WO2017197197A1 - Systems and methods for initializing a robot to autonomously travel a trained route - Google Patents

Systems and methods for initializing a robot to autonomously travel a trained route Download PDF

Info

Publication number
WO2017197197A1
WO2017197197A1 PCT/US2017/032285 US2017032285W WO2017197197A1 WO 2017197197 A1 WO2017197197 A1 WO 2017197197A1 US 2017032285 W US2017032285 W US 2017032285W WO 2017197197 A1 WO2017197197 A1 WO 2017197197A1
Authority
WO
WIPO (PCT)
Prior art keywords
robot
initialization object
location
route
initialization
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US2017/032285
Other languages
English (en)
French (fr)
Inventor
Jean-Baptiste Passot
Jaldert ROMBOUTS
Cody GRIFFIN
John Black
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Brain Corp
Original Assignee
Brain Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Brain Corp filed Critical Brain Corp
Priority to KR1020187035946A priority Critical patent/KR102444658B1/ko
Priority to JP2019511832A priority patent/JP6942177B2/ja
Priority to EP17796891.4A priority patent/EP3455686A4/en
Priority to CA3023557A priority patent/CA3023557A1/en
Priority to CN201780041844.2A priority patent/CN109416538A/zh
Publication of WO2017197197A1 publication Critical patent/WO2017197197A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/0088—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/40—Control within particular dimensions
    • G05D1/43—Control of position or course in two dimensions
    • A—HUMAN NECESSITIES
    • A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L11/00—Machines for cleaning floors, carpets, furniture, walls, or wall coverings
    • A47L11/40—Parts or details of machines not provided for in groups A47L11/02 - A47L11/38, or not restricted to one of these groups, e.g. handles, arrangements of switches, skirts, buffers, levers
    • A47L11/4011—Regulation of the cleaning machine by electric means; Control systems and remote control systems therefor
    • A—HUMAN NECESSITIES
    • A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L11/00—Machines for cleaning floors, carpets, furniture, walls, or wall coverings
    • A47L11/40—Parts or details of machines not provided for in groups A47L11/02 - A47L11/38, or not restricted to one of these groups, e.g. handles, arrangements of switches, skirts, buffers, levers
    • A47L11/4061—Steering means; Means for avoiding obstacles; Details related to the place where the driver is accommodated
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J11/00—Manipulators not otherwise provided for
    • B25J11/008—Manipulators for service tasks
    • B25J11/0085—Cleaning
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J19/00—Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
    • B25J19/02—Sensing devices
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1602—Programme controls characterised by the control system, structure, architecture
    • B25J9/161—Hardware, e.g. neural networks, fuzzy logic, interfaces, processor
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1628—Programme controls characterised by the control loop
    • B25J9/163—Programme controls characterised by the control loop learning, adaptive, model based, rule based expert control
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
    • B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
    • B25J9/1697—Vision controlled systems
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
    • G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0268—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means
    • G05D1/0272—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means comprising means for registering the travel distance, e.g. revolutions of wheels
    • A—HUMAN NECESSITIES
    • A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L2201/00—Robotic cleaning machines, i.e. with automatic control of the travelling movement or the cleaning operation
    • A47L2201/04—Automatic control of the travelling movement; Automatic obstacle detection
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D2101/00—Details of software or hardware architectures used for the control of position
    • G05D2101/10—Details of software or hardware architectures used for the control of position using artificial intelligence [AI] techniques
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D2111/00—Details of signals used for control of position, course, altitude or attitude of land, water, air or space vehicles
    • G05D2111/10—Optical signals
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D2111/00—Details of signals used for control of position, course, altitude or attitude of land, water, air or space vehicles
    • G05D2111/50—Internal signals, i.e. from sensors located in the vehicle, e.g. from compasses or angular sensors

Definitions

  • the present application relates generally to, inter alia, robotics systems and methods of utilizing the same. Specifically, in one aspect, the present disclosure relates to systems and methods for initializing and operating a robot to travel a trained route.
  • Another challenge that can occur in programming robots to travel routes is initializing the robots such that the robots can determine relatively quickly their positions in environments. Being able to make such determinations can be important for robots to accurately navigate routes by informing the robots where they are located within the environment.
  • Conventional systems and methods of determining positions involve users and/or technicians initially programming starting locations, and/or having the robots start at the same spot every time. In some cases, these systems and methods can be time consuming to implement and/or may not have the robustness that enables robots to navigate environments in a user-friendly manner.
  • systems and methods for detecting a position relative to an initialization object are disclosed.
  • detection can facilitate a robot learning a demonstrated route, and then autonomously navigating the demonstrated route.
  • a robot in one aspect, includes: a camera configured to take a first image of an initialization object while the robot is in a first location and a second image of the initialization object while the robot is in a second location; an odometry unit configured to determine positions of the robot with respect to the initialization object; and a controller.
  • the controller is configured to: initialize the odometry unit relative to the initialization object determined at least in part from the first image while the robot is in the first location, learn a route by user demonstration, beginning from the first location, where the learned route associates actions of the robot with positions of the robot relative to the initialization object determined by the odometry unit, initialize the odometry unit relative to the initialization object determined at least in part from the second image while the robot is in the second location, and navigate the robot autonomously beginning from the second location through at least a portion of the learned route while instructing the robot to perform one or more associated actions based at least in part with positions of the robot relative to the initialization object determined by the odometry unit.
  • the odometry unit includes at least one of an accelerometer, inertial measurement unit, lidar, odometer, gyroscope, visual odometer, and speedometer.
  • the robot includes a user interface configured to prompt a user to initiate navigation of the robot autonomously through the learned route while the robot is in the second location.
  • the robot is a floor cleaner.
  • the robot includes a sensor configured to take a scan lidar image of a scene, wherein the controller is further configured to associate the scan lidar image to the initialization object and verify the initialization object using the scan lidar image.
  • a method for operating a robot includes: taking a first image of an initialization object while the robot is in a first location; determining a first starting position of the robot relative to the initialization object based at least in part on the first image while the robot is in the first location; learning a route by user demonstration, beginning from the first location, where the learned route associates actions of the robot with positions of the robot relative to the initialization object, wherein the positions are determined at least in part from the first starting position; taking a second image of the initialization object while the robot is in a second location; determining a second starting position of the robot relative to the initialization object based at least in part on the second image while the robot is in the second location; navigating the robot autonomously beginning from the second location through at least a portion of the learned route while instructing the robot to perform one or more associated actions based at least in part with positions of the robot relative to the initialization object, wherein the positions are determined
  • taking the first image and taking the second image each include generating data with a sensor and storing the data in a data structure.
