X Development LLC

Redwood Robotics Inc.

Google Inc.

Other

Meka Robotics, LLC

Aaron Ladd Edsinger

Advait Jain

Anthony Sean Jules

Stefan Nusser

Lee Harris Magnusson

Philip Mullins

Anthony Rizzi Jules

Jeff Anthony Weber

James Nicholas Vines

Kevin Kemper

Nicolas Hudson

Justine Rembisz

Adrian Li

Andy Metzger

Matt Carney

Jeff Vora Weber

Attila Kabai

Methods, apparatus, systems, and computer-readable media are provided for generating spatial affordances for an object, in an environment of a robot, and utilizing the generated spatial affordances in one or more robotics applications directed to the object. Various implementations relate to applying vision data as input to a trained machine learning model, processing the vision data using the trained machine learning model to generate output defining one or more spatial affordances for an object captured by the vision data, and controlling one or more actuators of a robot based on the generated output. Various implementations additionally or alternatively relate to training such a machine learning model.

Generating and utilizing spatial affordances for an object in robotics applications

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.

Autonomous aerial personal assistant device

An example robot actuator utilizing a differential pulley transmission is provided. As an example, a differential pulley actuator includes input drive gears for coupling to a motor and timing pulleys coupled together through the input drive gears. Rotation of the input drive gears causes rotation of a first timing pulley in a first direction and rotation of a second timing pulley in a second direction opposite the first direction. The actuator also includes multiple idler pulleys suspended between the timing pulleys and the output pulley, and the multiple idler pulleys are held in tension between the timing pulleys via a first tension-bearing element and the output pulley via a second tension-bearing element. The first tension-bearing element loops around the timing pulleys and the multiple idler pulleys. The output pulley couple to a load, and is configured to apply motion of the multiple idler pulleys to the load.

Robot actuator utilizing a differential pulley transmission

Example implementations described herein may provide a pipeline from a model of a given object to a model of one or more fingertips that are specialized to grasp the given object. An example system may receive a three-dimensional geometric model of a given object. The system may also iterate over a plurality of fingertip geometries to determine a particular fingertip geometry that is compliant to a shape of the given object at a grasp point on the given object. The system may further iterate the particular fingertip geometry over a plurality of fingertip sizes to determine a particular fingertip size that is compliant to one or more dimensions of the given object at the grasp point of the given object; and the system may provide a model of one or more fingertips having the particular fingertip geometry and the particular fingertip size.

End-of-arm fingertip pipeline

A method operable by a computing device is provided. The method may include receiving a request for a given task to be performed by a robotic system. The method may also determining one or more subtasks required to perform the given task, where the one or more subtasks include one or more parameters used to define the one or more subtasks. The method may also include determining an arrangement of the one or more subtasks to perform the given task, and providing for display an indication of the one or more undefined parameters for the given task. The method may also include receiving an input defining the one or more undefined parameters for the given task, and executing the one or more subtasks in the determined arrangement and in accordance with the one or more defined parameters to cause the robotic system to perform the given task.

Methods and systems for tiered programming of robotic device

An example torque controlled actuator includes a frame, and one or more timing belt stages coupled in serial on the frame. The timing belt stages include an input stage for coupling to a motor and an output stage for coupling to a load, and the timing belt stages couple rotation of the motor to rotation of an output of the output stage. The torque controlled actuator also includes one or more belt idlers coupled to the frame that contact a timing belt of the output stage, and a strain gauge coupled to the frame to determine a tension of the timing belt of the output stage based on force applied by the timing belt of the output stage to the one or more belt idlers.

Robot manipulator with modular torque controlled link

A system is provided, including one or more servers in communication with a robotic system. The one or more servers may be configured to receive operational data from the robotic system, and determine one or more operational performance metrics based on the received operational data. The system may also include a first computing device in communication with the robotic system including a workstation authoring software application configured to program the given task to be completed by the robotic system, and determine one or more subtasks required for the robotic system to perform the given task. The system may also include a second computing device in communication with the robotic system including an operational dashboard software application configured to control various operations of the robotic system, and provide for display a visual representation of the operational data and the operational performance metrics on an interface of the second computing device.

