Abstract: Implementations are directed to training a machine learning model that, once trained, is used in performance of robotic grasping and/or other manipulation task(s) by a robot. The model can be trained using simulated training examples that are based on simulated data that is based on simulated robot(s) attempting simulated manipulations of various simulated objects. Portion(s) of the model can also be trained based on real training examples that are based on data from real-world physical robots attempting manipulations of various objects. The simulated training examples can be utilized to train the model to predict an output that can be utilized in a particular task—and the real training examples used to adapt at least a portion of the model to the real-world domain can be tailored to a distinct task. In some implementations, domain-adversarial similarity losses are determined during training, and utilized to regularize at least portion(s) of the model.

Abstract: Methods and apparatus related to generating a model for an object encountered by a robot in its environment, where the object is one that the robot is unable to recognize utilizing existing models associated with the robot. The model is generated based on vision sensor data that captures the object from multiple vantages and that is captured by a vision sensor associated with the robot, such as a vision sensor coupled to the robot. The model may be provided for use by the robot in detecting the object and/or for use in estimating the pose of the object.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for estimating a pose of an object of interest. One of the methods includes receiving an input including an image that represents a pose of an object, processing the image using a machine learning model to predict output for the image, based on the output from the machine learning model, determining a correspondence between pixels in the image and locations on a three-dimensional model of the object, and determining the pose of the object based on the correspondence. The input further includes pre-processing output after pre-processing data associated with the object. The determined pose are processed by a downstream module to generate an updated pose.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing deformation modeling with edge-to-edge constraints. One of the methods includes computing edge-to-edge constraints that match one or more edges in point cloud data to one or more edges in a nominal representation of a workpiece. A current estimate of the deformation data is repeatedly updated according to the computed edge-to-edge constraints. A robotic manipulation task is then performed according to the generated deformation data.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating synthetic data to be used in training machine learning models in an auto-scalable manner. In one aspect, an auto-scalable synthetic data generation system maintains a plurality of synthetic data generator replicas that are each configured to generate synthetic training examples; maintains a plurality of machine learning training workers that are each configured to obtain synthetic training examples and to use the synthetic training examples to concurrently perform operations required to update the machine learning model; determines, by an autoscaler of the synthetic data generation system, that a number of synthetic data generator replicas is insufficient to service a current demand level of the plurality of machine learning training workers; and in response, deploys, by the autoscaler, one or more additional synthetic data generator replicas in the synthetic data generation system.

Abstract: Methods and apparatus related to generating a model for an object encountered by a robot in its environment, where the object is one that the robot is unable to recognize utilizing existing models associated with the robot. The model is generated based on vision sensor data that captures the object from multiple vantages and that is captured by a vision sensor associated with the robot, such as a vision sensor coupled to the robot. The model may be provided for use by the robot in detecting the object and/or for use in estimating the pose of the object.

Abstract: Particular techniques for generating synthetic images and/or for training machine learning model(s) based on the generated synthetic images. For example, training a machine learning model based on training instances that each include a generated synthetic image, and ground truth label(s) for the generated synthetic image. After training of the machine learning model is complete, the trained machine learning model can be deployed on one or more robots and/or one or more computing devices.

Abstract: Methods for annotating objects within image frames are disclosed. Information is obtained that represents a camera pose relative to a scene. The camera pose includes a position and a location of the camera relative to the scene. Data is obtained that represents multiple images, including a first image and a plurality of other images, being captured from different angles by the camera relative to the scene. A 3D pose of the object of interest is identified with respect to the camera pose in at least the first image. A 3D bounding region for the object of interest in the first image is defined, which indicates a volume that includes the object of interest. A location and orientation of the object of interest is determined in the other images based on the defined 3D bounding region of the object of interest and the camera pose in the other images.

Abstract: Methods and apparatus related to generating a model for an object encountered by a robot in its environment, where the object is one that the robot is unable to recognize utilizing existing models associated with the robot. The model is generated based on vision sensor data that captures the object from multiple vantages and that is captured by a vision sensor associated with the robot, such as a vision sensor coupled to the robot. The model may be provided for use by the robot in detecting the object and/or for use in estimating the pose of the object.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for planning robotic movements to perform a given task while satisfying object pose estimation accuracy requirements. One of the methods includes generating a plurality of candidate measurement configurations for measuring an object to be manipulated by a robot; determining respective measurement accuracies for the plurality of candidate measurement configurations; determining a measurement accuracy landscape for the object including defining a high measurement accuracy region based on the respective measurement accuracies for the plurality of candidate measurement configurations; and generating a motion plan for manipulating the object in the robotic process that moves the robot, a sensor, or both, through the high measurement accuracy region when performing pose estimation for the object.

