Abstract: A gimbal control method includes obtaining a measured working parameter of a gimbal, where the gimbal includes an axial arm and a motor configured to drive the axial arm to rotate. The method further includes in response to the measured working parameter satisfying a preset condition: determining that an external force is applied to the gimbal, the external force being not applied by a controller; and stopping the gimbal to operate based on a first working parameter, the first working parameter being obtained when the gimbal is controlled by the controller without applying the external force.

Abstract: Technologies for monitoring input feature health for machine learning models and detecting anomalies are described. Embodiments include receiving, by a trained machine learning model and from an online system, a set of historical values and a current feature value of a time series feature, the time series feature represents one or more measured values of the online system. Embodiments include predicting an expected feature value and an expected range of values using the set of historical values. Embodiments include receiving the expected feature value and the expected range of values. Embodiments include receiving the current feature value. Embodiments include determining that an anomaly condition is present based on the current feature value, the expected feature value, and the expected range of values. Embodiments include generating an alert for the anomaly condition that includes the severity metric and the duration metric.

Abstract: The disclosed embodiments provide a system for streamlining machine learning. During operation, the system determines a resource overhead for a baseline version of a machine learning model that uses a set of features to produce entity rankings and a number of features to be removed to lower the resource overhead to a target resource overhead. Next, the system calculates importance scores for the features, wherein each importance score represents an impact of a corresponding feature on the entity rankings. The system then identifies a first subset of the features to be removed as the number of features with lowest importance scores and trains a simplified version of the machine learning model using a second subset of the features that excludes the first subset of the features. Finally, the system executes the simplified version to produce new entity rankings.

Abstract: Embodiments of the disclosed technologies receive, for a first machine learning task, a first set of raw features arranged according to a first schema, and, for a second machine learning task, a second set of raw features arranged according to a second schema different than the first schema. A multi-task raw feature set is created by storing, in a data store arranged according to a common schema, the first and second sets of raw features. A common feature that is common to both the first and second sets of raw features is identified. A multi-task transformed feature set and a model bundle are created. The multi-task transformed feature set is separated into first and second sets of transformed features. The first and second sets of transformed features and the model bundle can be used to create a trained multi-task machine learning model.

Abstract: An optical sensor packaging system and method can include: providing a substrate, the substrate including a redistribution pad; mounting an optical sensor to the substrate, the optical sensor including a photo sensitive material formed on a photo sensitive area of an active optical side of the optical sensor; wire-bonding the optical sensor to the substrate with a first bond wire connected from the active optical side to the redistribution pad; and encapsulating the optical sensor, the first bond wire, and the photo sensitive material with an over-mold, the over-mold formed with a top surface co-planar to a surface of the photo sensitive material, the over-mold forming a vertically extended border around the photo sensitive material and around the photo sensitive area, and the over-mold formed above the first bond wire.

Abstract: The disclosed embodiments provide a system for streamlining machine learning. During operation, the system determines a resource overhead for a baseline version of a machine learning model that uses a set of features to produce entity rankings and a number of features to be removed to lower the resource overhead to a target resource overhead. Next, the system calculates importance scores for the features, wherein each importance score represents an impact of a corresponding feature on the entity rankings. The system then identifies a first subset of the features to be removed as the number of features with lowest importance scores and trains a simplified version of the machine learning model using a second subset of the features that excludes the first subset of the features. Finally, the system executes the simplified version to produce new entity rankings.

Abstract: A giant six-legged polar research vehicle with tracked feet, including a platform, six legs arranged at six ends of the platform and six tracked feet arranged below the six legs. A monitoring device is arranged on a top cover. Six power compartments each having a steering device are arranged at six ends of a chassis in the platform. Each leg includes a main traveling device with an upper end and a lower end respectively connected to the steering device and a tracked foot, an auxiliary traveling device with an upper end and a lower end respectively connected to the chassis and the main traveling device, and a connecting device arranged on the main traveling device. The tracked foot includes a main flipping mechanism, an auxiliary flipping mechanism, a tracked foot slewing device, a crawler, a sliding plate and a suspension.

