth camera.

A control device includes a target position setting part, a trajectory estimation part, and a target position selector. The target position setting part determines a target position of an actor of a robot based on a form of an object located in an operating environment of the robot. The trajectory estimation part estimates a predicted trajectory of the actor based on motion of the actor up to the present, and estimates a trajectory of the actor from a current position to the target position as an approach trajectory using a predetermined function. The target position selector selects one target position based on a degree of similarity between the predicted trajectory and each of the approach trajectory.

A robot comprising: a chopstick, configured for at least four degrees of freedom of movement, a stiff body of shape and proportions approximate to a pool cue; an electromagnetic actuator, comprising a motor, for each degree of freedom of movement coupled with the stiff body, wherein the functional mapping from each actuator's motor current to torque output along an axis of motion is stored, and used in concert with a calibrated model of the robot for effective impedance control; and a 6-axis force/torque sensor mounted inline between the actuators and each chopstick.

A system and method is provided for determining grasping positions for two-handed grasps of industrial objects. The system may include a processor configured to determine a three dimensional (3D) voxel grid for a 3D model of a target object. In addition, the processor may be configured to determine at least one pair of spaced apart grasping positions on the target object at which the target object is capable of being grasped with two hands at the same time based on processing the 3D voxel grid for the target object with a neural network trained to determine grasping positions for two-handed grasps of target objects using training data. Such training data may include 3D voxel grids of a plurality of 3D models of training objects and grasping data including corresponding pairs of spaced-apart grasping positions for two-handed grasps of the training objects. Also, the processor may be configured to provide output data that specifies the determined grasping positions on the target object for two-handed grasps.

In some aspects, a system comprises a computer hardware processor and a non-transitory computer-readable storage medium storing processor-executable instructions for receiving, from one or more sensors, sensor data relating to a robot; generating, using a statistical model, based on the sensor data, first control information for the robot to accomplish a task; transmitting, to the robot, the first control information for execution of the task; and receiving, from the robot, a result of execution of the task.

A robot apparatus includes a grasping section that grasps an object, a recognition section that recognizes a graspable part and a handing-over area part of the object, and a grasp planning section that plans a path of the grasping section for handing over the object to a recipient by the handing-over area part. The robot apparatus further includes a grasp control section that controls grasp operation of the object by the grasping section in accordance with the planned path.

Training and/or use of a machine learning model for placement of an object secured by an end effector of a robot. A trained machine learning model can be used to process: (1) a current image, captured by a vision component of a robot, that captures an end effector securing an object; (2) a candidate end effector action that defines a candidate motion of the end effector; and (3) a target placement input that indicates a target placement location for the object. Based on the processing, a prediction can be generated that indicates likelihood of successful placement of the object in the target placement location with application of the motion defined by the candidate end effector action. At many iterations, the candidate end effector action with the highest probability is selected and control commands provided to cause the end effector to move in conformance with the corresponding end effector action. When at least one release criteria is satisfied, control commands can be provided to cause the end effector to release the object, thereby leading to the object being placed in the target placement location.