  • the method further includes beginning autonomous navigation based at least in part on at least one of a user input and a characteristic of the initialization object.
  • the method further includes selecting the learned route from a plurality of routes stored in memory.
  • a non-transitory computer readable storage medium having a plurality of instructions stored thereon.
  • the instructions being executable by a processing apparatus to operate a robot, the instructions configured to, when executed by the processing apparatus, cause the processing apparatus to: determine a starting position of the robot relative to an initialization object based at least in part on an image; and learn a route by user demonstration, where the learned route associates actions of the robot with positions of the robot relative to the initialization object, wherein the positions are determined at least in part from the starting position.
  • the non-transitory computer-readable storage medium includes instructions that are further configured to, when executed by the processing apparatus, cause the processing apparatus to instruct the robot to autonomously navigate the learned route.
  • the non-transitory computer-readable storage medium includes instructions that are further configured to, when executed by the processing apparatus, cause the processing apparatus to recognize a predetermined sequence of actions and autonomously navigate the learned route based at least in part on the predetermined sequence of actions.
  • a robotic scrubber is disclosed.
  • a robotic scrubber is configured to autonomously navigate a first learned route based at least in part on detection of a first initialization object and autonomously navigate a second learned route based at least in part on detection of a second initialization object.
  • the first learned route and second learned route are demonstrated to the robotic scrubber by a user.
  • the detection of the first initialization object and the detection of the second initialization object are from an initialization location.
  • the detection of the first initialization object is from a first initialization location and the detection of the second initialization object is from a second initialization location, and the robotic scrubber is brought to the first initialization location and the second initialization under user control.
  • an initialization object is disclosed.
  • the initialization object is configured for detection by a robot.
  • the robot detects the initialization object, the robot initializes a position relative to the initialization object.
  • the initialization object is a binary image. In another variant, the initialization object is asymmetrical. In another variant, the initialization object is associated with at least one learned route. In another variant, the initialization object is configured to cause autonomous navigation of the robot when the robot detects the initialization object.
  • some implementations of this disclosure can include an autonomously navigating robot comprising: a camera configured to take a first image of an initialization object while the robot is in a first location and a second image of the initialization object while the robot is in a second location; an odometry unit configured to determine positions of the robot with respect to the initialization object; and a controller configured to: initialize the odometry unit relative to the initialization object determined at least in part from the first image while the robot is in the first location, learn a route by user demonstration, beginning from the first location, where the learned route associates actions of the robot with positions of the robot relative to the initialization object determined by the odometry unit, initialize the odometry unit relative to the initialization object determined at least in part from the second image while the robot is in the second location, and navigate the robot autonomously beginning from the second location through at least a portion of the learned route while instructing the robot to perform one or more associated actions based
  • the odometry unit comprises at least one of an accelerometer, inertial measurement unit, lidar, odometer, gyroscope, visual odometer, and speedometer.
  • the initialization object is a binary image. In some implementations, the initialization object is a picture.
  • the autonomously navigating robot further comprises a user interface configured to prompt a user to initiate navigation of the robot autonomously through the learned route while the robot is in the second location.
  • At least one of the one or more associated actions comprises a turning of the robot. In some implementations, at least one of the one or more associated actions comprises the activation of a switch configured to turn on a brush for the robot. In some implementations, the robot is a floor cleaner.
  • the first location and the second location are substantially similar.
  • the one or more associated actions are further associated with trajectories of the robot.
  • the autonomously navigating robot further comprises a sensor configured to take a scan lidar image of a scene, wherein the controller is further configured to associate the scan lidar image to the initialization object and verify the initialization object using the scan lidar image.
  • some implementations of this disclosure can include a method for operating a robot comprising: taking a first image of an initialization object while the robot is in a first location; determining a first starting position of the robot relative to the initialization object based at least in part on the first image while the robot is in the first location; learning a route by user demonstration, beginning from the first location, where the learned route associates actions of the robot with positions of the robot relative to the initialization object, wherein the positions are determined at least in part from the first starting position; taking a second image of the initialization object while the robot is in a second location; determining a second starting position of the robot relative to the initialization object based at least in part on the second image while the robot is in the second location; navigating the robot autonomously beginning from the second location through at least a portion of the learned route while instructing the robot to perform one or more associated actions based at least in part with positions of the robot relative to the initialization object, wherein the positions are determined at least in part from the second starting position
  • taking the first image and taking the second image each comprise generating data with a sensor and storing the data in a data structure.
  • the method for operating the robot further comprises beginning autonomous navigation based at least in part on at least one of a user input and a characteristic of the initialization object.
  • At least one of the one or more associated actions comprises turning the robot. In some implementations, at least one of the one or more associated actions comprises switching on a brush. In some implementations, navigating the robot autonomously further comprises navigating the robot autonomously until the robot detects the initialization object.
  • the method for operating the robot further comprise selecting the learned route from a plurality of routes stored in memory.
  • At least one of the one or more associated actions comprises associating actions of the robot with trajectories.
  • some implementations of this disclosure can include a non-transitory computer-readable storage medium having a plurality of instructions stored thereon, the instructions being executable by a processing apparatus to operate a robot, the instructions configured to, when executed by the processing apparatus, cause the processing apparatus to: determine a starting position of the robot relative to an initialization object based at least in part on an image; and learn a route by user demonstration, where the learned route associates actions of the robot with positions of the robot relative to the initialization object, wherein the positions are determined at least in part from the starting position.
  • the initialization object is a binary image.
  • the associated actions comprise turning.
  • the associated actions comprise switching on a brush.
  • the non-transitory computer-readable storage medium includes instructions that are further configured to, when executed by the processing apparatus, cause the processing apparatus to instruct the robot to autonomously navigate the learned route.
  • the non-transitory computer-readable storage medium includes instructions that are further configured to, when executed by the processing apparatus, cause the processing apparatus to recognize a predetermined sequence of actions and autonomously navigate the learned route based at least in part on the predetermined sequence of actions.
  • the instructions of the non-transitory computer- readable storage medium are further configured to, when executed by the processing apparatus: determine an action of the robot at a determined location of the robot; and execute the action; wherein the action comprises a turning action of the robot.
  • the instructions of the non-transitory computer- readable storage medium are further configured to, when executed by the processing apparatus: determine an action of the robot at a determined location of the robot; and execute the action; wherein the action comprises the activation of a switch configured to turn on a brush.
