Abstract: Aspects of the present disclosure provide techniques to undo a portion of a mission recording of a robot by physically moving the robot back through the mission recording in reverse. As a result, after the undo process is completed, the robot is positioned at an earlier point in the mission and the user can continue to record further mission data from that point. The portion of the mission recording that was performed in reverse can be omitted from subsequent performance of the mission, for example by deleting that portion from the mission recording or otherwise marking that portion as inactive. In this manner, the mistake in the initial mission recording is not retained, but the robot need not perform the entire mission recording again.

Abstract: Techniques are described that determine motion of a robot's body that will maintain an end effector within a useable workspace when the end effector moves according to a predicted future trajectory. The techniques may include determining or otherwise obtaining the predicted future trajectory of the end effector and utilizing the predicted future trajectory to determine any motion of the body that is necessary to maintain the end effector within the useable workspace. In cases where no such motion of the body is necessary because the predicted future trajectory indicates the end effector will stay within the useable workspace without motion of the body, the body may remain stationary, thereby avoiding the drawbacks caused by unnecessary motion described above. Otherwise, the body of the robot can be moved while the end effector moves to ensure that the end effector stays within the useable workspace.

Abstract: A method of localizing a robot includes receiving odometry information plotting locations of the robot and sensor data of the environment about the robot. The method also includes obtaining a series of odometry information members, each including a respective odometry measurement at a respective time. The method also includes obtaining a series of sensor data members, each including a respective sensor measurement at the respective time. The method also includes, for each sensor data member of the series of sensor data members, (i) determining a localization of the robot at the respective time based on the respective sensor data, and (ii) determining an offset of the localization relative to the odometry measurement at the respective time. The method also includes determining whether a variance of the offsets determined for the localizations exceeds a threshold variance. When the variance among the offsets exceeds the threshold variance, a signal is generated.

Abstract: Disclosed herein are systems and methods directed to an industrial robot that can perform mobile manipulation (e.g., dexterous mobile manipulation). A robotic arm may be capable of precise control when reaching into tight spaces, may be robust to impacts and collisions, and/or may limit the mass of the robotic arm to reduce the load on the battery and increase runtime. A robotic arm may include differently configured proximal joints and/or distal joints. Proximal joints may be designed to promote modularity and may include separate functional units, such as modular actuators, encoder, bearings, and/or clutches. Distal joints may be designed to promote integration and may include offset actuators to enable a through-bore for the internal routing of vacuum, power, and signal connections.

Abstract: Consistent connection strategies for coupling accessories to a robot can help achieve certain objectives, e.g., to tolerate and correct misalignment during coupling of the accessory. In some embodiments, the connection strategy may enable certain accessories to connect to certain sides of a robot. When connected, an accessory may be rigid in yaw, lateral motion, and fore/aft motion, while remaining unconstrained in roll and pitch as well as vertical motion. A sensor may enable detection of the accessory, and a mechanical fuse may release the accessory when a force threshold is exceeded. A mechanical coupler of an accessory may include two connectors, each of which includes a receiving area configured to receive a pin on the robot and a latch configured to retain the pin within the receiving area. The pins (and the receiving areas) may be differently sized, and may be differently arranged.

Abstract: Methods and apparatus for determining a grasp strategy to grasp an object with a gripper of a robotic device are described. The method comprises generating a set of grasp candidates to grasp a target object, wherein each of the grasp candidates includes information about a gripper placement relative to the target object, determining, for each of the grasp candidates in the set, a grasp quality, wherein the grasp quality is determined using a physical-interaction model including one or more forces between the target object and the gripper located at the gripper placement for the respective grasp candidate, selecting, based at least in part on the determined grasp qualities, one of the grasp candidates, and controlling the robotic device to attempt to grasp the target object using the selected grasp candidate.

Abstract: Methods and apparatus for controlling lighting of a mobile robot are provided. A mobile robot includes a drive system configured to enable the mobile robot to be driven, a navigation module configured to provide control instructions to the drive system, a plurality of lighting modules, wherein each of the plurality of lighting modules includes a plurality of individually-controllable light sources, and a controller configured to control an operation of the plurality of individually-controllable light sources based, at least in part, on navigation information received from the navigation module.

Abstract: Methods and apparatuses for detecting one or more objects (e.g., dropped objects) by a robotic device are described. The method comprises receiving a distance-based point cloud including a plurality of points in three dimensions, filtering the distance-based point cloud to remove points from the plurality of points based on at least one known surface in an environment of the robotic device to produce a filtered distance-based point cloud, clustering points in the filtered distance-based point cloud to produce a set of point clusters, and detecting one or more objects based, at least in part, on the set of point clusters.

Abstract: Methods and apparatus for object detection and pick order determination for a robotic device are provided. Information about a plurality of two-dimensional (2D) object faces of the objects in the environment may be processed to determine whether each of the plurality of 2D object faces matches a prototype object of a set of prototype objects stored in a memory, wherein each of the prototype objects in the set represents a three-dimensional (3D) object. A model of 3D objects in the environment of the robotic device is generated using one or more of the prototype objects in the set of prototype objects that was determined to match one or more of the 2D object faces.

Abstract: A virtual bumper configured to protect a component of a robotic device from damage is provided. The virtual bumper comprises a plurality of distance sensors arranged on the robotic device and at least one computing device configured to receive distance measurement signals from the plurality of distance sensors, detect, based on the received distance measurement signals, at least one object in a motion path of the component, and control the robot to change one or more operations of the robot to avoid a collision between the component and the at least one object.

