Abstract: A mechanical arm calibration system and a mechanical arm calibration method are provided. The method includes: locating a position of an end point of a mechanical arm in a three-dimensional space to calculate an actual motion trajectory of the end point when the mechanical arm is operating; retrieving link parameters of the mechanical arm, randomly generating sets of particles including compensation amounts for the link parameters through particle swarm optimization (PSO), importing the compensation amounts of each of the sets of particles into forward kinematics after addition of the corresponding link parameters, to calculate an adaptive motion trajectory of the end point; calculating position errors between the adaptive motion trajectory and the actual motion trajectory of each of the sets of particles for a fitness value of the PSO to estimate a group best position; and updating the link parameters by the compensation amounts corresponding to the group best position.

Abstract: One variation of a method for deploying a mobile robotic system to scan inventory structures within a store includes: dispatching the mobile robotic system to navigate along inventory structures within the store during a setup cycle; at the mobile robotic system, while navigating along the inventory structures during the setup cycle, capturing a set of wireless connectivity metrics representing connectivity to a first wireless network; assembling the set of wireless connectivity metrics into a wireless connectivity map of the store; estimating a processing duration from start of the scan cycle to transformation of images of the inventory structures, captured by the mobile robotic system, into a stock condition of the store; and dispatching the mobile robotic system to autonomously capture images of the inventory structures within the store during a scan cycle preceding a scheduled restocking period in the store based on the processing duration.

Abstract: A companion robot is disclosed. In some embodiments, the companion robot may include a head having a facemask and a projector configured to project facial images onto the facemask; a facial camera; a microphone configured to receive audio signals from the environment; a speaker configured to output audio signals; and a processor electrically coupled with the projector, the facial camera, the microphone, and the speaker. In some embodiments, the processor may be configured to receive facial images from the facial camera; receive speech input from the microphone; determine an audio output based on the facial images and/or the speech input; determine a facial projection output based the facial images and/or the speech input; output the audio output via the speaker; and project the facial projection output on the facemask via the projector.

Abstract: The method can include performing inference using the system; and can optionally include training the system components. The method functions to automatically interpret flight commands from a stream of air traffic control (ATC) radio communications. The method can additionally or alternatively function to train and/or update a natural language processing system based on ATC communications. Additionally, the method can include or be used in conjunction with collision avoidance and/or directed perception for traffic detection associated therewith.

Abstract: A spray operation method includes obtaining two-dimensional position information of a target area and a three-dimensional model of the target area, and obtaining an operation route according to the two-dimensional position information and the three-dimensional model, where the operation route includes a plurality of waypoints, and at least one of the waypoints is associated with altitude information.

Abstract: For control of a surgical instrument in a surgical robotic system, multiple actuators establish a static pretension by actuating in opposition to each other in torque control. The static pretension reduces or removes the compliance and elasticity, reducing the backlash width. To drive the tool, the actuators are then moved in cooperation with each other in position mode control so that the movement maintains the static pretension while providing precise control.

Abstract: Systems and methods for drone air traffic control include, in an air traffic control system configured to manage Unmanned Aerial Vehicle (UAV) flight in a plurality of geographic regions, wherein the air traffic control system has one or more servers configured to manage each geographic region predetermined based on a geographic boundary, communicating to one or more UAVs via one or more wireless networks, wherein the one or more UAVs are configured to maintain their flight in the plurality of geographic regions based on coverage of or connectivity to the one or more wireless networks; obtaining data related to the one or more UAVs, wherein the data includes flight operational data, flight plan data, and sensor data related to obstructions and other UAVs; analyzing and storing the data for each geographic region; and managing flight of the one or more UAVs in corresponding geographic regions based on the data.

Abstract: Examples provide a system for system for customized item decanting. A robotic picker device is configured to remove a selected item from a selected case in an open configuration on a conveyor device. A decan manager, implemented on a processor, is configured to identify a destination tote in a set of totes for placement of the selected item, The robotic picker device places the selected item into the destination tote and the decant manager analyzes sensor data and item data associated with the selected item to confirm an identification of the selected item placed into the destination tote for inventory update.

Abstract: The delivery management system stores identification information identifying an article to be delivered to a user and storage place information indicating a storage place of the article in a living space of the user while associating the identification information and the storage place information with each other, and reads the identification information attached to the article and extracts the storage place information associated with the identification information when the article arrives at a delivery destination.

Abstract: A vehicle and a system method of operating the vehicle is disclosed. The system includes a monitoring module and a mitigation module operating on a processor. The monitoring module is configured to measure a degradation in an operation parameter of the vehicle, the vehicle operating in a first state based on a first value of a set of adaptive parameters. The mitigation module is configured to determine a threat to the vehicle due to operating the vehicle in the first state with the degradation in the operation parameter and adjust the set of adaptive parameters from the first value to a second value that mitigates the threat to the vehicle, wherein the processor operates the vehicle in a second state based on the second value.

Abstract: A vehicle control device includes a tracking unit estimating a motion of a moving object, a model selection unit selecting a motion model corresponding to a moving object type, an abnormality determination unit determining a presence or absence of an abnormality of the estimation of the motion of the moving object based on the estimated moving object motion and the motion indicated by the motion model, and a control unit. A control mode in which the control unit controls traveling of a host vehicle when the abnormality determination unit determines that the abnormality is present differs from a control mode in which the control unit controls the traveling of the host vehicle when the abnormality determination unit determines that the abnormality is absent.

