Abstract: A method to train a model for feature point detection involves obtaining a first image and a second image. The method involves generating a first score map for the first image and a second score map for the second image using the model. The method involves selecting a first plurality of interest points in the first image based on the first score map. The method involves selecting a second plurality of interest points in the second image based on the second score map. Pairwise matching of a first interest point among the first plurality of interest points with a second interest point among the second plurality of interest points is performed. Correctness of the pairwise matches is checked based on a ground truth, to generate a reward map. The score map and the reward map are compared and used to update the model.

Abstract: A suspension assembly for a cycle includes a plurality of links pivotably connected to one another by a plurality of pivots. A rotation sensor measures an angular relationship between two links and a setpoint of the suspension assembly is adjusted based on the angular measurements.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for robotics planning. One of the methods comprises receiving data defining multiple skills to be performed by one or more robots in an operating environment; invoking a projection function implemented by each skill, wherein the projection function generates a skill footprint specifying resources requested for performing the skill and a volume occupied by a corresponding entity used to perform the skill; determining that the skill footprints generated by the projection functions are in conflict; and in response, executing the skills in sequence.

Abstract: A method includes: detecting, through a sensor, a location and a movement direction of the user and an object around the user; specifying the type and the location of the detected object; if the object is a dangerous object and is located in the movement direction of the user, setting a relative position where the robot is to be located relative to the user to a lead position ahead of the user and in a direction different from the movement direction; driving at least one pair of legs or wheels of the robot to cause the robot to move to the lead position; and driving the at least one pair of legs or wheels to cause the robot to accompany the user in the lead position and induce a change in the movement direction of the user.

Abstract: Embodiments provide systems, method, and computer-readable storage media for performing robotic tasks in a distributed and coordinated manner. A fleet of robots may use a sequence of messages to appoint supervisors for a set of tasks, where the supervisor robots are responsible for ensuring that their supervised tasks are completed by other robots of the fleet. The supervisors may solicit requests from other robots to perform available tasks and select a robot to perform an available task. Once a robot is appointed, the supervisor and the worker may use messaging sequences to monitor the status of the task and participating robots (e.g., the supervisor and the worker). The monitoring may enable the supervisor to detect a failed worker and enable other robots to detect a failed supervisor.

Abstract: A device and a method for controlling a robotic device are described. The method includes: carrying out a sequence of actions by the robotic device using a robot control model; ascertaining an updated policy using the carried-out sequence of actions; projecting the updated policy onto a projected policy in such a way that for each state of the plurality of states of the projected policy: a similarity value according to a similarity metric between the projected policy and the updated policy is maximized, and a similarity value according to the similarity metric between the projected policy and the initial policy is greater than a predefined threshold value; adapting the robot control model for implementing the projected policy; and controlling the robotic device, using the adapted robot control model.

Abstract: This invention provides a system and method for allowing motion of a robotic manipulator on an AV truck in connecting to a native gladhand on a trailer front that represents and constructs a model of this free space on-the-fly, in the manner of an Obstacle Detection and Obstacle Avoidance (OD/OA) system and process. A robotic arm on an AV truck is adapted to connect a pneumatic line to a gladhand on the trailer front. A first 3D sensor generates a pointcloud, and is located at an elevated position on the truck to image the trailer front. A second 3D sensor also generates pointclouds at during robot motion, located adjacent to an end of the robotic arm. An occlusion mapping process generates an occlusion map of the trailer front, and a map update process updates the occlusion map to add and remove voxels therefrom to allow safe guidance of the robot.

Abstract: Disclosed is a robot cleaner including a light source for irradiating light, a sensor for sensing that the light irradiated from the light source is reflected, and a controller that processes an image using the light sensed by the sensor to calculate a distance value of an individual location of the corresponding image, wherein it is determined that there is an obstacle when there is a dead zone in the image processed by the controller.

Abstract: The present invention relates to a method for determining a touchdown position of a robot performed by a computing device, the method comprising the steps of: detecting a disturbance during swing of a first leg of the robot toward a first touchdown position, calculating a second touchdown position based on a dynamic state of the robot in response to the detection of the disturbance, identifying a step obstacle region based on terrain information of the robot and determining whether the second touchdown position belongs to the step obstacle region, and determining a third touchdown position out of the step obstacle region as the touchdown position of the first leg, in a case in which the second touchdown position belongs to the step obstacle region.

