Abstract: A glass tube processing parameter detection system includes a tube descending platform, a conveying apparatus, a plurality of clamping apparatuses, a tube descending basket, a terminal device, and a photographing apparatus. The conveying apparatus includes a conveyor belt and conveyor belt rotating shafts. The clamping apparatuses are disposed on the conveyor belt to fix the glass tube to be detected. The photographing apparatus is configured to photograph end face images corresponding to the glass tube to be detected, and send the end face images corresponding to the glass tube to be detected to the terminal device. The terminal device is configured to control start and stop of rotation of the conveyor belt rotating shafts, and determine the processing parameters corresponding to the glass tube to be detected according to the end face images corresponding to the glass tube to be detected.

Abstract: A method for detecting an acid-etching speed of core glass includes obtaining an initial end-face image and an acid-etched end-face image sequence corresponding to a sample microchannel plate blank, selecting a plurality of target microholes in the initial end-face image and intercepting a microhole acid-etching image sequence corresponding to each target microhole from the plurality of acid-etched end-face images according to position coordinates of each target microhole; determining a plurality of total optical flow motion fields corresponding to each target microhole based on a preset optical flow algorithm and the plurality of microhole acid-etching images corresponding to each target microhole, determining a maximum motion speed value among the plurality of total optical flow motion fields as a target motion speed value corresponding to the sample microchannel plate blank, and determining an acid-etching speed of core glass corresponding to the sample microchannel plate blank.

Abstract: A method for detecting dark spots on an end face of an optical fiber image transmission element comprises: obtaining a plurality of end face images for an optical fiber image transmission element to be detected by capturing different regions of the end face of the optical fiber image transmission element to be detected using a shooting apparatus; performing secondary windowed Fourier transform processing on each of the end face images based on a plurality of preset windows to generate a secondary masked image for each of the end face images, where the secondary masked image includes a plurality of closed regions; and determining the number of dark spots included in each of the end face images and an area value for each dark spot according to the number of closed regions included in each of the secondary masked images and the number of pixel points included in each closed region.

Abstract: A glass tube processing parameter detection method includes: acquiring an end face image of the glass tube; generating an edge binary image corresponding to the glass tube according to preset rules, a plurality of preset mask matrices, and the end face image, where the edge binary image includes a first circular ring composed of a plurality of first pixel points and a second circular ring composed of a plurality of second pixel points; determining a center coordinate of the first circular ring and a center coordinate corresponding to the second circular ring; determining an inner-circle processing diameter and an outer-circle processing diameter corresponding to the glass tube according to coordinates corresponding to each of the first pixel points and the second pixel points; determining a concentricity deviation value corresponding to the glass tube according to the center coordinate corresponding to the first circular ring and the second circular ring.

Abstract: A method for detecting an acid etching speed of core glass includes obtaining an initial end face image and an acid-etched end face image sequence corresponding to a sample microchannel plate blank; selecting a plurality of target microholes in the initial end face image, and intercepting a microhole acid-etching image sequence corresponding to each of the target microholes; determining a plurality of in-plane velocity fields corresponding to each of the target microholes based on a preset digital speckle correlation algorithm and the plurality of microhole acid-etching images corresponding to each of the target microholes; determining a maximum velocity value in the plurality of in-plane velocity fields as a target velocity value corresponding to the sample microchannel plate blank; and determining the acid-etching speed of core glass corresponding to the sample microchannel plate blank.

Abstract: A method for detecting dark spots on an end face of an optical fiber image transmission element includes: acquiring a plurality of end face images corresponding to an optical fiber image transmission element to be detected; performing secondary diffusion processing on each of the end face images according to a plurality of preset diffusion matrices to generate a secondary diffusion image corresponding to each of the end face images; and determining a number of dark spots comprised in each of the end face images and an area value corresponding to each of the dark spots according to a number of closed regions comprised in each of the secondary diffusion images and a number of pixel points comprised in each of the closed regions.

