Abstract: In a state estimation method for a legged robot, first sensor information and second sensor information of the legged robot are received. First state information of the legged robot for a period of time is determined, via a first Kalman filter, based on the first sensor information and the second sensor information. Third sensor information of the legged robot is received. Second state information of the legged robot is determined, via a second Kalman filter, based on the third sensor information and the first state information for the period of time. First state information of the legged robot at a current time is updated based on the second state information of the legged robot, to determine state information of the legged robot at the current time.

Abstract: In various examples, a technique for performing spatial mapping and planning with large language models using knowledge graphs may include querying a temporal knowledge graph to predict a next step that a robotic system is to traverse in an environment, wherein the temporal knowledge graph comprises a graph representation of entities and relationships in the environment, wherein the entities and relationships in the graph representation are produced based on sensor data captured from the environment. The technique also may include receiving additional sensor data captured from the environment. The technique further may include generating, via a machine learning model, spatial and temporal data associated with the additional sensor data. The technique still further may include updating, via the machine learning model, the temporal knowledge graph to include representations of the spatial and temporal data, based on a similarity between the spatial and temporal data and the graph representation.

Abstract: Among other things, we describe techniques for estimating a speed profile for a proposed trajectory for a vehicle and operating the vehicle along the proposed trajectory according to the speed profile, including a method for: obtaining, by a planning circuit on a vehicle, a proposed trajectory for the vehicle in response to a driving scenario; obtaining, by the planning circuit, an estimated speed profile, and a confidence score, wherein the confidence score represents a similarity of the estimated speed profile to an actual speed profile that would be generated by a control circuit for the proposed trajectory; determining whether the confidence score meets a confidence threshold; and in accordance with a determination that the confidence score exceeds the confidence threshold, operating, by a control circuit on the vehicle, the vehicle along the proposed trajectory.

Abstract: The present disclosure relates to a robot calibration method based on an equivalent kinematic model and a device thereof. The method includes: calibrating a robot real system with a degree of freedom of f?6 and containing multiple types of time-invariant error sources as a kinematic equivalent system with six degrees of freedom; wherein 6-dimensional pose vectors y of output members of the equivalent system and the real system are the same, and the equivalent system includes a forward position solution model: y=f(?(q,?),?0); constructing a function relationship between ? and ?; acquiring a pose measurement value of the real system; calculating an estimator {circumflex over (?)}i, of the adjustable constant parameter vector ? in a decoupling manner in a joint space; modifying a driving joint variable corresponding to a received motion trajectory instruction set; and driving the robot to move according to the modified driving joint variable.

Abstract: A surgical robotic system includes a robotic arm having an instrument drive unit and an access port configured to be secured to the robotic arm. The system also includes an instrument configured to couple to the instrument drive unit, where the instrument drive unit is configured to actuate the instrument. The system also includes a controller configured to: instruct the robotic arm to advance the instrument into the access port; instruct the instrument drive unit to calibrate the instrument inside the access port; determine a type of the access port during calibration of the instrument; and store the type of the access port in a memory.

Abstract: A covariate shift generally refers to the change of the distribution of the input data (e.g., noise distribution) between the training and inference regimes. Such covariate shifts can degrade the performance grasping neural networks, and thus robotic grasping operations. As described herein, an output of a grasp neural network can be transformed, so as to determine appropriate locations on a given object for a robot or autonomous machine to grasp.

Abstract: A computer-implemented method for retrieving flight data in a medical emergency includes receiving, by one or more processors of a first user device and via an application programming interface, an indication of a medical emergency aboard an aircraft; retrieving, by the one or more processors and responsive to receiving the indication of the medical emergency, flight data associated with the medical emergency from a flight management system associated with the aircraft; automatically categorizing, by the one or more processors, the retrieved flight data based at least on one or more flight data fields associated with the medical emergency to generate categorized flight data; and transmitting, by the one or more processors, at least some of the categorized flight data to a second user device external to the aircraft.

Abstract: An adaptive speed control system and method for adapting control of an electric motor under a changing load condition. A load torque observer determines a load torque value when the motor is operating at a constant speed, and the load torque value is used to adapt a torque. An inertia observer determines an inertia load value when the motor is operating at a changing speed, and the inertia load value is used to adapt a controller gain. An active disturbance input decoupler provides disturbance rejection when the motor is operating at a constant speed. An adaptive control switch switches the load torque observer between driving the inertia observer and driving the active disturbance input decoupler. The system may be configured for a multi-axis system in which multiple motors are each associated with a different axis of motion, and multiple adaptive speed control systems are each associated with a different motor.

Abstract: The present application relates to a pre-stacking device, a pre-stacking method, and a battery processing equipment. The pre-stacking device includes: at least two conveying members, each having at least two bearing positions for placing battery cells; a pre-stacking platform with a stacking station; and a gripping mechanism arranged between the pre-stacking platform and all the conveying members to move the battery cells on each of the conveying members to the stacking station; the gripping mechanism being configured to be capable of gripping corresponding battery cells from each of the conveying members according to a preset battery-cell polarity combination mode.

Abstract: Systems and methods for automated wire pickup using image-based robot guidance. The machine vision-based system includes: a robot arm; a tool head coupled to the distal end of the robot arm and including a wire gripper motor; a wire gripper movably coupled to the tool head and operatively coupled to the wire gripper motor; a wire holder configured for clamping a wire; camera means mounted to the distal end of the robot arm and having first and second fields of view which intersect in a volume of space that includes the tip of the wire gripper and the wire holder; and a computer system configured to control operation of the robot arm motors and wire gripper motor. More specifically, the computer system is configured for visually estimating the position and orientation of a contact-equipped wire being held by the wire holder and then generating robot guidance to enable automated pickup of the end section of the wire by the wire gripper.

