首页期刊导航|Robotics & Machine Learning Daily News
期刊信息/Journal information
Robotics & Machine Learning Daily News
NewsRx
Robotics & Machine Learning Daily News

NewsRx

Robotics & Machine Learning Daily News/Journal Robotics & Machine Learning Daily News
正式出版
收录年代

    'Rotary Locking Device Or Mechanism' in Patent Application Approval Process (USP TO 20240328454)

    108-111页
    查看更多>>摘要：The following quote was obtained by the news editors from the background informa tion supplied by the inventors: “In the common rotary locking device or mechanis m of the prior art, two connected rotatable bodies are typically connected via a round shaft, and then are locked or unlocked using the principle of triangular fixation (such as a movable support rod) or by a locking device or mechanism, su ch as an automatic door, a hidden bed, and a rotary support rod, which are expos ed outside a product, not only affecting the overall aesthetics of the product t o a large extent, but also occupying external space, and even impeding the norma l use of the product, so the common rotary locking device or mechanism is diffic ult to popularize in life. In addition, products involving rotary locking mechan isms, such as backrests of automated seats, robotic arms, robotic grippers, medi cal instruments, industrial robots, and joints of humanoid robots, are prevalent in the fields including industry, transportation, medical, aerospace, and artif icial intelligence. In the prior art, rotation is usually realized by means of p inion and rack drive, worm gear drive, belt drive or chain drive, and other lock ing devices are used to realize locking. Due to complex structure, high manufact uring cost, large size and other disadvantages, the rotary locking mechanism is limited in scope of application, especially in some products or mountings with s pecial limited in shape and appearance, such as surgical robotic hands, humanoid robot fingers, and small-sized robotic worms.”
      原文链接:
    • NETL
    • NSTL

    'Distributed Spoken Language Interface For Control Of Apparatuses' in Patent App lication Approval Process (USPTO 20240330590)

    111-114页
    查看更多>>摘要：The following quote was obtained by the news editors from the background informa tion supplied by the inventors: “Existing keyphrase recognition systems are typi cally based on machine-learning (ML) techniques. Such systems are generated by c ollecting a substantial amount of data of people with different accents speaking the keyphrase, and then training a machine-learning model, such as a neural net work, to provide a recognition when the keyphrase is spoken. Generating a keyphr ase recognition system in such a fashion is intensive in terms of both computing resources and human resources. As a result, generating a new keyphrase recognit ion system or modifying an existing one by adding new keyphrases tends to be bur densome.
      原文链接:
    • NETL
    • NSTL

    Patent Application Titled 'Method And Apparatus For Ai/Ml Model Monitoring' Publ ished Online (USPTO 20240334201)

    115-118页
    查看更多>>摘要：Reporters obtained the following quote from the background information supplied by the inventors: “Currently, a 3GPP framework for an AI/ML model for an air-int erface corresponding to each target use case regarding aspects such as performan ce, complexity, and potential specification impact is being explored. Some of th ese use cases focus on channel state information (CSI) feedback enhancement (e.g ., overhead reduction, improved accuracy, prediction), beam management (e.g., be am prediction in time, and/or spatial domain for overhead and latency reduction, beam selection accuracy improvement), and positioning accuracy enhancements for different scenarios including, for example, those with heavy non-line-of-sight (NLOS)conditions.
      原文链接:
    • NETL
    • NSTL

    Patent Issued for Privacy supporting messaging systems and methods (USPTO 121058 46)

    118-120页
    查看更多>>摘要：The following quote was obtained by the news editors from the background informa tion supplied by the inventors: “Modern devices, including mobile devices such a s smartphones and tablets, and computers, including laptops, desktops, and serve rs, are capable of running a variety of applications that can communicate with r emote systems via a network, such as the Internet. The content presented by thes e applications to users can vary based upon data exchanged over the Internet. In some scenarios, data received from the device may be about one or more aspects of the user, to allow delivered content to be targeted to the user.
      原文链接:
    • NETL
    • NSTL

    'Systems And Methods For Use In Object Processing Using Dynamic Wheel Assemblies ' in Patent Application Approval Process (USPTO 20240327145)

    121-124页
    查看更多>>摘要：The following quote was obtained by the news editors from the background informa tion supplied by the inventors: “The invention generally relates to automated, r obotic and other object processing systems such as sortation systems, and relate s in particular to automated and robotic systems intended for use in environment s requiring, for example, that a variety of objects (e.g., parcels, packages, an d articles, etc.) be processed and distributed to several output destinations.
      原文链接:
    • NETL
    • NSTL

    Patent Application Titled 'Blockwise Controlled Decoding Of Natural Language (Nl ) Based Output Generated Using A Large Language Model (Llm) To Reduce Latency In Rendering Thereof' Published Online (USPTO 20240330334)

    124-128页
    查看更多>>摘要：Reporters obtained the following quote from the background information supplied by the inventors: “Large language models (LLMs) are particular types of machine learning models that can perform various natural language processing (NLP) tasks , such as language generation, machine translation, and questionanswering. Thes e LLMs are typically trained on enormous amounts of diverse data including data from, but not limited to, webpages, electronic books, software code, electronic news articles, and machine translation data. Accordingly, these LLMs leverage th e underlying data on which they were trained in performing these various NLP tas ks. For instance, in performing a language generation task, these LLMs can proce ss a natural language (NL) based input that is received from a client device, an d generate a NL based output that is responsive to the NL based input and that i s to be rendered at the client device. However, in generating the NL based outpu t utilizing these LLMs, additional latency is introduced that may not be present absent utilizing these LLMs. This additional latency can prolong user interacti ons with these LLMs and detract from a user experience with these LLMs. Accordin gly, there is a need in the art for reducing latency in utilizing these LLMs.”
      原文链接:
    • NETL
    • NSTL

