Abstract: The present disclosure provides methods and systems of sampling-based object pose determination. An example method includes obtaining, for a time frame, sensor data of the object acquired by a plurality of sensors; generating a two-dimensional bounding box of the object in a projection plane based on the sensor data of the time frame; generating a three-dimensional pose model of the object based on the sensor data of the time frame and a model reconstruction algorithm; generating, based on the sensor data, the pose model, and multiple sampling techniques, a plurality of pose hypotheses of the object corresponding to the time frame, generating a hypothesis projection of the object for each of the pose hypotheses by projecting the pose hypothesis onto the projection plane; determining evaluation results by comparing the hypothesis projections with the bounding box; and determining, based on the evaluation results, an object pose for the time frame.

Abstract: The present disclosure provides an adaptive die testing apparatus and a formation method. The adaptive die testing apparatus includes a support frame capable of moving along a vertical direction; a movable panel capable of moving along a horizontal direction; an upper test socket installed on the support frame; and a lower test socket installed on the movable panel and capable of moving with the movable plate to directly below the upper test socket. A groove is formed on a lower surface of the upper test socket and/or an upper surface of the lower test socket. When the adaptive die testing apparatus is at a testing state, the lower surface of the upper test socket is in a close contact with the upper surface of the lower test socket, and a sealed chamber for placing a die to-be-tested is formed at a region of the groove.

Abstract: A computer-implemented method to generate a second machine learning model based on a first machine learning model, wherein the second machine learning model is structured for more efficient computation, is provided. The method includes processing an input with a hidden layer of a student machine-learned model to obtain an intermediate output. The method includes providing an encoded message descriptive of the input and the intermediate output for processing with a teacher machine-learned model. The method includes, responsive to providing the encoded message, obtaining a second encoded message descriptive of a second intermediate output of one or more hidden layers of the teacher machine-learned model. The method includes performing a knowledge distillation training process to train the student machine-learned model based on a difference between the intermediate output and the second intermediate output.

Abstract: Aspects of this technical solution are directed to multiplexed transmission by optical beam transmission. A method can include generating a first beam having a first frequency in an optical frequency range, generating a second beam including data and having a second frequency in the optical frequency range, converting the first beam and the second beam into a third beam including the data and having a third frequency based on a difference between the first frequency and the second frequency, and transforming the third beam into a fourth beam having a first angular distribution and including a first portion of the data, and into a fifth beam having a second angular distribution corresponding to the first angular distribution and including a second portion of the data.

Abstract: A configuration to address a bus stall during data packet transmission also allows for bus recover due to data packet transmission errors. If a downstream node is not ready to receive data from a buffer of an upstream node, a timer counts a timeout value. The time count increments on each clock cycle in which the downstream node is not ready to receive data. The buffer is cleared at the upstream node when the count reaches a predetermined threshold value. Alternately, the configuration also operates to receive a packet header from an upstream node, the packet header identifying a number of data in a packet. If no data downstream is received on a clock cycle, a counter triggers and is incremented for each cycle in which no date is received. When a threshold is reached dummy data fills a data packet to then transmit the data.

Abstract: A coated optical fiber includes an optical fiber having a core and an outer surface, and a homogeneous polymer coating in contact with the outer surface of the optical fiber. The optical fiber and the homogeneous polymer coating are UV transparent, and a refractive index of the outer surface of the optical fiber or the homogeneous polymer coating is up to 15% less than a refractive index of the core. Coating the optical fiber includes coating an outer surface with a polymerizable material and polymerizing the polymerizerable material to yield the coated optical fiber having a homogeneous polymer coating. The optical fiber and the homogeneous polymer coating are UV transparent, and a refractive index of the outer surface of the optical fiber or the homogeneous polymer coating is up to 15% less than a refractive index of a core of the optical fiber.

Abstract: Disclosed are a method and system for evaluating sonar self-noise at a ship design stage. The method includes: building a ship structure full-scale geometric simulation model; acquiring loss factors and sonar transducer space outfitting acoustic absorption coefficient material parameters; acquiring mechanical excitation, hydrodynamic excitation, and propeller excitation; inputting the loss factors and the sonar transducer space outfitting acoustic absorption coefficient material parameters into an established statistical energy evaluation model, and applying a mechanical excitation to a face plate of foundation of the built ship structure full-scale geometric simulation model, applying a hydrodynamic excitation to the surface of a ship hull, and applying a propeller excitation to a stern shaft to perform calculation of sonar self-noise of a ship to obtain total spectral density level of the sonar self-noise; and evaluating spectral density level calculation results by index requirements.