  • some implementations of this disclosure include a robotic scrubber configured to autonomously navigate a first learned route based at least in part on detection of a first initialization object and autonomously navigate a second learned route based at least in part on detection of a second initialization object.
  • the first learned route and second learned route are demonstrated to the robotic scrubber by a user.
  • the detection of the first initialization object and the detection of the second initialization object are detected from an initialization location.
  • the detection of the first initialization object is from a first initialization location and the detection of the second initialization object is from a second initialization location, and the robotic scrubber is brought to the first initialization location and the second initialization location under user control.
  • FIG. 1 A is an overhead view of an example route autonomously navigated by a robot in accordance with implementations of the present disclosure.
  • FIG. IB is an overhead view of the example route illustrated in FIG. 1A illustrating a user demonstrating the route to the robot in accordance with implementations of the present disclosure.
  • FIG. 1C is an overhead view of an alternative example route autonomously navigated by robot shown in FIGS. 1A and IB, where the robot avoids objects in accordance with some principles of the present disclosure.
  • FIG. ID is an overhead view of an example initial location that is a location on the floor for initializing a robot in accordance with some implementations of the present disclosure.
  • FIG. 2 is a process flow diagram of an exemplary method for training a robot to autonomously navigate a route in accordance with some implementations of the present disclosure.
  • FIG. 3A is a functional block diagram of an exemplary robot in accordance with some implementations of the present disclosure.
  • FIG. 3B is the functional block diagram from FIG. 3A with an example camera included as part of the exteroceptive sensor in accordance with some implementations of the present disclosure.
  • FIG. 3C is a side elevation view illustrating a robot detecting the initialization object from FIG. 3B in accordance with some principles of the present disclosure.
  • FIG. 4A illustrates various side elevation views of exemplary body forms for floor scrubbers in accordance with some principles of the present disclosure.
  • FIG. 4B illustrates various side elevation views of exemplary body forms for a robot in accordance with the principles of the present disclosure.
  • FIG. 5 illustrates various side elevation views of binary images that can be used as initialization objects in accordance with some principles of the present disclosure.
  • FIG. 6 is a process flow diagram of an exemplary method in which a robot can be initialized to autonomously navigate a learned route in accordance with some implementations of the present disclosure.
  • FIG. 7 is an overhead view of a robot positioned at an angle to an initialization object, where a sensor of the robot measures a distance and angle from an example point on the sensor to a point on the initialization object in accordance with some implementations of the present disclosure.
  • FIG. 8 is a side elevation view of an image representation of an initialization object in accordance with some implementations of the present disclosure.
  • FIG. 9A is an example user interface for receiving an input from a user to begin learning a route or to choose a route for navigation in accordance to some principles of the present disclosure.
  • FIG. 9B is a side elevation view of a user controlling a robot while teaching a route in accordance with some principles of the present disclosure.
  • FIG. 10 is an example user interface, which can be used for route selection in accordance with some principles of the present disclosure.
  • FIG. 11 is an overhead view of an environment with a plurality of initialization objects in accordance with some principles of the present disclosure.
  • FIG. 12 is a process flow diagram of an exemplary method for operating a robot in accordance with the principles of the present disclosure.
  • the present disclosure provides for improved systems and methods of operating a robot for autonomous navigation. More specifically, the present disclosure includes systems and methods for initializing the location of a robot for training and/or for autonomous navigation.
  • a robot can include mechanical or virtual entities configured to carry out complex series of actions automatically.
  • robots can be electro-mechanical machines that are guided by computer programs or electronic circuitry.
  • robots can include electro-mechanical machines that are configured for autonomous navigation, where the robot can move from one location to another with little to no user control.
  • Such autonomously navigating robots can include autonomous cars, floor cleaners (e.g., floor scrubbers, vacuums, etc.), rovers, drones, humanoid robots, and the like.
  • some of the systems and methods described in this disclosure can be implemented to a virtual environment, where a virtual robot can detect an initialization object in a simulated environment (e.g., in a computer simulation) with characteristics of the physical world. After detecting the initialization object, the robot can learn routes and then autonomously navigate the learned routes in the simulated environment and/or in the real world using systems and methods disclosed in this disclosure.
  • a virtual robot can detect an initialization object in a simulated environment (e.g., in a computer simulation) with characteristics of the physical world. After detecting the initialization object, the robot can learn routes and then autonomously navigate the learned routes in the simulated environment and/or in the real world using systems and methods disclosed in this disclosure.
  • the systems and methods of this disclosure at least: (i) reduce the need for environment-specific programming; (ii) reduce the need for highly skilled technicians to program robots; (iii) provide application-specific performance from a generally programmed robot; (iv) enable effective autonomous navigation of robots; and (v) provide a fast and user-friendly approach to initializing a robot for autonomous navigation.
  • Other advantages are readily discernable by one of ordinary skill given the contents of the present disclosure.
  • a user may not have to program every route beforehand.
  • this can allow a user to train a robot to navigate environments that the user had not anticipated beforehand.
  • a user may not utilize any particular expertise to train the robot.
  • a user may not have to know computer science and/or be educated on how to program the robot.
  • a user may just know how to perform the task that he/she desires the robot to do. For example, where the robot is a floor cleaner, the user may just know how to clean the floor, which he/she can demonstrate to the robot.
  • training a robot that can learn a navigable route can allow a robot to be specifically programmed to efficiently navigate a particular environment while also being generally programmed to perform in many environments.
  • this allows such robots to have the benefit of both being optimized in particular applications, yet having the ability, and flexibility, to perform in a variety of applications.
  • a robot can initialize to an initialization object, which can facilitate its route learning and autonomous navigation.
  • the disclosed initialization relative to the initialization objects presents marked improvements in the fields of robotics and/or autonomous navigation in that it allows the robot to determine accurately where it is in its environment and/or its trajectory and to perform actions based at least in part on that positioning and/or trajectory.
  • having a fast and effective way to initialize can allow a robot to associate actions to positions along its learned route. The robot can later replicate those actions when it autonomously navigates the learned route.
  • some implementations of this disclosure enable robots to navigate and/or perform actions relative to fixed locations, which can provide consistency and/or replicability. Consistency and/or replicability can be desirable in many industries, such as floor cleaning, where a robot performs substantially similar tasks repeatedly.
  • initialization objects can be symbols, images, shapes, items (e.g., furniture, statues, appliances, etc.), and/or any other objects.
  • a user can then use that initialization object to begin robotic route learning and/or begin the robot's autonomous navigation.
  • Having a visible initialization object can, in some cases, provide a user a sense or predictability in operating the robot, where the user can have the visible feedback of bringing the robot to the initialization object and seeing the robot behave in a predictable way.