    Patent Application Titled 'Method For Ascertaining A Descriptor Image For An Ima ge Of An Object' Published Online (USPTO20240331213)

    128-132页
    查看更多>>摘要：Reporters obtained the following quote from the background information supplied by the inventors: “In order to enable flexible production or processing of the o bjects by a robot, it is desirable for the robot to be able to handle an object regardless of the position in which the object is placed in the workspace of the robot. The robot should therefore be able to recognize which parts of the objec t are in which positions, so that it can, for instance, grab the object at the c orrect location, for example to fasten it to another object or weld the object a t the current location. This means that the robot should be able to recognize th e pose (position and orientation) of the object, for example from one or more im ages taken by a camera attached to the robot, or ascertain the position of locat ions for picking up or processing. One approach to achieving this is determining descriptors, i.e., points (vectors) in a predefined descriptor space, for parts of the object (i.e. pixels of the object represented in an image plane), wherei n the robot is trained to assign the same descriptors to the same parts of an ob ject regardless of a current pose of the object and thus recognize the topology of the object in the image, so that it is then known, for example, where which c orner of the object is located in the image. Knowing the pose of the camera then makes it possible to infer the pose of the object. The recognition of the topol ogy can be realized with a machine learning model (ML model) that is trained acc ordingly.
      原文链接:
    • NETL
    • NSTL

    Patent Issued for Method for handling spacer frames (USPTO 12104433)

    132-134页
    查看更多>>摘要：From the background information supplied by the inventors, news correspondents o btained the following quote: “Field of the Invention “The invention relates to a method for handling spacer frames for insulating gla ss. “Description of the Related Art “When producing insulating glass, i.e., spacer frames are used, which are compos ed of sections of hollow-section strips or are formed by bending hollow-section strips. “In the course of the production process for insulating glass, the spacer frames have to be transferred from a station for producing spacer frames and fed to a station for coating spacer frames with an adhesive and sealant material. In addi tion, the spacer frames that are coated with adhesive material and sealant have to be moved to a station for assembling insulating glass in order to be mounted there on a glass panel. “For moving spacer frames between the individual stations of a facility (“line”) for producing insulating glass, various devices are known for when spacer frame s are not moved by hand from one station to the next.
      原文链接:
    • NETL
    • NSTL

    'Item Sorting Apparatus And Interfaces Therefor' in Patent Application Approval Process (USPTO 20240326098)

    134-136页
    查看更多>>摘要：The following quote was obtained by the news editors from the background informa tion supplied by the inventors: “Contemporary item sorting systems and methods p resent a number of significant challenges. At one end of the spectrum, item sort ing has historically been accomplished by manual identification and sorting of i tems, for example employing people to manually select and sort items. This appro ach, however, is limited by the speed at which humans may identify and sort item s, and entails many of the risks commonly associated with manual sorting, such a s potential exposure to toxins, sharp objects, moving machinery, and other dange rs that may be presented by the items and/or machinery. Long-term use of people in this role also poses challenges such as the risk of repetitive stress injurie s. At the other end of the spectrum, systems for large-scale sorting of waste it ems, such as those employed at waste management sites to sort consumer waste int o bins or categories such as plastics of various types, cardboard and paper, rec yclable metals, and the like, are capable of processing large amounts of waste a utomatically, with little to no human intervention in the sorting process. Such systems often use optical sensors to identify waste items, and picking mechanism s to move the identified items to appropriate bins for processing. Item identifi cation is often inaccurate, however, with automatic identification methods commo nly unable to recognize unusual items, or items that have been discolored, dirti ed, crushed, warped, or otherwise deformed into difficult-to-recognize shapes. A ccordingly, efforts have been directed towards overcoming the challenges present ed at each end of the above spectrum, to generate more accurate and safer sortin g systems.”
      原文链接:
    • NETL
    • NSTL

    Patent Issued for Method and system for preprocessing optimization of streaming video data (USPTO 12108024)

    137-140页
    查看更多>>摘要：News editors obtained the following quote from the background information suppli ed by the inventors: “This disclosure relates generally to the optimization of p reprocessing of streaming video data and, more specifically, to the optimization of preprocessing parameters to improve a main output of a main artificial intel ligence model. “Cameras are beneficial for use in many areas of commercial and personal practice. For example, security cameras are used within (and outside) commercial wareho uses and on private personal property. Other applications use cameras along asse mbly lines for quality control purposes. With the increased capabilities of came ras having higher quality imagery (i.e., resolution) and a wider field of view, more area can be shown in the streaming video by the camera. A large portion of the frame/field of view may be of little or no interest to the consumer (e.g., a security or manufacturing company). However, current practices relay the entire ty of the streaming video (i.e., the entire frame/field of view) to the consumer , which can be time and resource consuming due to the need to transfer large fra me (i.e., field of view), high resolution video data.”
      原文链接:
    • NETL
    • NSTL