Abstract: Provided is a cleaning device including a suction port; at least one sensor configured to sense at least one object; a driver configured to open or close the suction port; a memory storing at least one instruction; and a processor configured to execute the at least one instruction stored in the memory. The processor executes the at least one instruction to sense at least one object within an area to be cleaned by the cleaning device by controlling the at least one sensor, identify a relative location of the sensed at least one object with respect to the cleaning device, determine at least a partial area within an entire area of the suction port as an open/close target area, based on the identified relative location of the at least one object, and open or close the open/close target area by controlling the driver.

Abstract: An integrated navigation system is configured to support cooperation of a plurality of different robot systems each including at least one automatic work robot operating at a production site. The integrated navigation system includes a job generation device configured to generate a job related to production based on information sent from a plurality of production facilities deployed at the production site, and a navigation device configured to generate a task, which is a work command to each of the plurality of different robot systems, based on the job, and to send the task of the robot system to each corresponding robot system.

Abstract: The disclosure generally pertains to systems and methods for use of a vehicle to conduct a site survey. An example method can include receiving navigation guidance to move a vehicle along a pre-defined path on a property for executing a site survey operation of the property. The site survey operation can include a staking procedure performed by a robotic arm attached to the vehicle, an image capture procedure performed by use of a camera in the vehicle, and/or a subterranean scanning procedure for detecting a buried object. In an example implementation, the staking procedure can include inserting at a first location on the property, a stake having affixed thereon, a barcode label. A barcode scanner can be used to read the barcode label for obtaining information associated with an object present at the first location.

Abstract: Modular, propelled cable-driven robotic platform systems and methods of operation are disclosed. The system includes a robotic platform suspended by a system of overhead cables, motorized cable reels and pulleys. The robotic platform is configured to be equipped with tool modules for performing respective tasks. Additionally, the tool modules each have a multirotor propulsion system. A master control computer coordinates operation of the motorized cable and propulsion systems as a function of sensor data captured by navigation sensors on-board the platform so as to maneuver the robotic platform inside an industrial plant. The system is configured to maneuver around pipings and avoid obstacles in the plant in order to maximize the effective workspace that the robotic platform can reach to perform operations including inspection or repair.

Abstract: Systems and methods for autonomous sterilization of vehicles includes coded logic or processing instructions to determine (i) when sterilization should occur (or not occur), (ii) where sterilization should occur (or not occur), and/or (iii) how sterilization should occur (e.g., which method to use, how long to conduct, and/or which parameter values/settings to employ). The sterilization system may autonomously determine when it is safe to sterilize an environment/location and/or how such sterilization should be carried out. Autonomous sterilization systems and/or processes described herein may permit facilities and/or services to remain open and/or available at higher rates than current offline processes permit, thereby increasing availability.

Abstract: A charging device and a safety function control circuit and method thereof are provided. When a charging device is not connected to a load, a converted voltage value of a power connection terminal of the charging device is kept to be lower than a safe voltage value so as to maintain a safe mode. The safety function control circuit includes a first control module and a second control module for constant voltage control. The first control module and the second control module perform matching control on a power conversion circuit of the charging device, and in case of a single fault of one of the control modules, the other module is still capable of keeping the converted voltage value to be less than the safe voltage value. Thus, it is ensured that the safe mode stays functional in case of a concurrency of both a single hardware fault and a single firmware fault.

Abstract: A system for generating electrical energy, for instance at an agricultural site. The system includes at least one mobile robot, referred to as “collecting robot” and a plurality of mobile robots, referred to as “producing robots”. Each collect robot includes at least one input port adapted for coupling with an output port of a producing robot and an output port for outputting electrical energy received from each producing robot coupled with the collecting robot. Each producing robot includes a photovoltaic generator, at least one input port adapted for coupling with an output port of another producing robot, and at least one output port for outputting electrical energy generated by the producing robot and received from each other producing robot coupled with the producing robot.

Abstract: A robotic system and method includes at least one robot actively or passively mobile between at least first and second locations, and a first safety configuration being defined at least for said first location. A first data carrier associated to the first safety configuration is located in said first location, and the robot comprises a reader adapted to read the first data carrier when the robot is in said first location.

