Abstract: Implementations claimed and described herein provide systems and methods for map routing matching system, wherein the one or more machine-learning models scores link candidates of location data points to determine a most probable route. In one implementation, use a machine-learning model to calculate a probability score for a plurality of link candidates. The machine-learning model scores each link candidates based on weighting set by training data, wherein the weighting associated with at least one of travel time, match distance, heading difference, speed conformity, road curvature, road classification, or travel distance versus time delay. A plurality of total probability scores derived from one or more combinations of probability scores for respective link candidates is calculated, wherein one link candidate is selected for each location data point.

Abstract: Methods, systems, and media comprising; a physical robot in a physical workcell; an onsite execution subsystem that is configured to control the physical robot using a real-time control subsystem; a cloud-based belief world subsystem that is configured to receive and store sensor data captured in the workcell, wherein the onsite execution subsystem is configured to use sensor data stored by the cloud-based belief world subsystem in order to control the robot using the real-time control subsystem.

Abstract: A motion planning system and method for motion planning of an autonomous vehicle are provided, wherein a motion planning system determines a plurality of configuration samples through a configuration space, wherein the configuration samples are determined in predefined sampling steps, wherein a sequence of configuration samples defines a possible trajectory of the vehicle in the configuration space, wherein the plurality of configuration samples in each sampling step, are determined in parallel by using at least one graphics processing unit of the motion planning system, wherein sequences of configuration samples are calculated from all configuration samples by predefined motion criteria, and wherein the calculated sequences of configuration samples are provided as vehicle trajectories for motion of the vehicle.

Abstract: A navigation method includes obtaining an initial navigation route, receiving a route erasing operation, and erasing a target section according to the route erasing operation. The target section includes at least a part of the initial navigation route corresponding to the route erasing operation. The method further includes determining a target navigation route according to the initial navigation route and the target section, and controlling an electronic device to perform navigation according to the target navigation route.

Abstract: Systems and methods for determining a shortest path between nodes are provided. The systems and methods involve the use of a modified multi-level overlay (MLO) algorithm that allows for multiple potential paths to be computed at a rate that allows for more efficient delivery route planning in real-time or otherwise. The MLO data structure is used to generate hub labels that are subsequently used to determine the shortest path. Additionally, the MLO data structure is modified to provide for a more efficient shortest path determination. For example, the modified MLO data structure may improve efficiency at least by optimizing partitioning parameters and performing label intersection, as is described below in additional detail. This optimization may reduce the search space used by the MLO data structure during a query stage when a shortest path associated with a routing determination is performed.

Abstract: Systems for automatic labeling and methods for using automatic labeling systems are disclosed. A system can include a camera, and a controller including one or more processors coupled to one or more memories. The system can acquire an image of a sample object via the camera, and identify first and second tokens depicted in the image or positioned over the image of the sample object. The system can designate a location of the first token as a target label location at which a label is to be applied to the sample object, and designate a location of the second token as an identifying information location that is identifiable to determine an orientation of the sample object.

Abstract: A robot-based assembly control method includes providing a plurality of neural network models comprising at least a reinforcement learning network model and a tensor fusion network model. Training data for the models includes visual data from the vision device, tactile data from the tactile sensor, motion feedback data from the robot and torque feedback data from the robot. An assembly control system based on the robot, which includes a clamping effector at an end of a movable part of the robot, wherein a tactile sensor and a soft rubber pad are provided on an inside clamping portion of the clamping effector from inside to outside is described. Based on fusing the visual information with the external forces of other dimensions indirectly obtained from changes in tactile signals, representation vectors can be used to generate appropriate instructions of robot motion for flexibly adjusting the insertion force to complete the assembly.

Abstract: Techniques for generating a local graph are described herein. A vehicle may receive a destination and generate a preferred path thereto. The vehicle may generate a local graph based on the preferred path which may define a subregion of the environment within which the vehicle may perform dynamic routing operations. For example, the vehicle may determine a local boundary which may define the subregion. Further, the vehicle may identify one or more junction(s) located within the local boundary and identify candidate (or alternative) driving lane(s) exiting such junction(s). The vehicle may generate a lane reference (e.g., optimized trajectory for the vehicle to follow to the destination) for each candidate driving lane. In such cases, the local boundary, the candidate driving lanes, and/or lane reference(s) may be data represented in the local graph. The vehicle may be controlled based on the local graph.

Abstract: An image acquisition device for a work device performing a task on a target. The image acquisition device includes: an imaging device capturing an image of the target; and an imaging tool that includes a model of the work device and is operable by an operator. The imaging device is attachable to and detachable from each of the work device and the imaging tool, such that that a relative position between the imaging device and the model and that between the imaging device and the work device are identical when the imaging device is attached to the model and the work device. The imaging device captures a part of the model and a part of the work device along with the target when the imaging device is respectively attached to the imaging tool and the work device, the part of the model corresponding to the part of the work device.