  • there can be a plurality of initialization objects where particular initialization objects can be associated with particular routes. This can enable a user to visually choose a route based on the initialization object, which can be advantageous especially in environments where operators may not be able to understand a particular language or use complex user interfaces.
  • a robot may learn to navigate a plurality of environments that are substantially similar using a plurality of routes. Accordingly, using an initialization object can facilitate navigation of particular environments by filtering and/or narrowing the number of routes and/or environments from which the robot may choose. This can reduce the chances a robot and/or a user incorrectly chooses a route, such as by choosing a route associated with a different environment.
  • a building can have a plurality of floors that are substantially similar, but with some variation. Each floor can have its own initialization object.
  • the robot initializes from the initialization object, it can filter the routes that it can navigate based at least in part on that initialization object, which can be associated with only a floor and/or particular routes.
  • present robotic navigation may involve a plurality of robotically detectable symbols in which the robot moves towards each of those symbols.
  • having too many symbols around can produce a negative perception to customers.
  • Having an initialization object, which can include symbols, images, shapes, items, etc. that may or may not already exist in the environment, can be advantageous in that a customer may not readily notice the initialization objects and/or be distracted by them.
  • a robot can initialize from fewer symbols, in some case a single symbol, and be able to autonomously navigate.
  • FIG. 1A illustrates an overhead view of an example route 106 autonomously navigated by robot 102 through implementations of this disclosure.
  • Robot 102 can autonomously navigate through environment 100, which can comprise various objects 108, 110, 112, 118.
  • Robot 102 can start at an initial location and end at end location 114.
  • the initial location can be determined by its relationship to initialization object 150.
  • the initial location can be any position in which robot 102 can detect initialization object 150.
  • a user may bring robot 102 to a particular place on the floor (e.g., a place demarcated and/or remembered by the user), as illustrated in FIG. ID, where initial location 104 is a position on the floor.
  • Bringing robot 102 to a particular place on the floor can be advantageous in that it can make it very likely that robot 102 can detect initialization object 150 and initialize.
  • Bringing robot 102 to a particular place on the floor, e.g., initial location 104 can also be helpful for a user to have a sense of routine and predictability in operating robot 102.
  • Initialization object 150 can include tangible objects (e.g., furniture, statues, appliances, etc.), markings, and/or digital displays.
  • initialization object 150 can be detected and/or recognized by robot 102 using at least in part exteroceptive sensor 300, which will be discussed more with reference to FIGS. 3A- B, as well as elsewhere throughout this disclosure.
  • robot 102 can determine robot 102's position relative to initialization object 150 (and/or a point therein), which can allow, among other things, robot 102 to initialize robot 102's odometry, such as odometry using at least in part an odometry unit 326, proprioceptive sensor 308, and other sensors which will be discussed more with reference to FIGS. 3A - 3B as well as elsewhere throughout this disclosure.
  • position has its ordinary and customary meaning. For example, in some cases, position can include a location in terms of displacement, coordinates, etc. of an object, robot 102, etc.
  • position can also include an orientation of an object, robot 102, etc. Accordingly, in some cases, the terms position and pose may be used interchangeably to include one or more of location, displacement, and orientation.
  • determining robot 102's position can enable robot 102 to use initialization object 150 (and/or a point therein) as a reference point for robot 102's odometry and/or sensors as robot 102 proceeds with learning a route, such as learning route 116 in environment 100 illustrated in FIG. IB. This use of initialization object 150 (and/or a point therein) as a reference point can later be again used by robot 102 in autonomously navigating learned route 116 as route 106.
  • Allowing robot 102 to initialize from an initialization object 150 can, in some implementations, reduce or eliminate the use of unattractive symbols.
  • robot 102 can be a robotic floor cleaner, such as a robotic floor scrubber, vacuum cleaner, steamer, mop, sweeper, and the like.
  • Environment 100 can be a space having floors that are desired to be cleaned.
  • environment 100 can be a store, warehouse, office building, home, storage facility, etc.
  • objects 108, 110, 112, 118 can be shelves, displays, objects, items, people, animals, or any other entity or thing that may be on the floor or otherwise imped the robot's ability to navigate through the environment.
  • Route 106 can be the cleaning path traveled by robot 102 autonomously. Route 106 can follow a path that weaves between objects 108, 110, 112, 118 as illustrated in example route 106.
  • robot 102 can go along the aisles of the store and clean the floors of the aisles.
  • routes are also contemplated, such as, without limitation, weaving back and forth along open floor areas and/or any cleaning path a user could use to clean the floor (e.g., if the user is manually operating a floor cleaner).
  • routes 106, 116, 126, illustrated in FIGS. 1A, IB, and 1C, respectively are meant merely as illustrative examples and can appear differently as illustrated.
  • robot 102 can begin at the initial location where it detects initialization object 150, which can be robot 102's starting point. Robot 102 can then clean along route 106 autonomously (e.g., with little or no control from user 904) until it reaches end location 114, where it can stop cleaning. End location 114 can be designated by a user 904, later described with reference to FIG. 9B.
  • end location 114 can be the location in route 106 after which robot 102 has cleaned the desired area of floor. In some cases, end location 114 can be the same, or substantially similar, as the initial location so that robot 102 performs substantially a closed loop in cleaning and ends up near its starting point at the initial location. In some cases, end location 114 can be a location for storage for robot 102, such as a temporary parking spot, storage room/closet, and the like. In some cases, end location 114 can be the point where user 904 decided to stop performing actions and training robot 102. For example, in the context of floor cleaners (e.g., floor scrubbers, vacuum cleaners, etc.), robot 102 may or may not clean at every point along route 106.
  • floor cleaners e.g., floor scrubbers, vacuum cleaners, etc.
  • robot 102 may only be operating in some portions of route 106 and not others.
  • robot 102 may associate certain actions (e.g., turning, turning on/off water, spraying water, turning on/off vacuums, moving vacuum hose positions, gesticulating an arm, raising/lowering a lift, moving a sensor, turning on/off a sensor, etc.) with particular positions and/or trajectories (e.g., while moving in a certain direction or in a particular sequence along route 106) along the demonstrated route.
  • certain actions e.g., turning, turning on/off water, spraying water, turning on/off vacuums, moving vacuum hose positions, gesticulating an arm, raising/lowering a lift, moving a sensor, turning on/off a sensor, etc.
  • particular positions and/or trajectories e.g., while moving in a certain direction or in a particular sequence along route 106 along the demonstrated route.