Abstract: Particular techniques for generating synthetic images and/or for training machine learning model(s) based on the generated synthetic images. For example, training a machine learning model based on training instances that each include a generated synthetic image, and ground truth label(s) for the generated synthetic image. After training of the machine learning model is complete, the trained machine learning model can be deployed on one or more robots and/or one or more computing devices.

Abstract: Particular techniques for generating synthetic images and/or for training machine learning model(s) based on the generated synthetic images. For example, training a machine learning model based on training instances that each include a generated synthetic image, and ground truth label(s) for the generated synthetic image. After training of the machine learning model is complete, the trained machine learning model can be deployed on one or more robots and/or one or more computing devices.

Abstract: Example embodiments may relate to methods and systems for selecting a grasp point on an object. In particular, a robotic manipulator may identify characteristics of a physical object within a physical environment. Based on the identified characteristics, the robotic manipulator may determine potential grasp points on the physical object corresponding to points at which a gripper attached to the robotic manipulator is operable to grip the physical object. Subsequently, the robotic manipulator may determine a motion path for the gripper to follow in order to move the physical object to a drop-off location for the physical object and then select a grasp point, from the potential grasp points, based on the determined motion path. After selecting the grasp point, the robotic manipulator may grip the physical object at the selected grasp point with the gripper and move the physical object through the determined motion path to the drop-off location.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for planning robotic movements to perform a given task while satisfying object pose estimation accuracy requirements. One of the methods includes generating a plurality of candidate measurement configurations for measuring an object to be manipulated by a robot; determining respective measurement accuracies for the plurality of candidate measurement configurations; determining a measurement accuracy landscape for the object including defining a high measurement accuracy region based on the respective measurement accuracies for the plurality of candidate measurement configurations; and generating a motion plan for manipulating the object in the robotic process that moves the robot, a sensor, or both, through the high measurement accuracy region when performing pose estimation for the object.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing image data. One of the method includes receiving an input image from a source domain, the input image showing an object to be manipulated by a robot in a robotic process; processing the input image to generate an intermediate representation of the input image, comprising: generating a gradient orientation representation and a gradient magnitude representation of the input image; and generating the intermediate representation of the input image from the gradient orientation representation and the gradient magnitude representation; processing the intermediate representation of the input image using a neural network trained to make predictions about objects in images to generate a network output that represents a prediction about physical characteristics of the object in the input image.

Abstract: Methods and apparatus related to generating a model for an object encountered by a robot in its environment, where the object is one that the robot is unable to recognize utilizing existing models associated with the robot. The model is generated based on vision sensor data that captures the object from multiple vantages and that is captured by a vision sensor associated with the robot, such as a vision sensor coupled to the robot. The model may be provided for use by the robot in detecting the object and/or for use in estimating the pose of the object.

Abstract: Methods and apparatus related to generating a model for an object encountered by a robot in its environment, where the object is one that the robot is unable to recognize utilizing existing models associated with the robot. The model is generated based on vision sensor data that captures the object from multiple vantages and that is captured by a vision sensor associated with the robot, such as a vision sensor coupled to the robot. The model may be provided for use by the robot in detecting the object and/or for use in estimating the pose of the object.

Abstract: Methods, apparatus, and computer readable media that are related to 3D object detection and pose determination and that may optionally increase the robustness and/or efficiency of the 3D object recognition and pose determination. Some implementations are generally directed to techniques for generating an object model of an object based on model point cloud data of the object. Some implementations of the present disclosure are additionally and/or alternatively directed to techniques for application of acquired 3D scene point cloud data to a stored object model of an object to detect the object and/or determine the pose of the object.

Abstract: Methods, apparatus, systems, and computer-readable media are provided for delegating object type and/or pose detection to a plurality of “targeted object recognition modules.” In some implementations, a method may be provided that includes: operating an object recognition client to facilitate object recognition for a robot; receiving, by the object recognition client, sensor data indicative of an observed object in an environment; providing, by the object recognition client, to each of a plurality of remotely-hosted targeted object recognition modules, data indicative of the observed object; receiving, by the object recognition client, from one or more of the plurality of targeted object recognition modules, one or more inferences about an object type or pose of the observed object; and determining, by the object recognition client, information about the observed object, such as its object type and/or pose, based on the one or more inferences.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a feature extraction neural network to generate domain-invariant feature representations from domain-varying input images. In one aspect, the method includes obtaining a training dataset comprising a first set of target domain images and a second set of real domain images that each have pixel-wise level alignment with a corresponding target domain image, and training the feature extraction neural network on the training dataset based on optimizing an objective function that includes a term that depends on a similarity between respective feature representations generated by the network for a pair of target and source domain images.