Abstract: One or more switchable magnetic systems is located near a battery attachment surface of an electric vehicle, where the battery attachment surface is configured to accept and electronically connect to a battery pack. The electric vehicle also includes a control system that operates the one or more switchable magnetic system switching from a first to a second state. In the first state the magnetic field of the switchable magnetic system secures the battery pack to the battery attachment surface so that the battery pack is electronically coupled to an electric motor of the electric vehicle. When the control system receives a message corresponding to removal of the battery pack, the control system switches the switchable magnetic system to the second state such that the battery pack is removable from the battery attachment surface.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, relating to an autonomous aerial personal assistant device (AAPAD). The AAPAD can include a base unit housing various electronic components, a balloon for providing lift to carry a weight of the base unit, and a positioning unit that maintains a position of the balloon based at least on feedback provided by a plurality of sensors and computers. In some implementations, a request to initiate a communication session can be received on the AAPAD. Data indicating a presence of the user can then be obtained. A preset distance between the user and the AAPAD can be determined. Instructions to one or more components of the AAPAD can then be transmitted to automatically navigate the AAPAD to travel toward a location of the user.

Abstract: An in-ear device includes a housing shaped to be held in an ear of a user, and be positioned around a canal of the ear so the canal is unobstructed by the in-ear device, and an audio package positioned within the housing to project sound into the canal when the in-ear device is disposed in the ear. A controller is disposed in the housing and coupled to the audio package, and the controller includes logic that when executed by the controller causes the in-ear device to perform operations, including emitting the sound from the audio package towards the canal of the ear.

Abstract: Embodiments described herein may relate to a system comprising a plurality of optical elements, comprising at least a first optical element and one or more secondary optical elements, a heliostat comprising the first optical element, where the heliostat is operable to move the first optical element to continuously reflect light from a non-stationary light source in a beam towards a first of the secondary optical elements, and where the secondary optical elements are arranged to re-direct the reflected beam of light towards an illumination target. The system further includes a controller configured to receive position data indicative of the position of the non-stationary light source over time, and in response to the position data, control at least the heliostat to continuously direct the beam of light towards the first of the secondary optical elements, such that the beam of light is continuously re-directed towards the illumination target.

Abstract: Robotic control systems and methods may include providing an end effector tool of a robotic device configured to perform a task on a work surface within a worksite coordinate frame. Unintended movement over time of the end effector tool with respect to the work surface and with respect to the worksite coordinate frame may be determined based on image data indicative of the work surface, first location data indicative of a first location of the end effector tool with respect to the worksite coordinate frame, and second location data indicative of a second location of the end effector tool with respect to the work surface. One or more control signals for the robotic device may be adjusted in order to counteract the unintended movements of the end effector tool with respect to the work surface and worksite coordinate frame.

Abstract: A method is provided that includes controlling a robotic gripping device to cause a plurality of digits of the robotic gripping device to move towards each other in an attempt to grasp an object. The method also includes receiving, from at least one non-contact sensor on the robotic gripping device, first sensor data indicative of a region between the plurality of digits of the robotic gripping device. The method further includes receiving, from the at least one non-contact sensor on the robotic gripping device, second sensor data indicative of the region between the plurality of digits of the robotic gripping device, where the second sensor data is based on a different sensing modality than the first sensor data. The method additionally includes determining, using an object-in-hand classifier that takes as input the first sensor data and the second sensor data, a result of the attempt to grasp the object.

Abstract: Methods, systems, and devices for managing robot resources are described. A robot receives from an application a request to reserve a particular set of physical resources of the robot. The robot then determines that each of the physical resources in the set are available to the application and, based on the determination, allocates exclusive use of the particular set of resources to the application by (i) generating a token corresponding to the set of resources, (ii) providing the token to the application, and (iii) updating token data that associates the token with the set of resources. The robot then controls access to the particular set of resources such that, while token data indicates that the token is valid, commands from applications that involve the set of resources are only executed when provided with the token corresponding to the allocation of access to the particular set of resources.

Abstract: A technique for encouraging a user towards a selected mental or emotional state includes: determining an influencing directive for encouraging the user towards the selected mental or emotional state; selecting environmental content, based upon the influencing directive, from a plurality of different types of environmental content; presenting the environmental content to the user via sensory immersion equipment, wherein the includes at least one environmental display for altering an ambience of a room; generating observation data based upon observing a user reaction to the environmental content with an observation system, the observation system including at least one sensor capable of observing the user reaction; and determining whether to adjust the environmental content based upon the observation data to encourage the user towards the selected mental or emotional state.

Abstract: An application specific integrated circuit (ASIC) chip includes: a systolic array of cells; and multiple controllable bus lines configured to convey data among the systolic array of cells, in which the systolic array of cells is arranged in multiple tiles, each tile of the multiple tiles including 1) a corresponding subarray of cells of the systolic array of cells, 2) a corresponding subset of controllable bus lines of the multiple controllable bus lines, and 3) memory coupled to the subarray of cells.

Abstract: An example apparatus includes a first disk that is rotatable and has a plurality of electro-permanent magnets disposed in a radial array on a surface of the first disk; and a second disk rotatably mounted adjacent to the first disk such that a gap separates the second disk from the first disk, where the second disk has a plurality of ferromagnetic elements disposed in respective radial array on a respective surface of the second disk. Applying an electric pulse to at least one electro-permanent magnet of the plurality of electro-permanent magnets changes a magnetic state of the electro-permanent magnet, thereby (i) generating an external magnetic field that traverses the gap between the first disk and the second disk and interacts with a corresponding ferromagnetic element of the plurality of ferromagnetic elements, and (ii) causing the second disk to rotate as the first disk rotates.