  • robot 102 can turn on a cleaning system in areas where user 904 demonstrated for robot 102 to clean, and turn off the cleaning system otherwise.
  • FIG. IB illustrates an overhead view of example user 904 demonstrating route 116 to example robot 102 before example robot 102 autonomously travels example route 106 in environment 100.
  • a user can start robot 102 at the initial location (e.g., the initial location relative to initialization object 150 and/or within initial location 104).
  • Robot 102 can then weave around objects 108, 110, 112, 118.
  • Robot 102 can finally end at end location 114.
  • route 106 can be exactly the same as route 116.
  • route 106 might not be precisely the same as route 116, but can be substantially similar.
  • robot 102 uses its sensors (e.g., sensors 300, 308 as will be described with reference to FIGS. 3 A - 3B and elsewhere throughout this disclosure) to sense where robot 102 is in relationship to robot 102's surroundings as it navigates. Such sensing may be imprecise in some instances, which may cause at least in part robot 102 to not navigate the precise route 116 demonstrated and that robot 102 was trained to follow. In some cases, changes to environment 100, such as the moving of shelves and/or changes in the items on the shelves, can cause robot 102 to deviate from route 116 when it autonomously navigates. As another example, as illustrated in FIG.
  • sensors 300, 308 as will be described with reference to FIGS. 3 A - 3B and elsewhere throughout this disclosure
  • 1C robot 102 may avoid objects 130, 132 by turning around them when autonomously navigating route 126, which can be another route travelled by robot 102 based at least in part on demonstrated route 116.
  • Objects 130, 132 might not have been present (and/or avoided) when the user demonstrated route 116.
  • objects 130, 132 may be temporarily placed and/or transient objects/items, and/or transient and/or dynamic changes to the environment 100.
  • user 904 may have done a poor job demonstrating route 116.
  • user 904 may have crashed and/or bumped into a wall, shelf, object, obstacle, etc.
  • robot 102 can store in memory (e.g., memory 302) one or more actions that it can correct, such as crashing and/or bumping to a wall, shelf, object, obstacle, etc.
  • memory e.g., memory 302
  • robot 102 can correct such actions and/or not perform them (e.g., not crash and/or bump into a wall, shelf, object, obstacle, etc.) when it is autonomously navigating.
  • user 904 can perform one or more actions that can be also demonstrated to and learned by robot 102 as robot 102 learns to navigate. These actions can include any actions that robot 102 may perform, such as turning, turning on/off water, spraying water, turning on/off vacuums, moving vacuum hose positions, gesticulating an arm, raising/lowering a lift, moving a sensor, turning on/off a sensor, etc.
  • user 904 can turn on and off the cleaning system of robot 102 in order to train robot 102 where to clean along route 116 (and subsequently where to clean when robot 102 autonomously cleans routes 106, 126) and/or in what trajectories.
  • Robot 102 can record these actions in memory 302 (later described in FIG. 3A) and later perform them when autonomously navigating. For example, in some implementations, robot 102 can associate one or more positions relative to initialization object 150 with the learned actions.
  • FIG. 2 illustrates a process flow diagram of an example method 200 for training robot 102 to autonomously navigate a route (e.g., route 106, 126).
  • Portion 202 can include positioning robot 102 to detect initialization object 150. This first placement of robot 102 into the initial location relative to initialization object 150 can be performed by user 904, who can be a janitor, custodian, and/or any other person or robot, who drives, remote controls, pushes, or otherwise controls robot 102 to a location (e.g., an initial location) to detect initialization object 150.
  • a location e.g., an initial location
  • Portion 204 can include demonstrating a navigation route to robot 102.
  • user 904 can demonstrate to robot 102 by, without limitation, driving, remote controlling, pushing, or otherwise controlling robot 102 along route 116. In this way, user 904 can demonstrate to robot 102 the desired route for travelling.
  • user 904 can demonstrate to robot 102 actions (e.g., turning, turning on/off water, spraying water, turning on/off vacuums, moving vacuum hose positions, gesticulating an arm, raising/lowering a lift, moving a sensor, turning on/off a sensor, etc.) to perform while travelling the route.
  • actions e.g., turning, turning on/off water, spraying water, turning on/off vacuums, moving vacuum hose positions, gesticulating an arm, raising/lowering a lift, moving a sensor, turning on/off a sensor, etc.
  • these actions can be associated with positions and/or trajectories along the learned route, such as positions and/or trajectories relative to initialization object 150.
  • demonstrated route 116 can be the desired route for cleaning the floor. In this way, user 904 can train robot 102 how to clean the floor.
  • Portion 206 can include positioning robot 102 to detect initialization object
  • This second placement of robot 102 to detect initialization object 150 can occur at a later point in time after portion 204, such as substantially right after the demonstration of portion 204, or at some later time, such as hours later, days later, weeks later, or whenever the user 904 desires to clean the floor.
  • Portion 208 can include initiating autonomous navigation.
  • robot 102 can travel along route 106 (or route 126 in some cases), which can be substantially similar to demonstrated route 116.
  • user 904 can select the demonstrated route on user interface 318, as will be described with reference to FIG. 10 as well as elsewhere throughout this disclosure.
  • robot 102 can then navigate route 106, (or a route substantially similar to route 106) autonomously from the initial location to end location 114.
  • FIG. 3 A illustrates a functional block diagram of example robot 102 in some implementations.
  • robot 102 can include controller 304, memory 302, user interface 318, exteroceptive sensor 300, odometry unit 308, proprioceptive sensor 308, as well as other components and subcomponents (e.g., some of which may not be illustrated).
  • controller 304 memory 302
  • user interface 318 user interface 318
  • exteroceptive sensor 300 e.g., odometry unit 308
  • proprioceptive sensor 308 e.g., some of which may not be illustrated.
  • FIG. 3A a specific implementation is illustrated in FIG. 3A, it is appreciated that the architecture may be varied in certain implementations as would be readily apparent to one of ordinary skill given the contents of the present disclosure.
  • Controller 304 can control the various operations performed by robot 102.
  • Controller 304 can include one or more processors (e.g., microprocessors) and other peripherals.
  • processors e.g., microprocessors
  • processor, microprocessor, and/or digital processor can include any type of digital processing device such as, without limitation, digital signal processors ("DSPs"), reduced instruction set computers (“RISC”), general-purpose (“CISC”) processors, microprocessors, gate arrays (e.g., field programmable gate arrays (“FPGAs”)), programmable logic device (“PLDs”), reconfigurable computer fabrics (“RCFs”), array processors, secure microprocessors, specialized processors (e.g., neuromorphic processors), and application-specific integrated circuits (“ASICs”).
