This is the first post of what I intended to be a weekly blog listing patents revolving around artificial intelligence, machine learning, robots and autonomous vehicles. I filter through newly issued patents issued every week by the USPTO and pick out the ones that I believe are interesting and hold weight in the AI/ML/Robotics/AV domains. This is intended to be a quick read on Friday afternoons and if you are interested in exploring any of the patents they are easily accessible on the USPTO website or Google Patents. I'm striving to describe the patents using a "in a nutshell" approach via a single sentence or two, followed by the Abstract and Background of each patent.
Why do this? Looking at recently issued patents (and published patent applications) provide a bellwether for technological trends and spur innovation. Just because a patent solves a particular problem one way does not preclude other, novel solutions. It's my hope that this blog in some small way is informative and spurs inventive thinking.
MACHINE LEARNING
My top pick is:
US 11468275 B1 Computer Vision Using A Prior Probability Distribution Selected Based On An Image Capture Condition
Assignee: Apple Inc.
In a nutshell: takes a previously generated probabilistic model for an image and applies it to a new image capture with the intent of improving accuracy of the ML model used.
Abstract
A machine learning (ML) model is trained and used to produce a probability distribution associated with a computer vision task. The ML model uses a prior probability distribution associated with a particular image capture condition determined based on sensor data. For example, given that an image was captured by an image capture device at a particular height above the floor and angle relative to the vertical world axis, a prior probability distribution for that particular image capture device condition can be used in performing a computer vision task on the image. Accordingly, the machine learning model is given the image as input as well as the prior probability distribution for the particular image capture device condition. The use of the prior probability distribution can improve the accuracy, efficiency, or effectiveness of the ML learning model for the computer vison task.
BACKGROUND
(2) Like human vision, computer vision tasks generally attempt to reason from one or more images. Humans, however, frequently have the benefit of at least some background knowledge or expectations that are useful to such reasoning. For example, humans often have implicit notions of where certain types of things might be in the areas around them. As a specific example, a human, even with eyes closed, may have a notion that a table is more likely to be ahead of him than above him on the ceiling. Computer vision tasks, in contrast, generally attempt to reason from images without the benefit of such background knowledge and expectations and thus can be less accurate, efficient, or effective than desired.
ARTIFICIAL INTELLIGENCE
My top pick is:
US 11467604 B2 Control Device And Method For A Plurality Of Robots
Assignee: LG Electronics Inc.
In a nutshell: measures the density of people in a high concentration area and deploys a moveable robot near the area. The robot may be a drone, robot, augmented reality device and the like.
Abstract
Disclosed is a device and method of controlling a plurality of robots. According to an embodiment, a device and method of controlling a plurality of robots periodically measures variations in the density of people per unit quarter and deploys a robot, which is positioned close to a high-density unit quarter and has a low workload, in the unit quarter. According to an embodiment, the artificial intelligence (AI) module may be related to unmanned aerial vehicles (UAVs), robots, augmented reality (AR) devices, virtual reality (VR) devices, and 5G service-related devices.
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. 10-2019-0177765, filed on Dec. 30, 2019, the contents of which are hereby incorporated by reference herein in its entirety.
TECHNICAL FIELD
(2) Embodiments of the disclosure relate to devices and methods for controlling a plurality of robots.
DISCUSSION OF RELATED ART
(3) To respond to increasing travelers in airports or train stations, robots are stationed to play a role as a smart hub to provide various services.
(4) In particular, artificial intelligence (AI) robots may take place of conventional computers, signage, or kiosks to handle the tasks that the legacy devices cannot, thus delivering an enhancement in service in view of both quality and quantity.
(5) AI robots are also useful as guides or info desks in busy places.
(6) However, conventional robots are set to serve only in specific areas assigned thereto, but not in the other areas.
(7) Thus, all of the robots deployed in an area may not efficiently be used, with only some on duty in a dense space overloaded.
(8) Further, robots may stray out of their working area and roam around in other areas without returning, failing to properly deliver the services they are supposed to.
AUTONOMOUS VEHICLES
My top pick is:
US 11466999 B1 Altering Autonomous Or Semi-Autonomous Vehicle Operation Based On Route Traversal Values
ASSIGNEE INFORMATION
NAME
Allstate Insurance Company
In a nutshell: calculates route traversal values to mitigate risk by selecting less risky routes.
Abstract
A method is disclosed for mitigating the risks associated with operating an autonomous or semi-autonomous vehicle by using calculated route traversal values to select less risky travel routes and/or modify vehicle operation. Various approaches to achieving this risk mitigation are presented. A computing device is configured to generate a database of route traversal values. This device may receive a variety of historical route traversal information, real-time vehicle information, and/or route information from one of more data sources and calculate a route traversal value for the associated driving route. Subsequently, the computing device may provide the associated route traversal value to other devices, such as a vehicle navigation device associated with the autonomous or semi-autonomous vehicle. An insurance company may use this information to help determine insurance premiums for autonomous or semi-autonomous vehicles by analyzing and/or mitigating the risk associated with operating those vehicles.
Background/Summary
Note: the background contains references to continuations and is not informative.
CROSS-REFERENCE TO RELATED APPLICATIONS
(1) This application is a continuation of U.S. application Ser. No. 15/784,922 filed Oct. 16, 2017, entitled “ALTERING AUTONOMOUS OR SEMI-AUTONOMOUS VEHICLE OPERATION BASED ON ROUTE TRAVERSAL VALUES,” which is a continuation of Ser. No. 15/408,938 filed Jan. 18, 2017, issued as U.S. Pat. No. 9,816,827 on Nov. 14, 2017, entitled “ALTERING AUTONOMOUS OR SEMI-AUTONOMOUS VEHICLE OPERATION BASED ON ROUTE TRAVERSAL VALUES,” which is a continuation of U.S. application Ser. No. 14/849,045 filed Sep. 9, 2015, issued as U.S. Pat. No. 9,587,952 on Mar. 7, 2017, entitled “ALTERING AUTONOMOUS OR SEMI-AUTONOMOUS VEHICLE OPERATION BASED ON ROUTE TRAVERSAL VALUES.” The contents of the above noted applications are hereby incorporated by reference in their entirety.
So that's it for this week. Comments are welcome!
コメント