1.1Four rotors for flight
Four brushless motors drive vertically oriented propellers in counter-rotating pairs. A flight controller varies motor speeds using inertial and altitude feedback to hold a hover or tilt the body for horizontal travel.
Generated automatically. Not reviewed by a person. Not legal advice or an opinion.
Questions about how individual parts of claim 1 might be changed. A question may describe something different from what the claim language recites, or a version of it; no view is offered on which.
Four brushless motors drive vertically oriented propellers in counter-rotating pairs. A flight controller varies motor speeds using inertial and altitude feedback to hold a hover or tilt the body for horizontal travel.
A transverse frame inside the shell supports the wing pivots and transfers flapping loads between them. Internal mounts hold the battery and circuit boards near the center of mass.
The central processor prepares image frames, audio features, gas sensor readings, and touch readings for inference. An accelerator with a fixed-point matrix engine processes these inputs using an 8-bit model, and the central processor updates selected model parameters from feedback.
A reader coil energizes passive tags and reads their identifiers when they enter its field. The processor matches each identifier to a stored object label and passes the label to the activity recognition model.
A wide-angle camera and microphone capture activity within the dock's viewing and listening area. The processor timestamps recognized activities and spoken requests and groups recurring events by time of day.
Further arrangements of the same subject matter, described here as of the publication date at the foot of this page.
A doll-shaped robot with two flapping wings accompanies a guide through an indoor exhibit gallery. A torso camera tracks a printed marker on the guide’s badge, and a forward range sensor measures clearance before the flight controller commands directional motion. Floor markers identify exhibit stops where the robot hovers and plays a prerecorded narration through a torso speaker. The robot stores the order of stops for each tour route in writable memory.
A geared brushless motor replaces the wing servo motors and turns a transverse crankshaft inside the torso. Mirrored connecting rods couple the crankshaft to the two wing roots and drive a symmetric flapping stroke. A magnetic encoder measures shaft speed, and the flight controller adjusts motor current to hold the commanded wingbeat rate. Two linear actuators shift the left and right wing pitch linkages independently to steer while the crankshaft continues turning.
An integer tensor accelerator replaces the graphics processor on the control board in the torso. The central processor captures camera frames, microphone samples, gas sensor readings, and touch readings and converts them into fixed-size inputs for trained models. The accelerator runs quantized models and returns user identities and activity labels to the central processor. The central processor updates user profiles in writable memory and runs the wing control loop on a separate timed task.
A stereo camera in the upper torso measures the distance to the recognized user at 30 frames per second. The robot holds a hover between 1.5 and 2.0 meters from the user and follows at up to 0.5 meters per second when separation exceeds 2.0 meters. Below 1.2 meters, it backs away when a rear range sensor confirms clearance, then resumes hovering at 1.5 meters. Between 1.2 and 1.5 meters, it holds position.
A charging station combines a broad landing pad, a printed alignment marker, and two recessed electrical contacts. A downward-facing camera in the doll’s torso estimates its pose relative to the marker, and the wing controller centers the feet over the pad during descent. Tapered guides align conductive shoe pads with the station contacts as the robot lands. The station enables charging after it detects a seated robot, and the robot stops its wing drive after confirming touchdown.
The central processor groups time-stamped room arrivals and spoken requests by recognized user and stores the groups in writable memory. The graphics processor runs speech recognition on microphone audio, and the central processor records the resulting request labels alongside room labels from camera observations. When the same user makes the same request in the same room within the same 30-minute period on three separate days, the software stores a routine entry. On a later matching visit, the robot offers the associated action once and records the user's response.
A brushless motor drives a reduction gear and two crank linkages that oscillate carbon-fiber wing spars covered with thin polyester film. Servo-driven linkages adjust wing pitch independently to control body attitude, and the controller varies flapping frequency to regulate lift.
A hollow molded torso carries the wing mounts and supports a head on a neck pivot. Pin joints at the shoulders, elbows, wrists, hips, knees, and ankles connect the paired arms, hands, legs, and feet.
A camera, digital microphone, metal-oxide gas sensor, and capacitive touch pads provide image, audio, chemical, and contact measurements. The central processor timestamps the readings and prepares feature windows, and the graphics processor runs a model that combines the four inputs.
Onboard flash stores instructions that track a selected person in camera frames and use depth measurements to estimate separation. A flight controller uses that estimate to maintain a set following distance while the processor records timestamps for recognized activities and spoken requests.
Passages reproduced word for word from two earlier-published U.S. patent documents, placed beside the language of claim 1. No characterization is offered. Each passage is marked A or B for the document it comes from; ¶ numbers are the paragraph numbers in the USPTO’s published full text of that document.
The body 110 is comprised of a head 120 preferably comprised of hair 121, ears 122, eyes 124, a nose 126, and a mouth 128.B · US 2024/0252938 A1 · ¶ 25
In one embodiment, the mobile application 350 is comprised of an artificial intelligence. The user can converse with the AI and ask a question into the microphone 330 wherein the AI then generates an answer to the question and plays the answer via the speaker 310 or from the smart device the mobile application 350 is installed on.B · US 2024/0252938 A1 · ¶ 31
The memory 104 can also include (e.g., store) a software model 105. The software model 105 can be, for example, an artificial intelligence (AI) model, machine learning (ML) model, analytical model, mathematical model, or a combination thereof.A · US 11,772,667 B1 · ¶ 42
An empty cell means only that no passage was reproduced there.
To collect sensor data that accurately and completely represents an area(s) of interest, it can be desirable to ensure that the sensors collecting such sensor data are operating as desired and remaining consistent in location relative to the vehicle.A · US 11,772,667 B1 · ¶ 2
One or more embodiments are related to operating a vehicle (e.g., autonomous vehicle) in response to detecting a faulty sensor using calibration parameters of the sensors.A · US 11,772,667 B1 · ¶ 1
Further, it provides users with a way to customize the device with interchangeable accessories that are representative of the respective celebrity.B · US 2024/0252938 A1 · ¶ 2
Nonetheless, it is to be appreciated that aspects of the present invention are also equally applicable to other like applications, devices, and methods of manufacture.B · US 2024/0252938 A1 · ¶ 2
patent.fyi, US 2026/0355642 A1, claim 1, first published 2026-11-19T14:02:27-05:00, https://patent.fyi/US20260355642/, SHA-256 f13bf9931a58d1879e010d4f775c9f6dc4dd682feab875f43f8947a4dfba9ba3