Abstract: A solar panel cleaning system includes a detection traveling vehicle configured to travel on a solar panel and detect dirt on the solar panel, a map generation part configured to generate a dirt map including a dirt position of the dirt on the solar panel, a cleaning path generation part configured to generate a cleaning path for performing cleaning of the solar panel, and a cleaning traveling vehicle configured to travel on the solar panel on the basis of the generated cleaning path and perform cleaning of the solar panel.

Abstract: A controller for an automated machine may include including: one or more processors configured to: determine that a group affiliation of the automated machine switched from a first group of automated machines to a second group of automated machines, the first group of automated machines being assigned to one or more first tasks, the second group of automated machines being assigned to one or more second tasks; generate a message for one or more network devices of the second group of automated machines in accordance with a communication protocol, the message including information about a task performing model of the automated machine, the task performing model being based on a result of performing at least one task of the one or more first tasks by the automated machine.

Abstract: Systems, apparatus, and methods are described for robotic learning and execution of skills. A robotic apparatus can include a memory, a processor, sensors, and one or more movable components (e.g., a manipulating element and/or a transport element). The processor can be operatively coupled to the memory, the movable elements, and the sensors, and configured to obtain information of an environment, including one or more objects located within the environment. In some embodiments, the processor can be configured to learn skills through demonstration, exploration, user inputs, etc. In some embodiments, the processor can be configured to execute skills and/or arbitrate between different behaviors and/or actions. In some embodiments, the processor can be configured to execute skills and/or behaviors using cached trajectories or plans. In some embodiments, the processor can be configured to execute skills requiring navigation and manipulation behaviors.

Abstract: Systems and methods are presented for operating a self-propelled device. In examples, an indication of a surface color on which the self-propelled device operates is received from one or more optical sensors. A color transition from a first color to a second color may be determined based on the received indication. Based on the determined color transition, an activity may be determined. For example, an activity may cause the self-propelled device to move, emit a sound, or illuminate a light, such as an LED. The determined activity may then be performed. In some examples, a hysteresis band may limit the effects of noise and other variations in the color signal. Accordingly, a color transition may occur when color values associated with the surface color indicated are within a first area but not within a second area.

Abstract: The notification system is configured to be capable of communicating with the outside, and is configured to notify the surroundings of the fact that the conveyance work is performed by a third party who has obtained permission when the vehicle is conveyed by the vehicle transportation device capable of autonomous traveling based on a notification instruction from the outside.

Abstract: Method, sterilization system and robotic system for improving duty cycle of Ultraviolet (UV) emitter modules disinfecting closed environment. Sterilization system includes robotic system and plurality of UV emitter modules. Robotic system receives signal corresponding to discharge of battery from one of plurality of UV emitter modules. Each of plurality of UV emitter modules includes UV light source for disinfecting closed environment. Robotic system detects location of UV emitter module amongst plurality of UV emitter modules and maneuvers to location of UV emitter module, mounts UV emitter module and maneuvers UV emitter module to charging dock for charging battery of UV emitter module. After charging, robotic system maneuvers UV emitter module to pre-defined area for disinfecting pre-defined area of closed environment.

Abstract: Implementations set forth herein relate to generating training data, such that each instance of training data includes a corresponding instance of vision data and drivability label(s) for the instance of vision data. A drivability label can be determined using first vision data from a first vision component that is connected to the robot. The drivability label(s) can be generated by processing the first vision data using geometric and/or heuristic methods. Second vision data can be generated using a second vision component of the robot, such as a camera that is connected to the robot. The drivability labels can be correlated to the second vision data and thereafter used to train one or more machine learning models. The trained models can be shared with a robot(s) in furtherance of enabling the robot(s) to determine drivability of areas captured in vision data, which is being collected in real-time using one or more vision components.

Abstract: An actuation mechanism for actuating an electromechanical end effector includes a housing, and a shaft assembly extending distally from the housing. The shaft assembly includes a shaft, a longitudinal knife bar, a first hub, and a second hub. The shaft is axially movable relative to the housing and configured to be coupled to the electromechanical end effector. Rotation of a first screw of the housing moves the first hub to effect longitudinal movement of the shaft. The longitudinal knife bar is axially movable relative to the shaft and configured to be coupled to a knife blade of the electromechanical end effector. Rotation of a second screw of the housing moves the second hub to effect axial movement of the longitudinal knife bar.