Abstract: Disclosed are a method and system for evaluating sonar self-noise at a ship design stage. The method includes: building a ship structure full-scale geometric simulation model; acquiring loss factors and sonar transducer space outfitting acoustic absorption coefficient material parameters; acquiring mechanical excitation, hydrodynamic excitation, and propeller excitation; inputting the loss factors and the sonar transducer space outfitting acoustic absorption coefficient material parameters into an established statistical energy evaluation model, and applying a mechanical excitation to a face plate of foundation of the built ship structure full-scale geometric simulation model, applying a hydrodynamic excitation to the surface of a ship hull, and applying a propeller excitation to a stern shaft to perform calculation of sonar self-noise of a ship to obtain total spectral density level of the sonar self-noise; and evaluating spectral density level calculation results by index requirements.

Abstract: Provided are a display panel and a display device. The display panel includes a base substrate as well as, at a side of the base substrate, a plurality of sub-pixels arranged in an array, a power bus, a plurality of first power lines extending along a column direction and a plurality of second power lines extending along a row direction, where each of the plurality of first power lines is electrically connected to the plurality of the sub-pixels arranged along the column direction, and the plurality of first power lines and the plurality of second power lines are electrically connected in overlapping regions of vertical projection on the base substrate. The display panel further includes a plurality of connection units.

Abstract: An electromechanical actuator for forced unlocking of a fuzzy jamming fault of a transmission mechanism comprises a bidirectional electromagnetic braking assembly, a circumferential rotating housing assembly, an axial spline bearing assembly, an LVDT linear displacement sensor assembly, a potentiometer linear displacement sensor assembly, and an electromechanical actuator assembly. The axial spline bearing assembly is mounted inside the circumferential rotating housing assembly for a reciprocating linear motion, and the axial spline bearing assembly is connected to an output rotating shaft of the electromechanical actuator assembly to form a spiral transmission pair; the bidirectional electromagnetic braking assembly is actively energized to generate radial and axial attraction forces, so that the circumferential rotating housing assembly is disengaged from a fixed connection with the electromechanical actuator assembly to rotate.

Abstract: The display panel includes pixel groups including multiple sub-pixels; and driving signal lines. One driving signal line corresponds to at least one pixel group of the pixel groups, and multiple driving signal lines sequentially output a charging enabling level to drive corresponding pixel groups. The sub-pixel includes a first color sub-pixel that includes a first sub-pixel and a second sub-pixel. The first and second sub-pixels are arranged along an arrangement direction of the pixel group, and the second sub-pixel and the first sub-pixel are spaced by other first color sub-pixels whose number is not greater than a preset number. In a first mode, the display panel receives a noise signal, and a phase difference between charging enabling levels provided by the driving signal lines corresponding to the first sub-pixel and the second sub-pixel is non-integer times a noise cycle.

Abstract: Disclosed by the present application are an optical lens and an electronic device comprising the optical lens. The optical lens comprises in succession from an object side to an image side along an optical axis: a first lens provided with a negative focal power, the image side surface of which is concave; a second lens provided with a focal power, the object side surface of which is concave and the image side surface of which is convex; a third lens provided with a positive focal power, the object side surface of which is convex and the image side surface of which is convex; a fourth lens provided with a focal power, the object side surface of which is convex; a fifth lens provided with a focal power, the image side surface of which is convex; and a sixth lens provided with a focal power.

Abstract: A target tracking method and system for use with a surgical robot is disclosed. The method includes: acquiring a visible light image and a depth image of a marker attached on a patient's body surface, where the marker is provided with a black and white checkerboard pattern, and a two-dimensional code is arranged inside squares of the checkerboard; performing two-dimensional code detection on the visible light image to obtain the checkerboard corners' 2D coordinates and the IDs of the two-dimensional codes on the marker; and obtaining 3D coordinates of checkerboard corners in the marker by using the depth image, 2D code corners' coordinates and 2D code ID. According to the 3D coordinates of the checkerboard corner, the position information of the tracked target in the 3D space is obtained.

