Method for automatically removing obstructions from robotic floor-cleaning devices

• US 10,524,627 B1
• Filed: 10/24/2017
• Issued: 01/07/2020
• Est. Priority Date: 10/05/2016
• Status: Active Grant

• Current Assignee: Ai Incorprated
• Original Assignee: Al, Inc.
• Inventors: Ebrahimi Afrouzi, Ali
• Primary Examiner(s): Entezari, Michelle M
• Application Number:
  US15/792,119
• Time in Patent Office: 805 Days
• Field of Search: None
• CPC Class Codes: A47L 2201/04 Automatic control of the travelling movement; Automatic obstacle detection
• CPC Class Codes: A47L 9/009 Carrying-vehicles; Arrangements of trollies or wheels; Means for avoiding mechanical obstacles
• CPC Class Codes: A47L 9/2805 Parameters or conditions being sensed
• CPC Class Codes: A47L 9/2852 Elements for displacement of the vacuum cleaner or the accessories therefor, e.g. wheels, casters or nozzles
• CPC Class Codes: A47L 9/2884 Details of arrangements of batteries or their installation
• CPC Class Codes: A47L 9/2889 Safety or protection devices or systems, e.g. for prevention of motor over-heating or for protection of the user
• CPC Class Codes: G05D 1/0214 in accordance with safety or protection criteria, e.g. avoiding hazardous areas monitoring the location of vehicles within a certain area, e.g. forbidden or allowed areas, in traffic control systems for road vehicles G08G1/13
• CPC Class Codes: G05D 1/0238 using obstacle or wall sensors G05D1/0246 and G05D1/0289 take precedence; lidar systems designed for anti-collision purposes G01S17/93
• CPC Class Codes: G05D 1/0246 using a video camera in combination with image processing means
• CPC Class Codes: G05D 1/0272 comprising means for registering the travel distance, e.g. revolutions of wheels measuring distance traversed on the ground by vehicles, e.g. using odometers G01C22/00
• CPC Class Codes: G05D 2201/0203 Cleaning or polishing vehicle
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06V 10/764 using classification, e.g. of video objects
• CPC Class Codes: G06V 10/82 using neural networks
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G06V 2201/07 Target detection
• CPC Class Codes: Y10S 901/01 Mobile robot
• CPC Class Codes: Y10S 901/47 Optical

Abstract

A method to automatically remove obstructions, such as electrical cords or wires, from robotic floor-cleaning devices after becoming entangled with wheels of the robotic device during operation. Upon sensing entanglement with an obstruction, the robotic floor-cleaning device is configured to retract one or more wheels into a wheel housing within the main housing of the robotic device. During retraction, the obstruction entangled around the one or more wheels is forcibly separated from the wheel by the corresponding wheel housing.

Claims

1. A method for automatically removing obstructions from a robotic floor-cleaning device wheel comprising:

defining one or more sets of conditions to actuate retraction of one or more wheels of the robotic floor-cleaning device into a corresponding wheel housing positioned within a main housing of the robotic floor-cleaning device;

interrupting a robotic floor-cleaning process upon satisfaction of the one or more sets of conditions for a predetermined amount of time;

retracting the one or more wheels into said corresponding wheel housing;

releasing the one or more wheels from said wheel housing after the predetermined amount of time; and

proceeding with the robotic floor-cleaning process,

wherein the wheel housing is designed with minimal amount of clearance between the wheel and adjacent wheel housing walls such that no other objects except a corresponding wheel may fit within the wheel housing during retraction and obstructions are forcibly pushed off and separated from the one or more wheels of the robotic floor-cleaning device by the wheel housing during retraction.

2. The method of claim 1 wherein a set of conditions for actuating retraction of one or more wheels includes any of:

a number of wheel rotations over a predetermined length of time falling below a predetermined threshold, and a current measurement drawn by a wheel exceeding a predetermined threshold.

3. The method of claim 2, wherein at least one odometer is used to monitor wheel rotations over time.

4. The method of claim 3, wherein a unique odometer is used to monitor each wheel and wheels are retracted individually when a corresponding odometer detects that wheel rotations over the predetermined length of time have fallen below the threshold.

5. The method of claim 2 wherein a separate wheel motor powers each wheel and the amount of current drawn by the separate wheel motors is monitored, and wheels are retracted individually when a corresponding motor detects that a current measurement drawn by a wheel exceeds the threshold.

6. The method of claim 1, further comprising using computer vision and deep learning to identify objects likely to obstruct wheels and determine when to actuate wheel retraction.

7. The method of claim 6, further comprising:

preloading a catalogue of objects likely to obstruct wheels into a database stored in a memory unit of the robotic floor-cleaning device;

extracting characteristics of the objects likely to obstruct wheels using computer vision;

iteratively capturing images of a work environment with an image sensor;

extracting characteristics of said images;

comparing the characteristics of said images to the characteristics of the objects in the catalogue; and

,

determining that an object likely to obstruct wheels is present if more than a predetermined percentage of characteristics in an image match with characteristics of an object in the catalogue.

8. The method of claim 7, further comprising actuating retraction of one or more wheels of the robotic floor-cleaning device when it has been determined that an object likely to obstruct wheels is present.

