Abstract: According to an embodiment, a picking system having a holder, a controller, and a sensor is provided. The holder holds the object. The controller controls a motion of the holder. The sensor acquires information of the object. A failure detector of the controller detects a failure in a generation of the holding operation plan or a failure in a holding operation on the object by the holder. A determiner of the controller performs a first retry determination of deciding an operation for a retry in a case in which the failure detector detects the failure in the generation of the holding operation plan, and performing a second retry determination of deciding an operation for a retry in a case in which the failure detector detects the failure in a pickup of the object by the holder, the second retry determination being different from the first retry determination.

Abstract: A method for performing robot dynamic path modification using a remote tool center point (RTCP) coordinate frame, for applications such as material dispensing. A processing tool is fixedly mounted in a workcell and the robot holds and moves the workpiece during the processing. The RTCP coordinate frame is defined at the tip of the tool. A nominal processing path on the workpiece is defined, and a prescribed offset distance from the tip of the tool to the workpiece is defined. A sensor measures the actual offset distance from the tool tip to the workpiece. A controller applies the offset in the RTCP coordinate frame and calculates robot motions causing the robot to move the workpiece nominal path past the tool tip at the offset distance. The controller converts the offset from the RTCP frame to a workcell frame, and performs inverse kinematics calculations to generate robot joint motion commands.

Abstract: A disclosed processing system includes a substrate placing stage, at least one measurer, and a determiner. The substrate placing stage is configured to place a substrate transferred by a transfer device thereon. The at least one measurer measures a load applied to the substrate placing stage from the substrate when the substrate is transferred from the transfer device onto the substrate placing stage. The determiner determines a suitability of a fixed state of the substrate to the transfer device based on the load measured by the at least one measurer.

Abstract: A navigation system is disclosed. The system may include a transceiver and a processor. The transceiver may be configured to receive a destination location, a destination location map and information associated with a site of interest in the destination location. The processor may determine an optimal drop-off location, from a plurality of drop-off locations, at the destination location based on the information associated with the site of interest and the destination location map. The processor may further determine a real-time location of a vehicle or a user device and generate navigation instructions to navigate the vehicle or the user device from the real-time location to the optimal drop-off location. The processor may further transmit the navigation instructions to the vehicle or the user device.

Abstract: The various embodiments described herein generally relate to systems and methods for operating one or more self-driving vehicles. In some embodiments, the self-driving vehicles may include a vehicle processor being operable to: control the vehicle to navigate an operating environment in an initial vehicle navigation mode; monitor for one or more trigger conditions indicating a possible change for the vehicle navigation mode; detect a trigger condition; determine a prospective vehicle navigation mode associated with the detected trigger condition; determine whether to change from the initial vehicle navigation mode to the prospective vehicle navigation mode; and in response to determining to change from the initial vehicle navigation mode to the prospective vehicle navigation mode, adjust one or more vehicle attributes corresponding to the prospective vehicle navigation mode, otherwise continue to operate the vehicle in the initial vehicle navigation mode.

Abstract: Presented herein are systems and methods for performing federated learning across vehicles. A computing device having one or more processors coupled with memory can maintain, on the memory, a first machine learning (ML) comprising a first plurality of parameters for determining values identifying a characteristic of a vehicle function on at least one of a plurality of vehicles. The computing device can receive a second plurality of parameters generated by a second ML used by each respective vehicle. The computing device can update the first plurality of parameters of the first ML in accordance with the second plurality of parameters received from each respective vehicle of the plurality of vehicles. The computing device can transmit, to a vehicle, the updated first plurality of parameters to update the second plurality of parameters of the second ML on the vehicle.

Abstract: Provided is a command value correction device that can reduce positioning error for a robot. A command value correction device according to one embodiment of the present invention corrects a command value for directing the orientation of a articulated robot that positions a tip end of an arm that has a plurality of joints, the device comprising: a robot model setting unit that sets a robot model that represents the articulated robot with an elastically deformable model; a support model setting unit that sets a support model that represents a support, to which the articulated robot is secured, with an elastically deformable model; a force calculation unit that calculates a force that acts on the support by the weight of the articulated robot if the orientation of the articulated robot follows the command value prior to correction; and a correction unit that corrects the command value so as to cancel out elastic deformation of the support model due to the force calculated by the force calculation unit.

Abstract: This robot control device capable of controlling a robot according to a lead-through control in which the orientation of the robot is varied in accordance with the magnitude and the direction of forces detected by a force sensor provided in the robot, wherein the robot control device: identifies a first state in which an additional device is installed at a distal end of a wrist of the robot, and a second state in which the additional device is not installed at said location; and switches a load setting of the robot when the lead-through control is executed in the first and second states.

Abstract: An apparatus includes a bendable body including at least a distal bending portion and a proximal bending portion, a first driving mechanism for bending the distal bending portion, a second driving mechanism for bending the proximal bending portion, a controller that controls the first driving mechanism to bend the distal bending portion and controls the second driving mechanism to bend the proximal bending portion, and a receiving unit that receives a command on an amount of bending of part or all of the bending portions. The controller is configured to switch between first control for bending only the distal bending portion according to the command received by the receiving unit and second control for bending the distal bending portion and the proximal bending portion so that the distal bending portion and the proximal bending portion have constant curvature according to the command received by the receiving unit.