Abstract: Implementations described herein relate to training and refining robotic control policies using imitation learning techniques. A robotic control policy can be initially trained based on human demonstrations of various robotic tasks. Further, the robotic control policy can be refined based on human interventions while a robot is performing a robotic task. In some implementations, the robotic control policy may determine whether the robot will fail in performance of the robotic task, and prompt a human to intervene in performance of the robotic task. In additional or alternative implementations, a representation of the sequence of actions can be visually rendered for presentation to the human can proactively intervene in performance of the robotic task.

Abstract: A robot control system determines which of a number of discretizations to use to generate discretized representations of robot swept volumes and to generate discretized representations of the environment in which the robot will operate. Obstacle voxels (or boxes) representing the environment and obstacles therein are streamed into the processor and stored in on-chip environment memory. At runtime, the robot control system may dynamically switch between multiple motion planning graphs stored in off-chip or on-chip memory. The dynamically switching between multiple motion planning graphs at runtime enables the robot to perform motion planning at a relatively low cost as characteristics of the robot itself change.

Abstract: An autonomous tray handling robotic system is disclosed. In various embodiments, data indicating a set of output stacks to be assembled is stored, each output stack including an associated set of trays each of a corresponding tray type. Operation of one or more robots is controlled, each robot being configured to grasp, move, and place one or more trays at a time, according to a plan, to iteratively pick trays from source stacks of trays and assemble the set of output stacks, including by building each output stack by successively placing on the output stack trays picked from one or more corresponding source stacks. Each of the robots comprises a robotic arm and a tray handling end effector configured to grasp, move, and place one or more trays without assistance from another robot.

Abstract: In one embodiment, a method includes accessing a target value for a gear system, where the target value includes a target backlash or a target preload, and where the gear system includes a driving gear, a driven gear, a preloading actuator coupled to the driving gear, and a preloading actuator controller, determining a measured value for the gear system, where the measured value includes a measured backlash or a measured preload, determining that an error value between the measured value and the target value exceeds a threshold error, and sending, by the preloading actuator controller, instructions to the preloading actuator to adjust the driving gear in response to determining the error value between the measured value and the target value exceeds the threshold error.

Abstract: Air traffic control systems and methods include communicating with passenger drones via one or more cell towers associated with the one or more wireless networks, wherein the passenger drones each include hardware and antennas adapted to communicate to the one or more cell towers, and wherein each passenger drone has a unique identifier in the air traffic control system; obtaining data associated with flight of each of the passenger drones based on the communicating; and managing the flight of each of the passenger drones based on the obtained data and performance of one or more functions associated with air traffic control, wherein each passenger drone is configured to constrain flight based on coverage of the one or more cell towers such that each passenger drone maintains communication on the one or more wireless networks.

Abstract: A robot collision avoidance motion optimization technique using a distance field constraint function. CAD or sensor data depicting obstacles in a robot workspace are converted to voxels, and a three-dimensional binary matrix of voxel occupancy is created. A corresponding distance map matrix is then computed, where each cell in the distance map matrix contains a distance to a nearest occupied cell. The distance map matrix is used as a constraint function in a motion planning optimization problem, where the optimization problem is convexified and then iteratively solved to yield a robot motion profile which avoids the obstacles and minimizes an objective function such as distance traveled. The distance field optimization technique is quickly computed and has a computation time which is independent of the number of obstacles. The disclosed optimization technique is easy to set up, as it requires no creation of geometry primitives to approximate robot and obstacle shapes.

Abstract: Utilization of user interface inputs, from remote client devices, in controlling robot(s) in an environment. Implementations relate to generating training instances based on object manipulation parameters, defined by instances of user interface input(s), and training machine learning model(s) to predict the object manipulation parameter(s). Those implementations can subsequently utilize the trained machine learning model(s) to reduce a quantity of instances that input(s) from remote client device(s) are solicited in performing a given set of robotic manipulations and/or to reduce the extent of input(s) from remote client device(s) in performing a given set of robotic operations. Implementations are additionally or alternatively related to mitigating idle time of robot(s) through the utilization of vision data that captures object(s), to be manipulated by a robot, prior to the object(s) being transported to a robot workspace within which the robot can reach and manipulate the object.

Abstract: Disclosed are methods, systems, and a non-transitory computer-readable medium for regenerating at least a portion of a flight plan of a vehicle. The method may include generating an adjustment to a speed, an altitude, and/or a heading for one or more locations along a flight path within at least one of a predetermined distance of the vehicle and a predetermined window of time, based on received speed data, altitude data, and flight path data, including a subset of points along each boom footprint included in the flight path data, and a permissible threshold boom value for each of the one or more locations. The method may also include regenerating a portion of a flight plan corresponding to the one or more locations, based on the generated adjustment to the speed, altitude, and/or heading for the one or more locations.

Abstract: Flying lane management systems and methods implemented in an air traffic control system communicatively coupled to one or more passenger drones via one or more wireless networks include initiating communication to the one or more passenger drones at a preflight stage for each, wherein the communication is via one or more cell towers associated with the one or more wireless networks, wherein the plurality of passenger drones each comprise hardware and antennas adapted to communicate to the plurality of cell towers; determining a flying lane for the one or more passenger drones based on a destination, current air traffic in a region under management of the air traffic control system, and based on detected obstructions in the region; and providing the flying lane to the one or more passenger drones are an approval to takeoff and fly along the flying lane.