9. The method of claim 7, wherein a set of conditions for actuating retraction of one or more wheels includes any of:

a number of wheel rotations over a predetermined length of time falling below a first predetermined threshold, and a current measurement drawn by a wheel exceeding a second predetermined threshold and wherein, upon determining that an object likely to obstruct wheels is present, increasing the first predetermined threshold and decreasing the second predetermined threshold.

10. A robotic floor-cleaning device comprising:

a chassis including a set of wheels;

a wheel housing disposed within the chassis and over each wheel into which each wheel may be retracted, the wheel housing is designed with a minimal amount of clearance between the wheel and adjacent wheel housing walls such that no other objects except a corresponding wheel may fit within the wheel housing during retraction of the wheel;

at least one motor to drive the wheels;

a method for detecting obstructions on wheels;

a battery to provide power to the robotic floor-cleaning device; and

,

a controller in communication with the motor and wheels, the controller including a processor operable to control the motor and wheels to steer movement of the robotic device and actuate retraction and release of the wheels;

wherein upon detection of an obstruction on a wheel, the wheel is retracted into the wheel housing and any obstructions thereon are forcibly separated from and pushed off of the wheel by the wheel housing.

11. The robotic floor-cleaning device of claim 10 wherein the method for detecting obstructions on wheels includes any of:

counting a number of wheel rotations over a predetermined length of time compared to a predetermined minimum; and

, measuring a current drawn by a wheel compared to a predetermined maximum, wherein an obstruction is detected on a wheel if the number of wheel rotations over a predetermined length of time is less than the predetermined minimum, or if the current drawn by a wheel is greater than the predetermined maximum.

12. The robotic floor-cleaning device of claim 11 wherein at least one odometer is used to count wheel rotations.

13. The robotic floor-cleaning device of claim 11 wherein a unique odometer is used to count wheel rotations of each wheel and wheels are retracted individually when a corresponding odometer detects that wheel rotations over the predetermined length of time are below the predetermined minimum.

14. The robotic floor-cleaning device of claim 11 wherein unique wheel motors power each wheel and the amount of current drawn by the unique wheel motors is monitored, and wheels are retracted individually when a corresponding wheel motor detects that a current drawn by a wheel exceeds the predetermined maximum.

15. The robotic floor-cleaning device of claim 10 further comprising:

a catalogue of images of objects likely to obstruct wheels stored in a memory unit;

an image sensor for capturing images of an environment of the robotic floor-cleaning device; and

,

an image processing unit for extracting a first set of characteristics of the images of objects in the catalogue, extracting a second set of characteristics of captured images, comparing the second set of characteristics to the first set of characteristics, and determining that an object likely to obstruct wheels is present in the environment if more than a predetermined percentage of characteristics in the first and second set match.

16. The robotic floor-cleaning device of claim 15 wherein retraction of one or more wheels of the robotic floor-cleaning device is actuated when it has been determined that an object likely to obstruct wheels is present.

17. The robotic floor-cleaning device of claim 15 wherein the method for detecting obstructions on wheels includes any of:

counting a number of wheel rotations over a predetermined length of time compared to a predetermined minimum; and

, measuring a current drawn by a wheel compared to a predetermined maximum, wherein an obstruction is detected on a wheel if the number of wheel rotations over a predetermined length of time is less than the predetermined minimum, or if the current drawn by a wheel is greater than the predetermined maximum, and wherein, if it is determined that an object likely to obstruct wheels is present in the environment, the predetermined minimum is increased and the predetermined maximum is decreased.

18. A method for removing obstructions from a robotic floor-cleaning device comprising:

defining one or more sets of conditions to actuate retraction of one or more wheels of the robotic floor-cleaning device into a corresponding wheel housing positioned within a main housing of the robotic floor-cleaning device;

retracting one or more wheels into said corresponding wheel housing upon satisfaction of the one or more sets of conditions for a predetermined amount of time;

returning the one or more wheels to a standard position after the predetermined amount of time; and

proceeding with a robotic floor-cleaning process as normal;

wherein the wheel housing is designed with minimal amount of clearance between the wheel and adjacent wheel housing walls such that substantially no other objects except a corresponding wheel may fit within the wheel housing during retraction and obstructions are forcibly pushed off and separated from the one or more wheels of the robotic floor-cleaning device by the wheel housing during retraction.

19. The method of claim 18 wherein a set of conditions for actuating retraction of one or more wheels includes any of:

a number of wheel rotations over a predetermined length of time falling below a predetermined threshold, and a current measurement drawn by a wheel exceeding a predetermined threshold.

20. The method of claim 18, further comprising:

preloading a catalogue of objects likely to obstruct wheels into a database stored in a memory unit of the robotic floor-cleaning device;

extracting characteristics of the objects likely to obstruct wheels using an image processor;

iteratively capturing images of a work environment with an image sensor;

extracting characteristics of said images with the image processor;

comparing the characteristics of said images to the characteristics of the objects in the catalogue;

determining that an object likely to obstruct wheels is present if more than a predetermined percentage of characteristics in an image match with characteristics of an object in the catalogue; and

,

actuating retraction of one or more wheels of the robotic floor-cleaning device if it is determined that an object likely to obstruct wheels is present.