Distributed system for management and analytics of robotic devices

A block-and-tackle transmission includes a timing belt input pinion, a timing belt, two or more shuttles, an output cable, and an output hub. The timing belt input pinion is for receiving input power. The timing belt is driven by the input pinion. The timing belt causes two shuttles of the two or more shuttles to move in opposing directions. Opposing ends of the output cable are pulled by two of the two or more shuttles. The output cable causes the output hub to transmit output power.

Low cost block and tackle robot transmission

An example modular reconfigurable workcell for quick connection of peripherals is described. In one example, a modular reconfigurable workcell comprises modular docking bays on a surface of the workcell that support attachment of docking modules in a fixed geometric configuration, and respective modular docking bays include electrical connections for a variety of power and communication busses of the docking modules to be attached. The workcell also includes an electrical subsystem for coupling the communication busses between the modular docking bays and providing power circuitry to the modular docking bays, and structural features in the modular docking bays to enable insertion of the docking modules in the fixed geometric configuration. The workcell also includes a processor for determining a geometric calibration of attached peripherals based on a location and the orientation of corresponding docking modules attached to the modular docking bays and based on an identification of the attached peripherals.

Modular reconfigurable workcell for quick connection of peripherals

Examples described may relate to methods and systems for gesture based switch for machine control. Some machines may be operated by computing devices that include touchscreens, and a graphical user interface (GUI) is provided for enabling control of the machines. Within examples, a disconnect switch can be incorporated into the GUI. As one example, to initiate operation of a machine, such as a robotic device, a user may be required to contact the touchscreen at a location and then trace out a pattern. To enable continued operation of the machine, the GUI may require the user to maintain contact with the touchscreen at a 'goal' position. If contact with the touchscreen is broken (e.g., for more than a threshold amount of time), the machine operation may be halted, and the process to initiate operation can be performed again on the touchscreen to cause the machine to resume operation.

Methods and systems for gesture based switch for machine control

A method operable by a computing device is provided. The method may include receiving a request for a given task to be performed by a modular reconfigurable workcell. The method may also include determining one or more peripherals required to perform the given task. The method may also include determining an optimal placement of the one or more peripherals based on the given task, wherein the one or more peripherals are coupled to the workcell in a fixed geometric configuration based on the determined optimal placement. The method may also include determining a first calibration of the one or more peripherals based on the orientation of the one or more peripherals relative to the workcell, and determining a second calibration of the one or more peripherals based on the optimal placement of the one or more peripherals with respect to each other.

Methods and systems for rapid configuration, calibration, and deployment of a modular reconfigurable mounting structure

Examples are provided that describe a motion based light display for a robotic arm. In one example, a robotic device comprising one or more components configured to be actuated for movement. The robotic device also includes one or more processors are configured to determine a motion per path metric of the one or more components based on a motion plan associated with the robotic device. The one or more processors are configured to determine one or more feedback characteristics based on the motion per path metric. The one or more feedback characteristics include information indicative of an effect of motion associated with the one or more components. The robotic device also includes an indicator coupled to the one or more components and configured to provide feedback about the one or more components based on the feedback characteristics indicative of the effect of motion.

Methods and systems to provide feedback based on a motion per path metric indicative of an effect of motion associated with components of a robotic device

Examples are provided that describe a cycloid transmission with an adjustable ring. An example cycloid transmission includes a disc and a motor shaft attached to the disc. The motor shaft is capable of rotating the disc around an outer ring of rollers. The outer ring of rollers surrounds the disc. As the disc is rotated, contact is made between the disc and the outer ring of rollers. A cycloid transmission also comprises an adjustable ring that is interposed between the motor shaft and the disc. A circumference of the adjustable ring can be adjusted in order to cause expansion of a radius of the disc. This expansion will result in increased contact of the disc with the outer ring of rollers and thereby lower backlash during rotation of the disc.

Cycloid transmission with an adjustable ring

A robotic device may: receive movement information associated with a plurality of subtasks performed by a manipulator of a robotic device, where the movement information indicates respective paths followed by the manipulator while performing the respective subtasks and respective forces experienced by the manipulator along the respective paths; determine task information for a task to be performed by the robotic device, where the task comprises a combination of subtasks of the plurality of subtasks, where the task information includes a trajectory to be followed by the manipulator, and forces to be exerted by the manipulator at points along the trajectory; and determine, based on the task information, torques to be applied over time to the manipulator via a joint coupled to the robotic device to perform the task.

Growing community of inventors

San Francisco, CA, United States of America

Aaron Ladd Edsinger