Abstract: A method of operating an autonomous cleaning robot is provided. The method includes receiving, at a handheld computing device, data representing a status of a debris collection bin of the autonomous cleaning robot, the status of the bin including a bin fullness reading. The method also includes receiving, at the handheld computing device, data representing a status of a filter bag of an evacuation station, the status of the filter bag including a bag fullness reading. The method also includes presenting, on a display of the handheld computing device, a first status indicator representing the bin fullness reading, and presenting, on the display of the handheld computing device, a second status indicator representing the bag fullness reading.

Abstract: A robot includes a driver; a camera; and a processor configured to: during an interaction session in which a first user identified in an image obtained through the camera is set as an interaction subject, perform an operation corresponding to a user command received from the first user, and determine whether interruption by a second user identified in an image obtained through the camera occurs, and based on determining that the interruption by the second user occurred, control the driver such that the robot performs a feedback motion for the interruption.

Abstract: An information processing device including: an information obtaining unit that obtains environmental information from sensor information obtained by a sensor with which an autonomous moving body is provided; an extraction unit that extracts, from the environmental information, specific environmental information to be saved; and an action control unit that controls an action of the autonomous moving body so as to output the specific environmental information to the outside.

Abstract: Methods, apparatus, and computer-readable media for determining and utilizing corrections to robot actions. Some implementations are directed to updating a local features model of a robot in response to determining a human correction of an action performed by the robot. The local features model is used to determine, based on an embedding generated over a corresponding neural network model, one or more features that are most similar to the generated embedding. Updating the local features model in response to a human correction can include updating a feature embedding, of the local features model, that corresponds to the human correction. Adjustment(s) to the features model can immediately improve robot performance without necessitating retraining of the corresponding neural network model.

Abstract: A method for cleaning the mop of a cleaning robot includes: obtaining mop usage information after the cleaning robot carries the mop to clean the floor, and selecting a target mop cleaning mode to clean the mop according to the mop usage information.

Abstract: A robot lawnmower and method for controlling the robot lawnmower based on the generation of a local map. The method including receiving an image from an imaging sensor onboard the robot lawnmower, the image including an area of ground in an upcoming path, applying a semantic segmentation algorithm to produce a segmented image from the received image, the segmented image including regions corresponding to features in the image, applying a perspective transform to the segmented image to obtain an overhead view transformed image, wherein the regions are preserved in the transformed image, determining, from the transformed image, positions of the regions relative to the current position of the robot lawnmower, plotting a local map of the environment of the robot lawnmower based on positions of the regions relative to a current position of the robot lawnmower; and controlling the robot lawnmower to navigate a lawn area using the local map.

Abstract: A floor element for an automated storage and retrieval system, which comprises a three-dimensional grid, a plurality of container handling vehicles, and multiple floor elements, each floor element arrangeable at a top end of a storage column on top of a vertical stack of containers, includes an upper surface with a substantially rectangular horizontal periphery suitable for being accommodated in a storage column. The floor element further includes lifting frame connecting elements, for releasable connection to a lifting frame, on a peripheral top section of the upper surface and a cut-out at each corner of the substantially rectangular horizontal periphery for interaction with guiding pins arranged on the lifting frame, and rail-connecting elements at the substantially rectangular horizontal periphery.

Abstract: The present disclosure provides a control method for an automatic cleaning equipment, wherein the automatic cleaning equipment includes a mobile device and a self-propelled cleaning device, in which the mobile device is used as a wireless network device to enable the self-propelled cleaning device to communicate with a hotspot function of the mobile device through a wireless network technology, the self-propelled cleaning device is connected to the cloud server through a mobile communication signal of the mobile device, and also the mobile device is connected to the cloud server by using a mobile communication technology.

Abstract: The information processing device 1D mainly includes a state-of-activity estimation unit 31D and a timing determination unit 32D. The state-of-activity estimation unit 31D estimates, based on information detected in a meeting room in which a meeting is being held, a state of activity of the meeting. The timing determination unit 32D determines a timing of mobile sales of a commodity to one or more participants of the meeting based on the state of activity estimated by the state-of-activity estimating unit 31D.