Abstract: A position obtaining device includes a driver and a processor. The processor, in response to a velocity of the device being moved by the driver being a predetermined velocity or less, derives a first attitude angle with a first likelihood as an attitude angle of the device; in response to a predetermined number of second light sources or more being captured in an image obtained by a second camera, derives a three-dimensional position of the device and a second attitude angle with a second likelihood lower than the first likelihood as the attitude angle of the device; determines a likelihood of the attitude angle and the three-dimensional position of the device estimated based on a result of the first attitude angle and a result of the three-dimensional position and the second attitude angle; and controls the velocity in accordance with the determined likelihood.

Abstract: An atrial septal defect occluder, including: a body (100, 510), including two disk surfaces arranged opposite to each other and provided with a through hole (130) along an axial direction of the body (100, 510); and a blocking assembly, including an openable choke film (200) or elastic body (520), the blocking assembly being in contact with the body (100, 510), wherein when the blocking assembly is in a closed state, the blocking assembly impedes liquid from flowing in the through hole (130), and when the blocking assembly is in an open state, the blocking effect is weakened or even eliminated, thus allowing a minimally invasive interventional device to pass through the occluder by means of the through hole (130).

Abstract: Disclosed herein are system, method, and computer program product embodiments for determining a precipitation rate. In some embodiments, a processor receives precipitation data from an array of sensors of a set of array of sensors disposed on a vehicle. A set of lidar sensors are disposed on the vehicle and the array of sensors is positioned within a predetermined distance of a respective lidar sensor of the set of lidar sensors. The processor generates a precipitation rate based on an average of the precipitation data received from the array of sensors and trains a model to control a function of the vehicle during precipitation using the precipitation rate.

Abstract: Implementations process, using a large language model, a free-form natural language (NL) instruction to generate to generate LLM output. Those implementations generate, based on the LLM output and a NL skill description of a robotic skill, a task-grounding measure that reflects a probability of the skill description in the probability distribution of the LLM output. Those implementations further generate, based on the robotic skill and current environmental state data, a world-grounding measure that reflects a probability of the robotic skill being successful based on the current environmental state data. Those implementations further determine, based on both the task-grounding measure and the world-grounding measure, whether to implement the robotic skill.

Abstract: An embodiment for executing a driver's expectations in view of contextual data and driver expectation data to improve driver safety. The embodiment may automatically establish a connection between an autonomous vehicle and a brain computer interface (BCI) device worn by a driver of the autonomous vehicle. The embodiment may automatically utilize sensors and the BCI device to gather contextual data, automatically process the contextual data, and conduct predictive analysis of correlated driver behavior. The embodiment may automatically detect a traffic context and generate a predicted driver expectation for the traffic context. The embodiment may, in response to detecting that the predicted driver expectation for the traffic context is safe, automatically execute the predicted driver expectation. The embodiment may, in response to detecting that the predicted driver expectation is unsafe, automatically execute a safety remedial action.

Abstract: Systems and methods of providing traction assistance in a vehicle are disclosed that include determining whether a slip condition exists on a route of the vehicle based on vehicle data, external data, or a combination thereof, determining an electric drive powertrain configuration that includes a minimum state of charge (SoC) for a battery system of the vehicle based on the slip condition, and communicating the electric drive powertrain configuration to an electric drive powertrain controller, which operates the electric drive powertrain in accordance with the electric drive powertrain configuration to provide the minimum SoC to the battery system prior to the vehicle travelling over a portion of the route where the slip condition exists.

Abstract: In some implementations, a method may include obtaining target fault data indicating a preselected fault for a vehicle having a hydrogen source. In addition, the method may include obtaining vehicle fault data indicating a fault of the vehicle. The method may include identifying a match between the fault and the preselected fault. Moreover, the method may include purging, in response to the identifying the match, hydrogen from the hydrogen source so as to reduce consequences of a battery thermal event by reducing the likelihood of a hydrogen-fueled fire.

Abstract: An offline teaching device includes an acquisition unit that acquires position information of a plurality of welding lines of a workpiece and a scanning range of a sensor, a generation unit that generates a plurality of three-dimensional regions to be scanned, and a control unit that disposes the plurality of welding lines and the plurality of three-dimensional regions in a virtual space and highlights an overlapping region of each of the plurality of three-dimensional regions. The overlapping region is deleted, a new auxiliary screen in which the plurality of welding lines and at least one three-dimensional region after the deletion of the overlapping region are disposed in a virtual space is generated, and a teaching program for scanning the at least one three-dimensional region is created and output.

Abstract: An apparatus for inspecting a host vehicle can include a memory configured to store failure diagnosis model information for each of vehicle parts, where the failure diagnosis model information is obtained while a test vehicle equipped with each of the vehicle parts in a failure state is driving, a sensor configured to generate information on driving conditions of a road on which the host vehicle is currently traveling and a status of the host vehicle, a controller configured to compare the information generated by the sensor with the failure diagnosis model information to determine whether a road surface condition is consistent with each other, and determine whether a vehicle part has failed in response that the road surface condition is consistent with each other, and a displayer configured to display a warning sign of a failure according to a control signal provided by the controller.

Abstract: A device for determining the positions of a plurality of recessed portions to be formed on a surface of a workpiece by means of scraping, in which a robot uses a scraper to shave the surface of the workpiece to make the same flat, is provided with: an input accepting unit for accepting input of shape information of the surface and pattern information of the plurality of recessed portions on the surface; and a position determining unit for automatically determining the position of each recessed portion on the surface on the basis of the shape information and the pattern information accepted by the input accepting unit.