Abstract: In one embodiment, a method includes receiving a natural language instruction from an interface device, receiving sensor data of a physical environment from a sensing system, converting the natural language instruction into instruction tokens and the sensor data into sensor tokens. The method further includes executing an autoregressive generative model on the instruction tokens and sensor tokens to output control tokens. The method further includes generating control commands based on the control tokens and transmitting the control commands to a robotic system. The control commands cause the robotic system to perform an action affecting the physical environment.

Abstract: Various embodiments of the present technology generally relate to robotic devices and artificial intelligence. More specifically, some embodiments relate to an artificial neural network training method that does not require extensive training data or time expenditure. The few-shot training model disclosed herein includes attempting to pick up items and, in response to a failed pick up attempt, transferring and generalizing information to similar regions to improve probability of success in future attempts. In some implementations, the training method is used to robotic device for picking items from a bin and perturbing items in a bin. When no picking strategies with high probability of success exist, the robotic device may perturb the contents of the bin to create new available pick-up points. In some implementations, the device may include one or more computer-vision systems.

Abstract: Various embodiments of the present technology generally relate to robotic devices, artificial intelligence, and computer vision. More specifically, some embodiments relate to an imaging process for detecting failure modes in a robotic motion environment. In one embodiment, a method of detecting failure modes in a robotic motion environment comprises collecting one or more images of a multiple scenes throughout a robotic motion cycle. Images may be collected by one or more cameras positioned at one or more locations for collecting images with various views. Images collected throughout the robotic motion cycle may be processed in real-time to determine if any failure modes are present in their respective scenes, report when failure modes are present, and may be used to direct a robotic device accordingly.

Abstract: Various embodiments of the technology described herein generally relate to systems and methods for trajectory optimization with machine learning techniques. More specifically, certain embodiments relate to using neural networks to quickly predict optimized robotic arm trajectories in a variety of scenarios. Systems and methods described herein use deep neural networks to quickly predict optimized robotic arm trajectories according to certain constraints. Optimization, in accordance with some embodiments of the present technology, may include optimizing trajectory geometry and dynamics while satisfying a number of constraints, including staying collision-free and minimizing the time it takes to complete the task.

Abstract: Various embodiments of the present technology generally relate to robotic devices and artificial intelligence. More specifically, some embodiments relate to a robotic device for picking items from a bin and perturbing items in a bin. In some implementations, the device may include one or more computer-vision systems. A computer-vision system, in accordance with the present technology, may use at least two two-dimensional images to generate three-dimensional (3D) information about the bin and items in the bin. Based on the 3D information, a strategy for picking up items from the bin is determined. When no strategies with high probability of success exist, the robotic device may perturb the contents of the bin to create new available pick-up points and re-attempt to pick up an item.

Abstract: Various embodiments of the technology described herein generally relate to systems and methods for trajectory optimization with machine learning techniques. More specifically, certain embodiments relate to using neural networks to quickly predict optimized robotic arm trajectories in a variety of scenarios. Systems and methods described herein use deep neural networks to quickly predict optimized robotic arm trajectories according to certain constraints. Optimization, in accordance with some embodiments of the present technology, may include optimizing trajectory geometry and dynamics while satisfying a number of constraints, including staying collision-free, and minimizing the time it takes to complete the task.

Abstract: Various embodiments of the present technology generally relate to robotic devices, artificial intelligence, and computer vision. More specifically, some embodiments relate to an imaging process for detecting failure modes in a robotic motion environment. In one embodiment, a method of detecting failure modes in a robotic motion environment comprises collecting one or more images of a multiple scenes throughout a robotic motion cycle. Images may be collected by one or more cameras positioned at one or more locations for collecting images with various views. Images collected throughout the robotic motion cycle may be processed in real-time to determine if any failure modes are present in their respective scenes, report when failure modes are present, and may be used to direct a robotic device accordingly.