Abstract: A diagnostic system for diagnosing a motor powered mechanism on board a vehicle includes a current detector for detecting current drawn by a motor of the motor powered mechanism, a hall effect sensor operatively coupled to an output of the motor for generating a hall effect position signal, and a controller configured to receive the detected motor current and the hall effect position signal and to determine an obstruction or fault condition associated with the motor powered mechanism based on the detected motor current and hall effect sensor signal.

Abstract: While operating a host vehicle in a lane of a road, a target vehicle operating in the lane in front of the host vehicle is represented with a virtual vehicle. Upon receiving, from an infrastructure element, operation data describing vehicle operation for the road, acceleration of the target vehicle is predicted based on the operation data. An adjustment to a speed for the virtual vehicle is determined based on the acceleration. The virtual vehicle is updated based on the adjustment to the speed. Upon determining that the virtual vehicle is between the target vehicle and the host vehicle, the host vehicle is operated based on the virtual vehicle.

Abstract: The present disclosure discloses a system and a method for controlling a robot to perform a task. The robot is a legged robot that comprises a trunk and a plurality of legs for supporting the trunk above a ground. The system is configured to collect a reference trajectory defining a sequence of states of the robot to perform the task and execute a stance controller configured to compute and apply reaction forces of the plurality of legs of the robot in contact with the ground to push the trunk based on the reference trajectory. The system is further configured to execute a swing controller of a plurality of swing controllers to swing a leg in response to detecting a relative position between the trunk and a foot of the leg of the robot indicative of an imbalance of a support of the trunk of the robot above the ground.

Abstract: A surgical robot (10) includes robot arms (130), an arm base (120), and a control device (30). Each of the robot arms includes a base portion (138), a tip portion (137) that can hold a medical instrument (150), and links (131 to 136). The link (131) adjacent to the base portion is connected to the base portion through a rotational joint (JA1). The control device controls the robot arm having at least seven degrees of freedom among the robot arms such that when viewed from a direction parallel to an axial direction of a rotation axis of the rotational joint, a first portion (134aa) of a first link (134) is located between a second portion (138a) of the base portion and a third portion (JA7a) of the tip portion.

Abstract: A control method by a computer, the control method acquires work information including information on work performed by a robot arm and information on an article as a target of the work; causes the robot arm to perform an operation of lifting up a predetermined unit of articles based on the work information; determines that there is a possibility that a plurality of articles are packaged together based on sensor data obtained when the operation is being performed; and when it is determined that there is a possibility that a plurality of articles are packaged together, settles a handling method for the article according to the sensor data.

Abstract: A robot comprises a memory, a processor, a body and a drive system which are coupled. The drive system comprises one of auger-based surface interface portions and continuous tread surface interface portions. The auger-based surface interface portions and the continuous tread surface interface portions are interchangeable to adapt the robot to one of different operating conditions and different uses. The processor is configured to: control movement of the robot, via the drive system, to traverse across a first surface, wherein the first surface comprises piled granular material, in response to the drive system being configured with the auger-based surface interface portions; and control movement of the robot via the drive system to traverse across a second surface, which is a solid or semi-solid surface other than the piled granular material, in response to the drive system being configured with the continuous tread surface interface portions.

Abstract: In one aspect, a robot control method includes: during the process of a robot advancing from a target elevator-waiting place to a target elevator-taking place in an elevator car, acquiring the distance between the robot and an obstacle; and if the distance is greater than a preset distance threshold value, controlling the robot to advance to the target elevator-taking place at a preset speed until the robot enters the elevator car.

Abstract: Various aspects of an object weight estimation system are described, to enable weight estimation of an object to be transferred from an object source (e.g., a human or an ambulatory robot) that holds the object using one or more arms. One technique includes: tracking an object to be transferred from the object source to a robot; identifying, from at least one image captured by the robot, a posture of the object source; identifying respective segments of the object source from the posture of the object source; determining an estimated weight of the object based on a center of gravity of the object source, as the center of gravity of the object source is calculated in real-time from a balance of the respective segments; and causing a responsive action with the robot, based on the estimated weight of the object.

Abstract: The disclosed technology is generally directed to a method for controlling a satellite. In one example of the technology, the method may include receiving a set of parameters associated with an orientation of the satellite and executing a reinforcement learning model to control the orientation of the satellite to stabilize the satellite based on each of the set of parameters. The executing of the reinforcement learning model includes predicting a set of actions to be executed by the satellite to control the orientation of the satellite based on each of the set of parameters, executing the set of actions, and predicting an outcome of the executing of the set of actions to generate a set of rewards. The method may further include controlling the orientation of the satellite to stabilize the satellite based on the set of actions and the set of rewards.

Abstract: A quadrupedal walking robot includes: a main body including a seat portion on which a person is to sit astride; a handle to be gripped by the person seated on the seat portion; four leg portions coupled to the main body and configured to perform a bending operation, each of the four leg portions including two or more joints; a plurality of actuators configured to drive the plurality of joints; an operation device disposed on the handle and configured to receive an input of a command related to an operation of the quadrupedal walking robot; and a controller configured to control the plurality of actuators in accordance with the command received from the operation device.

Abstract: For view-dependent control, different desired imaging catheter views are defined. A robotic catheter system may alter from one view to another view based on activation but otherwise hands-free by the operator. This pre-definition of views and swapping views intra-operatively provides procedure monitoring information while avoiding radiation and minimizing interruption of the procedure (e.g., needle placement and puncture).