Abstract: The walking robot is an electromechanical device. The walking robot is an automatic device. The walking robot is a mobile device. The walking robot incorporates a superior structure, an inferior structure, and a control circuit. The superior structure attaches to the inferior structure. The control circuit mounts in either the superior structure or the inferior structure. The inferior structure provides the motive forces the move the walking robot. The inferior structure allows the walking robot to walk. The control circuit generates the motive forces and coordinating systems that allow the inferior structure to walk.

Abstract: Described herein are various systems and processes for vehicles for delivery of real-time, on-demand orders for perishable goods, and operations thereof. The automated delivery vehicle may include a food transport container and various sensors configured to determine the location and/or surroundings of the vehicle. The automated delivery vehicle may further include a display and/or other output source configured to provide outputs to persons surrounding the automated delivery vehicle. Such outputs may include, for example, an information graphical representation (e.g., provided on a display), an audio output, and/or data communicated to a wireless device.

Abstract: An information processing device is provided including a recognition unit that executes recognition processing used to determine an action of an autonomous operating body on the basis of sensor information collected, in which the recognition unit includes a feedback recognizer that recognizes feedback from a user on behavior executed by the autonomous operating body, and the feedback recognizer recognizes a degree of the feedback on the basis of recognition results of a contact action and a non-contact action by the user for the autonomous operating body.

Abstract: A method is disclosed for reducing impact forces to legged robots as a result of traversing a terrain. The method includes employing actuators and compliant elements to use the contact of a first portion of a foot assembly with a terrain during a step to reduce the vertical velocity of a subsequent portion of the foot assembly so that the vertical velocity of the subsequent portion as it touches the terrain is substantially zero relative to the terrain. Foot assemblies that employ these methods are also provided.

Abstract: A method includes obtaining, using at least one processor of an electronic device, an image-query understanding model. The method also includes obtaining, using the at least one processor, an image and a user query associated with the image, where the image includes a target image area and the user query includes a target phrase. The method further includes retraining, using the at least one processor, the image-query understanding model using a correlation between the target image area and the target phrase to obtain a retrained image-query understanding model.

Abstract: An example implementation includes (i) receiving sensor data that indicates topographical features of an environment in which a robotic device is operating, (ii) processing the sensor data into a topographical map that includes a two-dimensional matrix of discrete cells, the discrete cells indicating sample heights of respective portions of the environment, (iii) determining, for a first foot of the robotic device, a first step path extending from a first lift-off location to a first touch-down location, (iv) identifying, within the topographical map, a first scan patch of cells that encompass the first step path, (v) determining a first high point among the first scan patch of cells; and (vi) during the first step, directing the robotic device to lift the first foot to a first swing height that is higher than the determined first high point.

Abstract: A dual-manipulator control method is configured to be used in a dual-manipulator control system including a first manipulator, a second manipulator, and a central control module. The first manipulator and the second manipulator are controlled by the central control module, and the central control module is configured to execute the dual-manipulator control method. The dual-manipulator control method includes: generating a first instruction sequence to control the first manipulator and a second instruction sequence to control the second manipulator; and controlling the first manipulator and the second manipulator based on the first instruction sequence and the second instruction sequence. Thus, the working efficiency is improved.

Abstract: A management system for a heterogenous robot fleet provides standard application programmatic interfaces (APIs) for use by inventory management systems in generating complex applications, wherein calls to the API are formatted in a standard format and cause the management system to orchestrate cross-fleet operations that involve multiple different types of robots that are operating in a common facility. Also, the management system for a heterogenous robot fleet provides a unified view of the heterogenous robot fleets of the facility.

Abstract: Systems and methods for transferring, storing, and/or processing materials, such as fuel or propellant, in space, are disclosed. A representative system includes a flexible container that is changeable between a stowed configuration in which the flexible container is contained within a satellite, and a deployed configuration in which the flexible container extends away from the satellite. The system can include a tanker with a storage container to dock with and refuel a satellite. Another representative system includes a controller programmed with instructions that position a spacecraft with a storage container in a first orbit, transfer the spacecraft to a second orbit, dock the spacecraft with a satellite in the second orbit, transfer material between the storage container and the satellite, undock the spacecraft from the satellite, and, optionally, return the spacecraft to the first orbit. An androgynous coupling system with mechanical and fluid connectors facilitates docking and material transfer.