Abstract: Various embodiments of the technology described herein generally relate to systems and methods for trajectory optimization with machine learning techniques. More specifically, certain embodiments relate to using neural networks to quickly predict optimized robotic arm trajectories in a variety of scenarios. Systems and methods described herein use deep neural networks to quickly predict optimized robotic arm trajectories according to certain constraints. Optimization, in accordance with some embodiments of the present technology, may include optimizing trajectory geometry and dynamics while satisfying a number of constraints, including staying collision-free and minimizing the time it takes to complete the task.

Abstract: Various embodiments of the present technology generally relate to robotic devices, computer-vision systems, and artificial intelligence. More specifically, some embodiments relate to a computer-vision system for robotic devices. In some implementations, a computer-vision is coupled to a robotic arm for picking items from a bin and perturbing items in a bin. A computer-vision system, in accordance with the present technology, may use at least two two-dimensional (2D) images to generate three-dimensional (3D) information about the bin and items in the bin. A method of training artificial neural networks to generate 3D information from 2D images includes obtaining ground truth data for a scene, obtaining at least two images of the scene, and providing the ground truth data and the at least two images to the artificial neural network configured to generate a depth map from the images based on the ground truth data.

Abstract: Various embodiments of the present technology generally relate to robotic devices, artificial intelligence, and computer vision. More specifically, some embodiments relate to an imaging process for detecting failure modes in a robotic motion environment. In one embodiment, a method of detecting failure modes in a robotic motion environment comprises collecting one or more images of a multiple scenes throughout a robotic motion cycle. Images may be collected by one or more cameras positioned at one or more locations for collecting images with various views. Images collected throughout the robotic motion cycle may be processed in real-time to determine if any failure modes are present in their respective scenes, report when failure modes are present, and may be used to direct a robotic device accordingly.

Abstract: Various embodiments of the technology described herein generally relate to systems and methods for trajectory optimization with machine learning techniques. More specifically, certain embodiments relate to using neural networks to quickly predict optimized robotic arm trajectories in a variety of scenarios. Systems and methods described herein use deep neural networks to quickly predict optimized robotic arm trajectories according to certain constraints. Optimization, in accordance with some embodiments of the present technology, may include optimizing trajectory geometry and dynamics while satisfying a number of constraints, including staying collision-free and minimizing the time it takes to complete the task.

Abstract: Various embodiments of the present technology generally relate to robotic devices and artificial intelligence. More specifically, some embodiments relate to a robotic device for picking items from a bin and perturbing items in a bin. The robotic device may include one or more picking elements and one or more perturbation elements for disturbing a present arrangement of items in the bin. In an exemplary embodiment, a perturbation element comprises a compressed air valve. In some implementations, the robotic device may also include one or more computer-vision systems. Based on image data from the one or more computer-vision systems, a strategy for picking up items from the bin is determined. When no strategies with high probability of success exist, the robotic device may perturb the contents of the bin to create new available pick-up points.

Abstract: Various embodiments of the present technology generally relate to robotic devices and artificial intelligence. More specifically, some embodiments relate to an artificial neural network training method that does not require extensive training data or time expenditure. The few-shot training model disclosed herein includes attempting to pick up items and, in response to a failed pick up attempt, transferring and generalizing information to similar regions to improve probability of success in future attempts. In some implementations, the training method is used to robotic device for picking items from a bin and perturbing items in a bin. When no picking strategies with high probability of success exist, the robotic device may perturb the contents of the bin to create new available pick-up points. In some implementations, the device may include one or more computer-vision systems.

Abstract: Various embodiments of the present technology generally relate to robotic devices and artificial intelligence. More specifically, some embodiments relate to a robotic device for picking items from a bin and perturbing items in a bin. The robotic device may include one or more picking elements and one or more perturbation elements for disturbing a present arrangement of items in the bin. In an exemplary embodiment, a perturbation element comprises a compressed air valve. In some implementations, the robotic device may also include one or more computer-vision systems. Based on image data from the one or more computer-vision systems, a strategy for picking up items from the bin is determined. When no strategies with high probability of success exist, the robotic device may perturb the contents of the bin to create new available pick-up points.