Abstract: Techniques are provided for making data stored in channels of shared memory available across a communication network that couples multiple distinct processing units (e.g., of a robot). In various implementations, a channel may be established in shared memory of a first processing unit. A publisher process executing on the first processing unit may publish data to the channel, and a remote subscriber process executing on a second processing unit may subscribe to the channel. A persistent network connection over a communication network may be established between the first and second processing units. Data published to the channel by the publisher process may be read directly from the local memory of the first processing unit by or on behalf of one or more local subscriber processes, and may be transmitted over the communication network to the remote subscriber process using the persistent network connection.

Abstract: A nuclear magnetic resonance (NMR) apparatus includes a pair of permanent magnets spaced apart from each other and defining a sample space, the permanent magnets providing a magnetic field in the sample space, an array of magnetic field sensors arranged relative to the sample space to provide, during operation, information about a homogeneity of the magnetic field, an array of magnetic coils arranged relative to the sample space so that each magnetic coil, during operation, generates a magnetic field that changes the magnetic field in the sample space, and an electronic controller programmed to receive information about the magnetic field from the array of magnetic field sensors and to variably energize the magnetic coils in the array of magnetic coils so that the magnetic fields generated by the array of magnetic coils reduces inhomogeneities of the magnetic field.

Abstract: Implementations are provided for improved robot reachability maps. Multiple reachability maps for a robot may be indexed on a distance of an end effector from a plane and a tilt of the end effector. Each reachability map may include multiple cells, each representing a base pose of the robot. A target pose of the end effector may be used to determine a target distance/tilt, which is used to select a given reachability map. A given cell of the given reachability map may be used to determine a first candidate set of joint parameters for the robot. In some implementations, one or more additional, neighboring cells of the plurality of cells may be used to determine additional candidate set(s) of joint parameters. A given set of joint parameters to be implemented to achieve the target pose of the end effector may be selected and used to operate the robot.

Abstract: Systems and methods are provided for specifying safety rules for robotic devices. A computing device can determine information about any actors present within a predetermined area of an environment. The computing device can determine a safety classification for the predetermined area based on the information. The safety classification can include: a low safety classification if the information indicates zero actors are present within the predetermined area, a medium safety classification if the information indicates any actors are present within the predetermined area all are of a predetermined first type, and a high safety classification if the information indicates at least one actor present within the predetermined area is of a predetermined second type. After determining the safety classification for the predetermined area, the computing device can provide a safety rule for operating within the predetermined area to a robotic device operating in the environment.

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. At least portions 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: Example embodiments include determining a map of an environment of a robotic vehicle. The map includes locations of a plurality of mapped landmarks within the environment and a false detection source region within the environment. The embodiments further include detecting a plurality of candidate landmarks, and determining which of the detected candidate landmarks correspond to one of the plurality of mapped landmarks and which correspond to false detections. The embodiments additionally include estimating a pose of the robotic vehicle within the environment. The embodiments further include determining which of the detected candidate landmarks determined to correspond to false detections fall within the false detection source region. The embodiments still further include determining a confidence level of the pose estimate based on which of the detected candidate landmarks determined to correspond to false detections fall within the false detection source region.

Abstract: The disclosure provides for a system that includes a plurality of stations equipped for free-space optical communications (FSOC) in a network and a central control system. At least one station in the plurality of stations includes a wavelength selectable switch, an OEO module, and one or more first processors. The one or more first processors are configured to control the wavelength selectable switch, process an electrical signal that is extracted using the OEO module, and communicate with the central control system. The central control system includes one or more second processors that are configured to receive data regarding FSOC communication conditions at the plurality of stations, determine a path between stations through the network based on the received data, and transmit instructions to the plurality of stations.

Abstract: A reusable mechanism is disclosed for coupling two robotic appendages, such that an unintended force acting against a side of one of the appendages may decouple the appendages. The mechanism includes a revolved male dovetail mated to a revolved female dovetail. The mechanism may further include a channel within the male dovetail and a detent that inhibits rotation of the male dovetail in relation to the female dovetail.

Abstract: Methods, systems, and apparatus for receiving a reference to an object located in an environment of a robot, accessing mapping data that indicates, for each of a plurality of object instances, respective probabilities of the object instance being located at one or more locations in the environment, wherein the respective probabilities are based at least on an amount of time that has passed since a prior observation of the object instance was made, identifying one or more particular object instances that correspond to the referenced object, determining, based at least on the mapping data, the respective probabilities of the one or more particular object instances being located at the one or more locations in the environment, selecting, based at least on the respective probabilities, a particular location in the environment where the referenced object is most likely located, and directing the robot to navigate to the particular location.

Abstract: Techniques are provided to estimate of location or position of objects that are depicted in an image of a scene. Some implementations include obtaining an image of a scene; identifying an object within the image of the scene; obtaining a three-dimensional model that corresponds to the object that was identified within the image of the scene, the three-dimensional model being obtained from the database of three-dimensional models; determining, based on data from the three-dimensional model, an estimated depth of the object within the scene; generating or updating a three-dimensional representation of the scene based at least on the estimated depth of the object within the scene; and providing the three-dimensional representation of the scene, including at least a portion of the three-dimensional representation of the scene that was generated or updated based on the three-dimensional model of the object, to the scene analyzer.