Here are this week’s patent picks and some brief commentary. The categories are Autonomous Vehicles, Machine Learning and Intelligent Sensors.
Autonomous Vehicles
Zoox appears to be very active in the autonomous vehicle patent space with six issued patent this week. The patents address sensor mitigation, potentially expanding a drivable area if an area is obstructed, detecting blemishes on optical devices, identifying and removing false positives from lidar sensor output, a vehicle health monitor and processing sound to enhance teleoperation of a vehicle.
My Patent Pick:
Sensor Obstruction Detection And Mitigation
DOCUMENT ID
US 11479213 B1
DATE PUBLISHED
2022-10-25
ASSIGNEE INFORMATION
NAME
Zoox, Inc.
In a Nutshell
Detects sensor obstructions and attempts to work them or clear them if possible.
Abstract
A system may be configured to detect an obstruction on sensor at least partially obstructing or distorting a portion of the field of view of the sensor. The system may also be configured to mitigate the effects of the obstruction or at least partially remove the obstruction from the sensor. The system may be configured to receive one or more signals from a sensor configured to generate signals indicative of an environment, which may include one or more objects, in which the sensor is present. The system may be configured to determine, for example, classify, based at least in part on the one or more signals, an obstruction or distortion on a surface of the sensor, and initiate a response, based at least in part on the determination, to mitigate effects of the obstruction and/or at least partially remove the obstruction.
Background/Summary
BACKGROUND
(1) Sensors may be used to generate sensor data indicative of objects in an environment. However, the effectiveness of the sensor may be compromised if the operating view of the sensor is partially or completely obstructed by an obstruction, such as, for example, dirt, bird droppings, bugs, lens scratches, or moisture (e.g., rain, snow, ice, condensation, etc.). Such obstructions may block a portion of the field of view of the sensor and/or generate sensor data that provides an inaccurate representation of one or more objects present in the environment in which the sensor is located. This situation may be particularly problematic when, for example, the sensor is used to provide sensor data for autonomous operation of a machine without the presence or assistance of an operator that might be able to detect and/or remove the obstruction from the sensor. For example, an autonomous vehicle operating without an operator relies at least in part on data generated by sensors for proper operation. If one or more of the sensors has an obstruction that affects the data generated by the sensor, the autonomous vehicle may take actions based on inaccurate information related to the environment through which it is travelling, such that, for example, the autonomous vehicle is unable to detect or identify objects, or locate the positions and/or trajectories of objects in the environment. This may hinder operation of such vehicles.
Machine Learning
Active in the machine learning space are Amazon, Wells Fargo Banking, Allstate Insurance Company among others. Amazon dominates with over a dozen patents issued.
My Pick:
In a Nutshell:
Attaches filter values to recorded audio input to determine the best configuration of microphones to improve microphone placement.
Beamforming Using Filter Coefficients Corresponding To Virtual Microphones
DOCUMENT ID
US 11483646 B1
DATE PUBLISHED
2022-10-25
APPLICANT INFORMATION
NAME
Amazon Technologies, Inc.
Abstract
Techniques for improving beamforming using filter coefficient values corresponding to virtual microphones are described. A system may define “virtual” microphone positions and determine corresponding filter coefficient values. These filter coefficient values may be applied to input audio data captured by actual physical microphones, enabling the system to improve performance of beamforming and/or to reduce a number of physical microphones without degrading performance. Offline testing and simulations may be performed to identify the best combination of virtual microphones and/or filter coefficient values for a particular look-direction. For example, the simulations may identify that a first filter coefficient corresponding to a first virtual microphone and a first direction will be associated with a first physical microphone and the first direction. During run-time processing, a device may generate beamformed audio data for the first direction by applying the first filter coefficient to input audio data captured by the first physical microphone.
Background/Summary
BACKGROUND
(1) In audio systems, beamforming refers to techniques that are used to isolate audio from a particular direction. Beamforming may be particularly useful when filtering out noise from non-desired directions. Beamforming may be used for various tasks, including isolating voice commands to be executed by a speech-processing system.
(2) Speech recognition systems have progressed to the point where humans can interact with computing devices using speech. Such systems employ techniques to identify the words spoken by a human user based on the various qualities of a received audio input. Speech recognition combined with natural language understanding processing techniques enable speech-based user control of a computing device to perform tasks based on the user's spoken commands. The combination of speech recognition and natural language understanding processing techniques is commonly referred to as speech processing. Speech processing may also convert a user's speech into text data which may then be provided to various text-based software applications.
(3) Speech processing may be used by computers, hand-held devices, telephone computer systems, kiosks, and a wide variety of other devices, such as those with beamforming capability, to improve human-computer interactions.
Intelligent Sensors
In a Nutshell
I found this patent really interesting. It uses neural networks that take input as vibration sensor data and video images, and can predict vibration patents in a washing machine and compensate for them without an elaborate sensor suite.
Method And Apparatus For Compensating Vibration Of Deep-learning Based Washing Machine
DOCUMENT ID
US 11479894 B2
DATE PUBLISHED
2022-10-25
ASSIGNEE INFORMATION
NAME
LG ELECTRONICS INC.
Abstract
Provided are a method and an apparatus for analyzing a vibration of a deep-learning based washing machine. In the method for analyzing a vibration of a deep-learning based washing machine according to an embodiment of the present invention, a washing tub of the washing machine includes a specific shape pattern, an artificial neural network model is learned from a video image obtained by photographing the shape pattern through a camera and a vibration value sensed through the vibration sensor, and thus, by using the artificial neural network model, it is possible to predict a vibration value of the washing machine using the camera of the washing machine even without a vibration sensor. According to the present invention, a smart washing machine without the vibration sensor such as 6-axis gyro sensor can be implemented. The AI device of the present invention can be associated with an unmanned aerial vehicle (UAV), a robot, an augmented reality (AR) device, a virtual reality (VR) device, and a device related to a 5G service.
Background/Summary
CROSS-REFERENCE TO RELATED APPLICATIONS
(1) Pursuant to 35 U.S.C. § 119(a), this application claims the benefit of earlier filing date and right of priority to Korean Patent Application No. 10-2019-0099965, filed on Aug. 15, 2019, the contents of which are hereby incorporated by reference herein in its entirety.
BACKGROUND OF THE INVENTION
Field of the Invention
(2) The present invention relates to a method and an apparatus for analyzing a vibration of a deep-learning based washing machine, and particularly, a method and an apparatus for analyzing vibration of a deep-learning based washing machine capable of predicting a vibration value from an image through a camera.
Related Art
(3) In general, a washing machine refers to various devices for treating fabrics by applying physical and/or chemical action to laundry such as clothes and bedding. The washing machine includes an outer tub which stores washing water and an inner tub which stores the fabrics and is rotatably installed in the in the outer tub. In general, a washing method by the washing machine includes a washing process of rotating the inner tub to wash the fabrics and a dewatering process of dewatering the fabrics using a centrifugal force of an inner tub.
(4) When the washing machine is operated, vibration are generated according to an operation of a motor. In this case, an accurate prediction of the vibration can be used as an importance evidence related to the operation of the washing machine to determine whether to enter an operation of a next step through an increase in RPM of a washing tub or determine whether to stop a current step and perform restart.
(5) In a washing machine of the related art, in order to measure the vibration value, a method of obtaining control information based on a voltage and a current or sensing the vibration value through a 6-axis gyro sensor.
Comentários