  • DSPs digital signal processors
  • RISC reduced instruction set computers
  • CISC general-purpose
  • microprocessors e.g., gate arrays (e.g., field programmable gate arrays (“FPGAs”)), programmable logic device (“PLDs”)
  • Controller 304 can be operatively and/or communicatively coupled to memory 302.
  • Memory 302 can include any type of integrated circuit or other storage device configured to store digital data including, without limitation, read-only memory (“ROM”), random access memory (“RAM”), non-volatile random access memory (“NVRAM”), programmable read-only memory (“PROM”), electrically erasable programmable read-only memory (“EEPROM”), dynamic random-access memory (“DRAM”), Mobile DRAM, synchronous DRAM (“SDRAM”), double data rate SDRAM (“DDR/2 SDRAM”), extended data output RAM (“EDO”), fast page mode RAM (“FPM”), reduced latency DRAM (“RLDRAM”), static RAM (“SRAM”), “flash” memory (e.g., NAND/NOR), memristor memory, pseudostatic RAM (“PSRAM”), etc.
  • ROM read-only memory
  • RAM random access memory
  • NVRAM non-volatile random access memory
  • PROM programmable read-only memory
  • EEPROM electrically
  • Memory 302 can provide instructions and data to controller 304.
  • memory 302 can be a non-transitory, computer-readable storage medium having a plurality of instructions stored thereon, the instructions being executable by a processing apparatus (e.g., controller 304) to operate robot 102.
  • the instructions can be configured to, when executed by the processing apparatus, cause the processing apparatus to perform the various methods, features, and/or functionality described in this disclosure.
  • controller 304 can perform logical and arithmetic operations based on program instructions stored within memory 302.
  • memory 302 can store a library 324 of images of initialization object 150.
  • this library 324 can include images of initialization object 150 and/or substantially similar objects to initialization object 150 in different lighting conditions, angles, sizes, distances, clarity (e.g., blurred, obstructed/occluded, partially off frame, etc.), colors, surroundings, etc.
  • the images in library 324 can be taken by a sensor (e.g., camera 304 or any other camera) or generated automatically, such as with a computer program that is configured to generate/simulate (e.g., in a virtual world) library images of initialization object 150 or substantially similar objects (e.g., which can generate/simulate these library images entirely digitally or beginning from an actual image of initialization object or substantially similar objects) from different lighting conditions, angles, sizes, distances, clarity (e.g., blurred, obstructed/occluded, partially off frame, etc.), colors, surroundings, etc.
  • Library 324 can be used to train robot 102 to identify initialization object 150 in many conditions and/or to identify the position of robot 102 as will be discussed more with reference to FIG.
  • the number of images in library 324 can depend at least in part on one or more of the number of available images of initialization object 150 and/or substantially similar objects, the variability of the surrounding environment in which robot 102 will operate, the complexity of initialization object 150 and/or substantially similar objects, the variability in appearance of initialization object 150 and/or substantially similar objects, and the amount of available storage space (e.g., in memory 302 or on a server).
  • library 324 can contain approximately 1, 5, 10, 100, 1000, 10,000, 100,000, 1,000,000, 10,000,000, or any number of images of initialization object 150 and/or substantially similar objects.
  • library 324 may be stored in a network (e.g., cloud, server, etc.) and may not be saved within memory 302.
  • exteroceptive sensor 300 can comprise systems and/or methods that can detect characteristics within and/or around robot 102.
  • Exteroceptive sensor 300 can comprise a plurality and/or a combination of sensors.
  • Exteroceptive sensor 300 can include sensors that are internal to robot 102 or external, and/or have components that are partially internal and/or partially external.
  • exteroceptive sensor 300 can include exteroceptive sensors such as sonar, lidar, radar, lasers, cameras (including video cameras, infrared cameras, 3D cameras, etc.), antenna, microphones, and/or any other sensor known in the art.
  • exteroceptive sensor 300 can collect raw measurements (e.g., currents, voltages, resistances gate logic, etc.) and/or transformed measurements (e.g., distances, angles, detected points in obstacles, etc.). In some implementations, exteroceptive sensor 300 can be configured to detect initialization object 150. Exteroceptive sensor 300 can generate data based at least in part on measurements. Such data can be stored in data structures, such as matrices, arrays, etc. In some implementations, the data structure of the sensor data can be called an image.
  • odometry unit 326 can be configured to determine the odometry of robot 102.
  • odometry unit 326 can include proprioceptive sensor 308, which can comprise sensors such as accelerometers, inertial measurement units ("IMU"), odometers, gyroscopes, speedometers, cameras (e.g. using visual odometry), clock/timer, and the like.
  • Proprioceptive sensor 308 can be used by odometry unit 326 for odometry to facilitate autonomous navigation of robot 102.
  • This odometry can include robot 102's position (e.g., where position includes robot's location, displacement and/or orientation, and can sometimes be interchangeable with the term pose as used herein) relative to the initial location (e.g., the initial location relative to initialization object 150 and/or within initial location 104).
  • proprioceptive sensor 308 can collect raw measurements (e.g., currents, voltages, resistances gate logic, etc.) and/or transformed measurements (e.g., distances, angles, detected points in obstacles, etc.).
  • Such data can be stored in data structures, such as matrices, arrays, etc.
  • the data structure of the sensor data can be called an image.
  • user interface 318 can be configured to enable a user (e.g., user 904) to interact with robot 102.
  • user interfaces 318 can include touch panels, buttons, keypads/keyboards, ports (e.g., universal serial bus ("USB"), digital visual interface ("DVI”), Display Port, E-Sata, Firewire, PS/2, Serial, VGA, SCSI, audioport, high-definition multimedia interface (“HDMI”), personal computer memory card international association (“PCMCIA”) ports, memory card ports (e.g., secure digital (“SD”) and miniSD), and/or ports for computer-readable medium), mice, rollerballs, consoles, vibrators, audio transducers, and/or any interface for a user to input and/or receive data and/or commands, whether coupled wirelessly or through wires.
  • USB universal serial bus
  • DVI digital visual interface
  • Display Port Display Port
  • E-Sata Firewire
  • PS/2 Serial, VGA, SCSI
  • HDMI high-definition multimedia interface
  • PCMCIA personal
  • User interface 318 can include a display, such as, without limitation, liquid crystal display (“LCDs”), light-emitting diode (“LED”) displays, LED LCD displays, in-plane-switching (“IPS”) displays, cathode ray tubes, plasma displays, high definition (“HD”) panels, 4K displays, retina displays, organic LED displays, touchscreens, surfaces, canvases, and/or any displays, televisions, monitors, panels, and/or devices known in the art for visual presentation.