Abstract: A smart connection sheet includes a conductive connection member and a cut-off apparatus. An end of the conductive connection member is adapted to connect to a terminal of a battery core of a battery pack, and another end of the conductive connection member is adapted to connect to a terminal of an adj acent battery core of the battery pack, to implement a series connection between the two adjacent battery cores in the battery pack. When receiving a cut-off signal, the cut-off apparatus cuts off the conductive connection member, to disconnect the two battery cores connected to the conductive connection member.

Abstract: A surface tracking-based surgical robot system for drilling operations includes surface scanning equipment, a robot arm, and a workstation. To perform a drilling operation, surface information for the target surgery area and the surgical tool held by the robot arm is simultaneously collected with the same surface scanning equipment. A preoperative 3D image of a surgical area and a CAD model of the surgical tool are registered to the collected surface information, through which the relative position between a surgical path and the surgical tool is obtained. Using this relative position, the robot arm aligns to the predefined surgical path using a feedback control method running on a workstation. The drilling operation is then autonomously executed by the robot arm or manually performed under the guidance of the surgical tool held by the robot arm.

Abstract: Embodiments of the present disclosure provide a memory detection method and detection apparatus, for detecting a current-leakage bitline. The method includes: a memory including a plurality of memory cells, a plurality of sense amplifiers, and the sense amplifier including a power line providing a low potential voltage and a power line providing a high potential voltage; writing first memory data to each of the memory cells; performing a reading operation after the first memory data is written; acquiring a first test result based on a difference between first real data and the first memory data; performing the reading operation again to read second real data in each of the memory cells; acquiring a second test result based on a difference between the second real data and second memory data; and acquiring a specific position of the current-leakage bitline based on the second test result and the first test result.

Abstract: An emulation system traverses trace buffers to read data captured from a design under test (DUT). The emulation system receives a request to read at least a portion of DUT data. The emulation system reads a header of the latest sample of the DUT data, where header of each sample of the DUT data includes one or more pointers to a previously stored sample. The samples of the DUT data are partitioned into frames and sectors. The emulation system can identify samples of the DUT data using the pointers in the header of the samples and compare time stamps of the samples against a specified time stamp in the received request. After identifying a sample having the specified time stamp, the emulation system may read the sample for output to the user (e.g., reconstructing a waveform using the sample).

Abstract: Embodiments of the present disclosure provide a memory detection method and detection apparatus, for detecting a current-leakage bitline. The method includes: a memory including a plurality of memory cells, a plurality of sense amplifiers, and the sense amplifier including a power line providing a low potential voltage and a power line providing a high potential voltage; writing first memory data to each of the memory cells; performing a reading operation after the first memory data is written; acquiring a first test result based on a difference between first real data and the first memory data; performing the reading operation again to read second real data in each of the memory cells; acquiring a second test result based on a difference between the second real data and second memory data; and acquiring a specific position of the current-leakage bitline based on the second test result and the first test result.

Abstract: A system including a main neural network for performing one or more machine learning tasks on a network input to generate one or more network outputs. The main neural network includes a Mixture of Experts (MoE) subnetwork that includes a plurality of expert neural networks and a gating subsystem. The gating subsystem is configured to: apply a softmax function to a set of gating parameters having learned values to generate a respective softmax score for each of one or more of the plurality of expert neural networks; determine a respective weight for each of the one or more of the plurality of expert neural networks; select a proper subset of the plurality of expert neural networks; and combine the respective expert outputs generated by the one or more expert neural networks in the proper subset to generate one or more MoE outputs.

Abstract: A coated optical fiber includes an optical fiber having a core and an outer surface, and a homogeneous polymer coating in contact with the outer surface of the optical fiber. The optical fiber and the homogeneous polymer coating are UV transparent, and a refractive index of the outer surface of the optical fiber or the homogeneous polymer coating is up to 15% less than a refractive index of the core. Coating the optical fiber includes coating an outer surface with a polymerizable material and polymerizing the polymerizerable material to yield the coated optical fiber having a homogeneous polymer coating. The optical fiber and the homogeneous polymer coating are UV transparent, and a refractive index of the outer surface of the optical fiber or the homogeneous polymer coating is up to 15% less than a refractive index of a core of the optical fiber.