Abstract: A cleaning robot may determine a three-dimensional model of a physical environment based on data collected from one or more sensors. The cleaning robot may then identify a surface within the physical environment to clean. Having identified that surface, the robot may autonomously navigate to a location proximate to the surface, position an ultraviolet light source in proximity to the surface, and activate the ultraviolet light source for a period of time.

Abstract: A model of a physical environment may be determined based at least in part on sensor data collected by one or more sensors at a robot. The model may include a plurality of constraints and a plurality of data values. A trajectory through the physical environment may be determined for an ultraviolet end effector coupled with the robot to clean one or more surfaces in the physical environment. The ultraviolet end effector may include one or more ultraviolet light sources. The ultraviolet end effector may be moved along the trajectory.

Abstract: An interactive robot having mechanical torso, limbs, and head assembled for movement with multiple degrees of freedom to enable life-like movements and responses. The robot's head may further include LED displays as the eyes and mouth, and a speaker associated with the mouth. These features enable life-like audio and visual responses, including complex facial expressions and conversational audio interaction.

Abstract: A robot may identify a human located proximate to the robot in a physical environment based on sensor data captured from one or more sensors on the robot. A trajectory of the human through space may be predicted. When the predicted trajectory of the human intersects with a current path of the robot, an updated path to a destination location in the environment may be determined so as to avoid a collision between the robot and the human along the predicted trajectory. The robot may then move along the determined path.

Abstract: A cleaning robot may determine a three-dimensional model of a physical environment based on data collected from one or more sensors. The cleaning robot may then identify a surface within the physical environment to clean. Having identified that surface, the robot may autonomously navigate to a location proximate to the surface, position an ultraviolet light source in proximity to the surface, and activate the ultraviolet light source for a period of time.

Abstract: A model of a physical environment may be determined based at least in part on sensor data collected by one or more sensors at a robot. The model may include a plurality of constraints and a plurality of data values. A trajectory through the physical environment may be determined for an ultraviolet end effector coupled with the robot to clean one or more surfaces in the physical environment. The ultraviolet end effector may include one or more ultraviolet light sources. The ultraviolet end effector may be moved along the trajectory.

Abstract: Methods, apparatus, systems, and computer-readable media are provided for enabling users to approximately identify a space within an environment inhabited by a plurality of objects that user wishes for a robot to manipulate. In various implementations, an approximation of a space within an environment may be identified based on user input. The actual space within the environment may then be extrapolated based at least in part on the approximation and one or more attributes of the environment. A plurality of objects that are co-present within the space and that are to be manipulated by a robot may be identified. The robot may then be operated to manipulate the identified plurality of objects.

Abstract: Some implementations are directed to methods and apparatus for determining, based on sensor data generated during physical manipulation of a robot by a user, one or more grasp parameters to associate with an object model. Some implementations are directed to methods and apparatus for determining control commands to provide to actuator(s) of a robot to attempt a grasp of an object, where those control commands are determined based on grasp parameters associated with an object model that conforms to the object. The grasp parameter(s) associated with an object model may include end effector pose(s) that each define a pose of a grasping end effector relative to the object model and/or translational force measure(s) that each indicate force applied to an object by a grasping end effector, where the force is at least partially the result of translation of an entirety of the grasping end effector.

Abstract: Methods, apparatus, systems, and computer-readable media are provided for enabling users to approximately identify a space within an environment inhabited by a plurality of objects that user wishes for a robot to manipulate. In various implementations, an approximation of a space within an environment may be identified based on user input. The actual space within the environment may then be extrapolated based at least in part on the approximation and one or more attributes of the environment. A plurality of objects that are co-present within the space and that are to be manipulated by a robot may be identified. The robot may then be operated to manipulate the identified plurality of objects.