  • LCDs liquid crystal display
  • LED light-emitting diode
  • IPS in-plane-switching
  • cathode ray tubes plasma displays
  • HD high definition
  • 4K displays retina displays
  • organic LED displays organic LED displays
  • touchscreens touchscreens
  • canvases and/or any displays, televisions, monitors, panels, and/or devices known in the art for visual presentation.
  • user interface 318 can be positioned on the body of robot 102.
  • user interface 318 can be positioned away from the body of robot 102, but can be communicatively coupled to robot 102 (e.g., via communication units including transmitters, receivers, and/or transceivers) directly or indirectly (e.g., through a network, server, and/or a cloud).
  • the wireless coupling can include wireless transmissions configured to send/receive a transmission protocol, such as BLUETOOTH®, ZIGBEE®, Wi-Fi, induction wireless data transmission, radio frequencies, radio transmission, radio- frequency identification (“RFID”), near-field communication ("NFC”), global system for mobile communications ("GSM”), infrared, network interfaces, cellular technologies such as 3G (3GPP/3GPP2), high-speed downlink packet access (“HSDPA”), high-speed uplink packet access (“HSUPA”), time division multiple access (“TDMA”), code division multiple access (“CDMA”) (e.g., IS-95A, wideband code division multiple access (“WCDMA”), etc.), frequency hopping spread spectrum (“FHSS”), direct sequence spread spectrum (“DSSS”), global system for mobile communication (“GSM”), Personal Area Network (“PAN”) (e.g., PAN/802.15), worldwide interoperability for microwave access (“WiMAX”), 802.20, long term evolution (“LTE”) (e.
  • networks, servers, and/or clouds can include network interfaces.
  • Network interfaces can include any signal, data, or software interface with a component, network, or process including, without limitation, those of the FireWire (e.g., FW400, FW800, FWS800T, FWS1600, FWS3200, etc.), universal serial bus (“USB”) (e.g., USB 1.X, USB 2.0, USB 3.0, USB Type-C, etc.), Ethernet (e.g., 10/100, 10/100/1000 (Gigabit Ethernet), 10-Gig-E, etc.), multimedia over coax alliance technology ("MoCA”), Coaxsys (e.g., TVNETTM), radio frequency tuner (e.g., in-band or OOB, cable modem, etc.), Wi-Fi (802.11), WiMAX (e.g., WiMAX (802.16)), PAN (e.g., PAN/802.15), cellular (e.g., 3G, LTE/
  • FireWire
  • Wi-Fi can include one or more of IEEE-Std. 802.11, variants of IEEE-Std. 802.11, standards related to IEEE-Std. 802.11 (e.g., 802.1 1 a/b/g/n/ac/ad/af/ah/ai/aj/aq/ax/ay), and/or other wireless standards.
  • Wired coupling can include wired connections, such as any cable that has a signal line and ground.
  • cables can include Ethernet cables, coaxial cables, Universal Serial Bus ("USB"), Firewire, and/or any wired connection known in the art.
  • USB Universal Serial Bus
  • Such protocols can be used by robot 102 to communicate to internal systems (e.g., communications between any components and/or subcomponents of robot 102) and/or external systems (e.g., computers, smart phones, tablets, data capture systems, mobile telecommunications networks, clouds, servers, and/or the like).
  • exteroceptive sensor 300 includes camera 306.
  • FIG. 3B illustrates the functional block diagram from FIG. 3A with example camera 306 included in example exteroceptive sensor 300.
  • Camera 306 can be an optical instrument for recording images. In some cases, images can be still photographs, or a sequence of images (e.g., frames) forming a video or movie.
  • light can enter an enclosed box (e.g., the body of camera 306) through a lens 314.
  • the light can be recorded on a light-sensitive medium as an image.
  • the medium can store the image digitally, in some cases, or in analog form.
  • a shutter mechanism of camera 306 can control the length of time that light can enter Camera 306.
  • Camera 306 can take a series of such images in succession, such as 24 or more frames per second. Each image taken can comprise pixels.
  • camera 306 can be a red, green, blue (“RGB”) camera, which can be configured to sense three basic color components (e.g., red, green, and blue). Light received can be exposed onto imaging sensors 312, which can comprise charge-coupled devices ("CCDs"), complementary metal-oxide semiconductor (“CMOS”), or other materials. Imaging sensors 312 can convert the light into electrons (e.g., accumulated charge) for each pixel of an image, storing the image.
  • Camera 306 can also be other cameras, such as cameras that can measure depth, including ranging cameras, flash lidars, time-of-flight (“ToF”) cameras, and RGB-Depth (“RGB-D”) cameras. Camera 306 can also comprise other specialized cameras for taking images outside of the visible spectrum, such as infrared cameras, hyperspectral cameras, bi-spectral cameras and other cameras that can image and/or process energy outside of the visible spectrum.
  • Camera 306 can have a number of physical and performance properties/characteristics.
  • camera 306 can comprise lens 314.
  • Lens 314 can comprise a wide angle lens, standard lens, medium telephoto/portrait lens, telephoto lens, macro lens, fisheye lens, and/or any other lens known in the art.
  • Lens 314 can have a characteristic focal length 316.
  • Focal length 316 can include the distance between lens 314 and image sensor 312 when a subject (e.g., initialization object 150) is in focus. In many cases, focal length 316 can be expressed in millimeters (e.g., 15 mm, 28 mm, 30 mm, 50 mm, 100 mm, 150mm, 210 mm, 300 mm, or any focal length for a lens).
  • focal length 314 can be adjusted across a range (e.g., 10 mm to 100mm, etc.).
  • Horizontal angle of view 310 can be the maximum view camera 306 can view through lens 314, and is expressed in degrees. In some cases, the angle of view 310 can change with focal length 316. Horizontal angle of view can be measured in degrees, such as 6, 12, 34, 47, 63, 90, 94, 180 and other degrees.
  • camera 306 can have other angles of view, such as angles of view measured in the vertical and diagonal directions, along with horizontal angle of view 310. Together, these angles of view can form a conical, or other, shape shown as field of view 320, an example of which is illustrated in FIG. 3C.
  • Lens 314 can also have other properties/characteristics, such as an optical center, which can be the point on the principal axis of lens 314 through which light passes without undergoing any deviation.
  • Lens 314 can also have lens distortion effects where images taken by camera 306 using lens 314 can appear to bend and/or have other deviations from their natural appearance.
