Abstract: A device for nullifying images captured by drones utilizing a high power light to overexpose said images. One or more cameras installed in the device capture images of the area surrounding the device. Computer vision and deep learning are utilized to identify drones in the captured images. If a drone is identified, the location thereof is estimated. A high power light installed in the device is directed at the estimated location of the drone to overexpose any images that are being taken of the area around the device by the drone.
Abstract: A method for identifying objects by an autonomous robot. Images of a work environment are captured by an image sensor positioned on the autonomous robot. An object dictionary containing image data regarding a set of objects is preloaded into the system. Captured images are transmitted to an image processor that processes the images to generate feature vector and identify objects within the images by comparison to objects in the object dictionary. Upon identifying an object in an image as an object from the object dictionary, the system executes a set of preprogrammed responses, which may include altering a movement path to avoid colliding with or driving over the object.
Abstract: Provided is a process, including: obtaining, with one or more processors, first depth data, wherein: the first depth data indicates a first distance from a robot at a first position to a surface of, or in, a workspace in which the robot is disposed, the first depth data indicates a first direction in which the first distance is measured, the first depth data indicates the first distance and the first direction in a frame of reference of the robot, and the frame of reference of the robot is different from a frame of reference of the workspace; translating, with one or more processors, the first depth data into translated first depth data that is in the frame of reference of the workspace; and storing, with one or more processors, the translated first depth data in memory.
Type:
Grant
Filed:
October 17, 2018
Date of Patent:
September 24, 2019
Assignee:
AI Incorporated
Inventors:
Ali Ebrahimi Afrouzi, Sebastian Schweigert, Lukas Fath, Chen Zhang
Abstract: A front suspension wheel for mobile robotic devices that can be compressed into or decompressed out of a main body of a mobile robotic device to facilitate driving the mobile robotic device over obstacles, thresholds and the like. The wheel will provide the mobile robotic device with information such as how fast the wheel is traveling. The wheel is easily removable by hand by the user.
Type:
Application
Filed:
March 16, 2018
Publication date:
September 19, 2019
Applicant:
AI Incorporated
Inventors:
Ali Ebrahimi Afrouzi, Masoud Nasiri Sarvi, Scott McDonald
Abstract: A method for increasing the rate of debris collection of a robotic vacuum through increasing the number of times a robotic vacuum's cleaning apparatus passes over a work surface during each pass of the device. The device's main cleaning apparatus is installed on a plate that rotates within the housing of the device. The drive unit is housed separately from the rotating plate so that the device's normal movement patterns will be uninterrupted by the rotation of the cleaning apparatus. Ideally, the cleaning apparatus will be caused to rotate two or more times over an area before the robotic vacuum has driven completely through the area.
Type:
Grant
Filed:
January 23, 2018
Date of Patent:
September 17, 2019
Assignee:
AI Incorporated
Inventors:
Ali Ebrahimi Afrouzi, Soroush Mehrnia, Azadeh Afshar Bakooshli
Abstract: A method for estimating distances from a baseplate to arbitrary objects or surfaces around the baseplate. Two laser emitters emitting converging collimated laser beams, an image sensor, and an image processor are provided on the baseplate. Images of the laser projections on surfaces are captured by the image sensor. Captured images are sent to the image processor which extracts the pixel data of the laser projections within the images. The distance between the laser projections of the two laser emitters is analyzed and compared to figures in a preconfigured table that relates distances between laser projections with corresponding distances of the baseplate to projection surfaces in order to estimate a distance of the baseplate to the projection surface in the images.
Abstract: The present invention introduces a method for guiding or directing the heading of a mobile robotic device. A light pattern is projected from a light emitting unit disposed on a robotic device. The angle of the plane of the projection of the light pattern with respect to a heading of the robotic device is preset, but may be any angle as desired by a manufacturer or operator. A camera positioned in a plane parallel to the plane of the light pattern projection captures images of the projected light pattern on surfaces substantially opposite the light emitting unit. Images are processed to check for reflection symmetry about a vertical centerline of the images. Upon detecting an image that does not have reflection symmetry, the robotic device turns to adjust its heading with relation to the surfaces in the environment on which the light pattern is projected. Turning amounts and directions are provided to the controller and may be based on analysis of the last image captured.
Abstract: A path planning method for a robotic floor-cleaning device in which a user's commands are repeated autonomously by the robotic floor-cleaning device at a later time.