Abstract: Some robotic arms may include vacuum-based grippers. Detecting the seal quality between each vacuum assembly of the gripper and a grasped object may enable reactivation of some vacuum assemblies, thereby improving the grasp. One embodiment of a method may include activating each of a plurality of vacuum assemblies of a robotic gripper by supplying a vacuum to each vacuum assembly, determining, for each of the activated vacuum assemblies, a first respective seal quality of the vacuum assembly with a first grasped object, deactivating one or more of the activated vacuum assemblies based, at least in part, on the first respective seal qualities, and reactivating each of the deactivated vacuum assemblies within a reactivation interval.

Abstract: An apparatus for decoupling angular adjustments about perpendicular axes is described herein. The apparatus comprises a first plate, a second plate offset from the first plate in a first direction, a first pivot disposed between the first and second plates, a second pivot disposed between the first and second plates. The second pivot is offset from the first pivot in a second direction perpendicular to the first direction. The third pivot is disposed between the first and second plates. The third pivot is offset from the first pivot in a third direction perpendicular to both the first and second directions. The apparatus further includes a first wedge at least partially disposed between the second pivot and the second plate. The first wedge is configured to adjust a first angle between the first and second plates, the first angle being about a first axis extending along the third direction.

Abstract: Method and apparatus for object detection by a robot are provided. The method comprises analyzing using a set of trained detection models, one or more first images of an environment of the robot to detect one or more objects in the environment of the robot, generating at least one fine-tuned model by training one or more of the trained detection models in the set, wherein the training is based on a second image of the environment of the robot and annotations associated with the second image, wherein the annotations identify one or more objects in the second image, updating the set of trained detection models to include the generated at least one fine-tuned model, and analyzing using the updated set of trained detection models, one or more third images of the environment of the robot to detect one or more objects in the environment.

Abstract: A robot system includes: an upper body section including one or more end-effectors; a lower body section including one or more legs; and an intermediate body section coupling the upper and lower body sections. An upper body control system operates at least one of the end-effectors. The intermediate body section experiences a first intermediate body linear force and/or moment based on an end-effector force acting on the at least one end-effector. A lower body control system operates the one or more legs. The one or more legs experience respective surface reaction forces. The intermediate body section experiences a second intermediate body linear force and/or moment based on the surface reaction forces. The lower body control system operates the one or more legs so that the second intermediate body linear force balances the first intermediate linear force and the second intermediate body moment balances the first intermediate body moment.

Abstract: A method for perception and fitting for a stair tracker includes receiving sensor data for a robot adjacent to a staircase. For each stair of the staircase, the method includes detecting, at a first time step, an edge of a respective stair of the staircase based on the sensor data. The method also includes determining whether the detected edge is a most likely step edge candidate by comparing the detected edge from the first time step to an alternative detected edge at a second time step, the second time step occurring after the first time step. When the detected edge is the most likely step edge candidate, the method includes defining, by the data processing hardware, a height of the respective stair based on sensor data height about the detected edge. The method also includes generating a staircase model including stairs with respective edges at the respective defined heights.

Abstract: Methods, apparatus, systems, and computer-readable media are provided for generating a spatio-temporal object inventory based on object observations from mobile robots and determining, based on the spatio-temporal object inventory, monitoring parameters for the mobile robots for one or more future time periods. Some implementations relate to using the spatio-temporal object inventory to determine a quantity of movements of objects that occur in one or more areas of the environment when one or more particular criteria are satisfied—and using that determination to determine monitoring parameters that can be utilized to provide commands to one or more of the mobile robots that influence one or more aspects of movements of the mobile robots at future time periods when the one or more particular criteria are also satisfied.

Abstract: An example method may include i) determining a first distance between a pair of feet of a robot at a first time, where the pair of feet is in contact with a ground surface; ii) determining a second distance between the pair of feet of the robot at a second time, where the pair of feet remains in contact with the ground surface from the first time to the second time; iii) comparing a difference between the determined first and second distances to a threshold difference; iv) determining that the difference between determined first and second distances exceeds the threshold difference; and v) based on the determination that the difference between the determined first and second distances exceeds the threshold difference, causing the robot to react.

Abstract: An example implementation for determining mechanically-timed footsteps may involve a robot having a first foot in contact with a ground surface and a second foot not in contact with the ground surface. The robot may determine a position of its center of mass and center of mass velocity, and based on these, determine a capture point for the robot. The robot may also determine a threshold position for the capture point, where the threshold position is based on a target trajectory for the capture point after the second foot contacts the ground surface. The robot may determine that the capture point has reached this threshold position and based on this determination, and cause the second foot to contact the ground surface.

Abstract: A robot includes a drive system configured to maneuver the robot about an environment and data processing hardware in communication with memory hardware and the drive system. The memory hardware stores instructions that when executed on the data processing hardware cause the data processing hardware to perform operations. The operations include receiving image data of the robot maneuvering in the environment and executing at least one waypoint heuristic. The at least one waypoint heuristic is configured to trigger a waypoint placement on a waypoint map. In response to the at least one waypoint heuristic triggering the waypoint placement, the operations include recording a waypoint on the waypoint map where the waypoint is associated with at least one waypoint edge and includes sensor data obtained by the robot. The at least one waypoint edge includes a pose transform expressing how to move between two waypoints.

Abstract: A method of planning a path for an articulated arm of robot includes generating a directed graph corresponding to a joint space of the articulated arm. The directed graph includes a plurality of nodes each corresponding to a joint pose of the articulated arm. The method also includes generating a planned path from a start node associated with a start pose of the articulated arm to an end node associated with a target pose of the articulated arm. The planned path includes a series of movements along the nodes between the start node and the end node. The method also includes determining when the articulated arm can travel to a subsequent node or the target pose, terminating a movement of the articulated arm towards a target node, and initiating a subsequent movement of the articulated arm to move directly to the target pose or the subsequent node.