Abstract: An optical sensor packaging system and method can include: providing a substrate, the substrate including a redistribution pad; mounting an optical sensor to the substrate, the optical sensor including a photo sensitive material formed on a photo sensitive area of an active optical side of the optical sensor; wire-bonding the optical sensor to the substrate with a first bond wire connected from the active optical side to the redistribution pad; and encapsulating the optical sensor, the first bond wire, and the photo sensitive material with an over-mold, the over-mold formed with a top surface co-planar to a surface of the photo sensitive material, the over-mold forming a vertically extended border around the photo sensitive material and around the photo sensitive area, and the over-mold formed above the first bond wire.

Abstract: An optical sensor packaging system and method can include: providing an embedded substrate, the embedded substrate including an embedded chip coupled to a redistribution pad with a redistribution line connecting therebetween; mounting an optical sensor to the embedded substrate, the optical sensor including a photo sensitive material formed on a photo sensitive area of an active optical side of the optical sensor; wire-bonding the optical sensor to the embedded substrate with a first bond wire connected from the active optical side to the redistribution pad; and encapsulating the optical sensor, the first bond wire, and the photo sensitive material with an over-mold, the over-mold formed with a top surface co-planar to a surface of the photo sensitive material, the over-mold forming a vertically extended boarder around the photo sensitive material and around the optical sensing area, and the over-mold formed above the first bond wire.

Abstract: A giant six-legged polar research vehicle with tracked feet, including a platform, six legs arranged at six ends of the platform and six tracked feet arranged below the six legs. A monitoring device is arranged on a top cover. Six power compartments each having a steering device are arranged at six ends of a chassis in the platform. Each leg includes a main traveling device with an upper end and a lower end respectively connected to the steering device and a tracked foot, an auxiliary traveling device with an upper end and a lower end respectively connected to the chassis and the main traveling device, and a connecting device arranged on the main traveling device. The tracked foot includes a main flipping mechanism, an auxiliary flipping mechanism, a tracked foot slewing device, a crawler, a sliding plate and a suspension.

Abstract: A giant gully-crossing vehicle for polar scientific expeditions, including a gully-crossing bridge and a vehicle body. The vehicle body includes a stage and a travelling portion. The stage includes a top platform, a front bottom plate, a rear bottom plate and a multifunctional lock platform assembly. The front bottom plate and the rear bottom plate are capable of sliding independently relative to the top platform. The travelling portion includes a first travelling component, a second travelling component, a third travelling component, a fourth travelling component, a fifth travelling component and a sixth travelling component which are symmetrically provided at left and right sides of the stage and below the stage. Bridge bodies of the bridge components can be assembled to two gully-crossing bridges, and are connected to the multifunctional lock platform assembly when crossing gullies.

Abstract: An optical sensor packaging system and method can include: providing an embedded substrate, the embedded substrate including an embedded chip coupled to a redistribution pad with a redistribution line connecting therebetween; mounting an optical sensor to the embedded substrate, the optical sensor including a photo sensitive material formed on a photo sensitive area of an active optical side of the optical sensor; wire-bonding the optical sensor to the embedded substrate with a first bond wire connected from the active optical side to the redistribution pad; and encapsulating the optical sensor, the first bond wire, and the photo sensitive material with an over-mold, the over-mold formed with a top surface co-planar to a surface of the photo sensitive material, the over-mold forming a vertically extended boarder around the photo sensitive material and around the optical sensing area, and the over-mold formed above the first bond wire.

Abstract: The present invention belongs to the field of laser technology, particularly relates to a semiconductor laser, including a substrate, and lasers, fast axis collimation components, slow axis collimation components, steering compression optical systems, a polarization beam combination prism, a focusing lens and an optical fiber provided on the substrate, wherein the lasers can be arranged in two rows or one row. And lasers of the same row are all located in a same plane. Each laser is sequentially provided with a fast axis collimation component and a slow axis collimation component in the direction of an optical path. The lasers of the same row correspond to a group of steering compression optical systems used to steer and compress the light beams collimated by the fast axis collimation components and the slow axis collimation components. The polarization beam combination prism is used for combining two beams of lasers having been steered and compressed by two groups of the steering compression optical systems.