Abstract: A mower is provided and includes: a case; a charging circuit provided in the case, and a combination of a wheel assembly and a charging interface assembly arranged on the case. The combination is electrically connected to the charging circuit. The combination includes: a wheel assembly; and a charging interface assembly electrically connected to the charging circuit. The wheel assembly and the charging interface assembly are integrated molding structure.

Abstract: An autonomous mobile device (AMD) may move around a physical space to perform tasks. The AMD may explore the physical space to determine map data characterizing the physical space. An exploration system maintains a world snapshot that is representative of the AMD with respect to the physical space. A plurality of planning algorithm modules are available, each able to achieve a particular exploration goal or operate under particular conditions such as initial motion, loop closure, frontier exploration, and so forth. One or more planning algorithms are selected and provided with the world snapshot data. These planning algorithms provide responses and may also determine plan data. The resulting plan data is then selected for execution. The execution of the selected plan data operates the AMD to explore the physical space.

Abstract: The present application relates to a multi-sensor-fusion-based autonomous mobile robot indoor and outdoor navigation method and a robot. The method includes: acquiring inertial measurement data and three-dimensional point cloud data of a robot at a current position; determining a pose change of the robot based on the inertial measurement data of the robot at the current position; performing distortion correction on the three-dimensional point cloud data of the robot at the current position based on the pose change of the robot; and matching the three-dimensional point cloud data after the distortion correction with a navigation map, to determine the current position of the robot. With the method, the robot can be accurately positioned.

Abstract: A remote operation spray device configured to depress-operate a spray nozzle of a spray device from a remote location. Device generates adequate pushing force required for depression and, by achieving weight reduction of the device proper, enables easy mounting on an aerial vehicle or other mobile unit. Device includes a main body fixedly attached to a mobile unit, a holder fixedly retaining a spray cartridge and a spray mechanism for jetting spray. The spray mechanism includes a base member, a depressing member pivotably attached to the base member for pressing down a spray nozzle, a pushed member provided at a distal end of the depressing member, and a pusher installed on the base member for pushing the pushed member to move horizontally. The horizontal pushing action of the pusher pushes and moves the pushed member to pivot the depressing member and press the spray nozzle vertically downward.

Abstract: A base attachment module (BAM) of a small aerial vehicle (SAV) for attaching the SAV to an unprepared work surface has a surface meeting docking assembly (DA) with contact points for contacting the work surface, and a contracting suction cup (CSC) in a position surrounded by the contact points such that it extends beyond a plane (or other surface contour) of the points in a fully extended operating position, and is retracted below the plane in a retracted position. Actuation of the CSC when in contact with the surface, allows for the DA to be brought into contact with the contact points in a controlled manner, for a high stiffness attachment, as is needed for deployment of robotic tasks from a SAV.

Abstract: A includes a driver, and a processor configured to: based on an occurrence of an event for performing an operation, acquire target information corresponding to the operation and context information related to the robot, based on at least one of the target information or the context information, select an action plan, acquire action information based on the action plan, and control the driver such that an operation corresponding to the target information is performed based on the action information. The action plan is selected as a first action plan of performing an operation according to first action information stored in advance in the memory, a second action plan of performing an operation according to second action information generated based on the acquired target information and the acquired context information, and a third action plan of performing an operation according to third action information learned based on an operation pattern of the robot.

Abstract: An autonomous working apparatus includes a main body mechanism, a moving mechanism, a working mechanism and a control module. The main body mechanism includes a support body, and the working mechanism is configured to be installed on the support body. The working mechanism has an operating member and a height variable mechanism. The height variable mechanism is configured to be movably connected with the operating member. The operating member is configured to be rotatably connected to the support body, and the operating member is configured to operatably drive the height variable mechanism to move along the height adjustment direction. The force between the operating member and the height variable mechanism is caused by the gravity of the height variable mechanism, and the force between the operating member and the support body is caused by the gravity of said operating member and of said height variable mechanism.