Abstract: A computer-implemented control system and method for dynamically managing airflow in an air lubrication system of a vessel are disclosed. Real-time sensor data including vessel motion, acceleration, and speed-through-water are processed by a controller executing a machine learning model to predict near-future sea-state conditions. The controller accesses and interpolates computational fluid dynamics (CFD) efficiency profiles to determine target airflow values and generate valve and compressor commands that modulate airflow to multiple outlets beneath the hull. Feedback from hull and pressure sensors is used to stabilize the air layer, while a learning-based prediction engine continuously refines CFD profiles to improve operational efficiency and reduce parasitic load. The system enhances vessel hydrodynamic performance and fuel economy under varying environmental conditions.

Abstract: Techniques for operating a kinematic structure by manual motion of a link coupled to the kinematic structure include a system comprising a first manipulating means and processing means. The processing means is for detecting whether a first means for accessing an internal worksite is mounted to the first manipulating means; detecting an input indicating the system is to be in a set-up mode; and inhibiting, in response to detecting that the first means is mounted to the first manipulating means, transition of the system to the set-up mode.

Abstract: A control device for controlling an industrial machine, the control device comprising: an image operating unit that displays an image captured by a visual sensor on a display screen, and accepts a user operation with respect to the image; and a motion control unit that controls a motion of the industrial machine so as to move a visual field of the visual sensor displayed as an image on the display screen, in accordance with the accepted user operation.

Abstract: A teaching device is used in an industrial machine system in which information is acquired using a sensor and a task is performed with respect to a workpiece. The teaching device includes: an execution history screen display unit for displaying an execution history screen that presents one or more pieces of execution history pertaining to an execution result of a program performing sensor-based processing, and for accepting an operation to select a single piece of execution history; and a modification screen display unit for restoring the selected single piece of execution history to a program and displaying a modification screen for modifying a configuration parameter of the program in which the execution history is restored.

Abstract: A method for controlling an autonomous, mobile robot which is designed to navigate independently in a robot deployment area, using sensors and a map. According to one embodiment, the method comprises detecting obstacles and calculating the position of detected obstacles based on measurement data received by the sensors, and controlling the robot to avoid a collision with a detected obstacle, the map comprising map data that represents at least one virtual blocked region which, during the control of the robot, is taken into account in the same way as an actual, detected obstacle.

Abstract: A robotic cosmetic application device, and techniques for controlling the device, are provided. The device may include sensors that capture real-time data associated with a user's face and robotic arms equipped with makeup applicator tools. The device may receive an indication of a makeup look selected by the user; analyze the real-time data associated with the user's face in order to generate a three-dimensional map associated with the face of the user, identify facial features of the user's face on the three-dimensional map; and control the robotic arms to automatically adjust the makeup applicator tools with respect to the user's face, to apply one or more cosmetic products to the user's facial features in order to achieve the makeup look selected by the user.

Abstract: A method for the safety control, through direct teaching, of a robotised system comprises a learning step, wherein a processing unit determines a relative distance (RD) between at least one link (L) of the robot manipulator and an operator (O) and controls whether the relative distance (RD) of the at least one link (L) exceeds a predefined distance threshold value (TV); wherein the predefined distance threshold value (TV) is equal to or greater than the distance covered by the robot manipulator in the amount of time needed to stop starting from a respective maximum linear speed (VMAX); in case the relative distance (RD) is smaller than the predefined distance threshold value (TV), the method entails stopping the robot.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for robotic control using demonstrations to learn category-level manipulation task. One of the methods includes obtaining a collection of object models for a plurality of different types of objects belonging to a same object category and training a category-level representation in a category-level space from the collection of object models. A category-level trajectory is generated the demonstration data of a demonstration object. For a new object in the object category, a trajectory projection is generated in the category-level space, which is used to cause a robot to perform the robotic manipulation task on the new object.

Abstract: A control device for an electrified vehicle includes: an acquisition unit for acquiring a temperature of a motor which is a driving power source of an electrified vehicle and an index value indicating ease of temperature rise of the motor; an execution unit that executes a notification process for notifying a driver of electrified vehicle of a warning when a temperature of the motor is equal to or higher than a first temperature and lower than a second temperature; a limitation unit that limits a torque of the motor when a temperature of the motor is equal to or higher than the second temperature; and a setting unit that sets the first temperature to a lower temperature than a temperature when the index value is equal to or higher than a threshold value when the index value is equal to or higher than the threshold value.

Abstract: A navigation system includes an evaluation unit for evaluating a degree of a health state quality for a driver of a moving body, and a route guidance unit for searching a route from a current position of the moving body to a destination, and for performing guidance of the searched route, the route guidance unit determines a search condition of a route to the destination based on an evaluation result of a health state of the driver in the evaluation unit.

Abstract: In a voice output device, the storage unit stores priority predetermined for each type of marks used in voice guidance at guidance points. The guidance voice generation unit determines the mark to be used for voice guidance at the guidance point based on the priority, for each of the guidance points included in a guide route of a mobile body, and generates guidance voice using the determined mark. The voice output unit outputs the generated guidance voice at an utterance point where the guidance voice for the guidance point is to be uttered.