Abstract: The disclosure provides a method of restoring a bright dot of an organic electroluminescent device, an organic electroluminescent device and a manufacturing method thereof, and a display apparatus. The method of restoring the bright dot of the organic electroluminescent device comprises blocking transmission of holes or electrons in the sub-pixel unit corresponding to the bright dot. By blocking transmission of holes or electrons in the sub-pixel unit corresponding to the bright dot, it is possible to block a flow of current towards an organic functional layer in the sub-pixel unit corresponding to the bright dot such that the sub-pixel unit will not emit light, thus converting the bright dot into a dark dot.

Abstract: The present invention belongs to the field of laser technology, particularly relates to a semiconductor laser, including a substrate, and lasers, fast axis collimation components, slow axis collimation components, steering compression optical systems, a polarization beam combination prism, a focusing lens and an optical fiber provided on the substrate, wherein the lasers can be arranged in two rows or one row. And lasers of the same row are all located in a same plane. Each laser is sequentially provided with a fast axis collimation component and a slow axis collimation component in the direction of an optical path. The lasers of the same row correspond to a group of steering compression optical systems used to steer and compress the light beams collimated by the fast axis collimation components and the slow axis collimation components. The polarization beam combination prism is used for combining two beams of lasers having been steered and compressed by two groups of the steering compression optical systems.

Abstract: Wafer-level package devices are described that include multiple die packaged into a single wafer-level package device. In an implementation, a wafer-level package device includes a semiconductor device having at least one electrical interconnection formed therein. At least one semiconductor package device is positioned over the first surface of the semiconductor device. The semiconductor package device includes one or more micro-solder bumps. The wafer-level package device further includes an encapsulation structure disposed over and supported by the semiconductor device for encapsulating the semiconductor package device(s). When the semiconductor package device is positioned over the semiconductor device, each micro-solder bump is connected to a respective electrical interconnection that is formed in the semiconductor device.

Abstract: Nanorods assemblies that have lengths in excess of 50 microns to meters are formed from contacting rice-shaped colloidal superparticles that are aligned along the long axis of the colloidal superparticles. The rice-shaped colloidal superparticles are formed from a multiplicity of nanorods with a high degree of association that is end to end to form colloidal superparticles that are in excess of three microns in length and have a length to diameter ratio of about three or more. Methods of preparing the rice-shaped colloidal superparticles employ mixing with an additional ligand to the nanorods to bias the self assembly of the nanorods by solvophobic interactions. Methods of preparing the nanorods assemblies include the infusion of the rice-shaped colloidal superparticles into microchannels patterned on a substrate, wherein the rice-shaped colloidal superparticles' long axes align in the microchannels.

Abstract: A fuel cell system may have at least one sensor including a pair of electrodes disposed on a substrate. The sensor may be configured to produce an output signal having a magnitude that is proportional to a relative humidity in a vicinity of the sensor and, if liquid water is on the sensor, proportional to an amount of the liquid water on the sensor.

Abstract: A WLP device is provided with a flange shaped UBM or an embedded partial solder ball UBM on top of a copper post style circuit connection.

Abstract: Nanorods assemblies that have lengths in excess of 50 microns to meters are formed from contacting rice-shaped colloidal superparticles that are aligned along the long axis of the colloidal superparticles. The rice-shaped colloidal superparticles are formed from a multiplicity of nanorods with a high degree of association that is end to end to form colloidal superparticles that are in excess of three microns in length and have a length to diameter ratio of about three or more. Methods of preparing the rice-shaped colloidal superparticles employ mixing with an additional ligand to the nanorods to bias the self assembly of the nanorods by solvophobic interactions. Methods of preparing the nanorods assemblies include the infusion of the rice-shaped colloidal superparticles into microchannels patterned on a substrate, wherein the rice-shaped colloidal superparticles' long axes align in the microchannels.