  • Lens distortion effects can have radial and tangential components. For example, radial distortions can be due to light rays further from the center of lens 314 bending too much compared to light rays that pass closer to the center. Tangential distortion can be due to defects resulting from lens 314 not being exactly parallel to the imaging plane (e.g., the projective plane).
  • Camera 306 can also have one or more filters that narrow received light to one or more spectral bands.
  • camera 306 can have an infrared filter that allows cameras 306 to image light in the infrared spectrum and may reflect light in other spectrums.
  • FIG. 3C illustrates a side view of the example robot 102 detecting an initialization object 150.
  • robot 102 can be a floor cleaner (e.g., floor scrubber, vacuum cleaner, sweeper, etc.) configured to clean the floor of a store, warehouse, office building, home, storage facility, etc.
  • robot 102 can have brush 322 configured to clean the floor beneath and/or around robot 102. Brush 322 can be turned on/off by a switch.
  • FIG. 4A illustrates example body forms for a floor scrubber. These are non-limiting examples meant to further illustrate the variety of body forms, but not to restrict robot 102 to any particular body form or even to a floor scrubber.
  • Example body form 402 has an upright shape with a small frame where a user can push behind body form 402 to clean a floor. In some cases, body form 402 can have motorized propulsion that can assist a user in cleaning, but can also allow for autonomous movement of body form 402.
  • Body form 404 has a larger structural shape than body form 402.
  • Body form 404 can be motorized enabling it to move with little to no user exertion upon body form 404 besides steering.
  • the user may steer body form 404 as it moves.
  • Body form 406 can include a seat, pedals, and a steering wheel, where a user can drive body form 406 like a vehicle as body form 406 cleans.
  • Body form 408 can have a shape that is larger than body form 406 and can have a plurality of brushes.
  • Body form 410 can have a partial or fully encased area where a user sits as he/she drives body form 410.
  • Body form 412 can have a platform where a user stands while he/she drives body form 412.
  • FIG. 4B illustrates some additional examples of body forms of robot 102.
  • body form 414 illustrates an example where robot 102 is a stand-up shop vacuum.
  • Body form 416 illustrates an example where robot 102 is a humanoid robot having an appearance substantially similar to a human body.
  • Body form 418 illustrates an example where robot 102 is a drone having propellers.
  • Body form 420 illustrates an example where robot 102 has a vehicle shape having wheels and a passenger cabin.
  • Body form 422 illustrates an example where robot 102 is a rover.
  • initialization object 150 in this illustrative example is a sign having an image, such as a distinct picture, positioned on wall 350.
  • initialization object 150 is not limited to images, but can include symbols, images, shapes, items (e.g., furniture, statues, appliances, etc.), and/or any other objects.
  • Initialization object 150 can also include tangible objects, such as chairs, columns, ornaments, etc.
  • Initialization object 150 can include human-readable symbols, such as text.
  • the text can describe characteristics of the initialization object 150 such as the location, an alphanumeric associated with the location (e.g., including locations 1,2,3,4, locations A,B,C,D, locations 1A, 1B, 1C, etc.), a description of the route associated with initialization object 150, and/or any other characteristic and/or aspect of initialization object 150.
  • initialization object 150 can be a screen.
  • the screen can include a display, such as, without limitation, LCDs, LED displays, LED LCD displays, IPS displays, cathode ray tubes, plasma displays, HD panels, 4K displays, retina displays, organic LED displays, touchscreens, surfaces, canvases, screen ink technology, and/or any displays, televisions, monitors, panels, and/or devices known in the art for visual presentation.
  • a screen can display an image that can be changed. This ability to change can give flexibility for a user to change routes and/or other actions initialized by initialization object 150.
  • the image displayed by initialization object 150 can correspond to a particular route.
  • a user 904 can select the image corresponding to a route on the screen of initialization object 150.
  • robot 102 can detect the image on the screen.
  • robot 102 can associate the route with the image it detected on the screen (e.g., that a user 904 created and/or chose from a plurality of routes).
  • robot 102 can recall the route associated with the image detected on the screen (e.g., that a user 904 created and/or chose from a plurality of routes).
  • the screen is a touch screen
  • a user 904 can draw the image on the screen.
  • user 904 can illustrate any symbols, images, etc. that he/she desires to associate with a learned route, and can later recall the route by later selecting or redrawing that substantially similar symbol, images, etc.
  • initialization object 150 can be an object that robot 102 can recognize through sensor 300.
  • initialization object 150 can be instantiated in the visible spectrum (e.g., visible by a human), as with the case where camera 306 is an RGB camera or it can be instantiated outside of the visible spectrum (e.g., not visible by humans), where camera 306, and/or sensor 300, can detect initialization object 150 using infrared cameras, hyperspectral cameras, and/or other cameras and/or sensors that can image or detect energy outside of the visible spectrum.
  • initialization object 150 can transmit a signal, such as a radio transmission (e.g., BLUETOOTH®) or other signal protocol to enable detection of initialization object 150 by robot 102.
  • a radio transmission e.g., BLUETOOTH®
  • initialization object 150 can make a sound, such as a chirp, beep, noise, and/or any other sound.
  • the sound can be unique to the initialization object 150, wherein other initialization objects can make other sounds.
  • the sounds can be detected by sensor 300 (e.g., by a microphone), which can use the sound to identify and/or triangulate the location of initialization object 150.
  • initialization object 150 can include both human visible and invisible elements.
  • having both human visible and invisible elements can facilitate a user directing robot 102 to detect initialization object 150.
  • the visible portion can be smaller and/or more inconspicuous than the invisible portion.
  • sensor 300 can be positioned at height 330 from the ground, where height 330 can be determined at least in part on the expected position of initialization object 150 and/or where sensor 300 should be in order to capture initialization object 150 within field of view 320.
  • FIG. 5 illustrates example binary images 502, 504, 506 that can be used where example initialization object 150 is a sign having an image.
  • Example binary images 502, 504, 506 are for illustration only and are not meant to limit the image that can be used where initialization object 150 is a sign.
  • initialization object 150 can include any symmetrical or asymmetrical image.
  • binary image 502 has a white border 508 framing white shapes 510, 512 with a black background 546.
  • white border 508 can give corners and/or edges that can facilitate detection. With white border 508, black background 546 can be more clearly set apart from the surrounding for detection by robot 102.
  • some binary images such as example binary image 504, may not have a border.
  • White shapes 510, 512 in some cases, can be machine-generated blocky shapes, where the shapes are composed of contiguous substantially square and/or rectangular blocks.
  • a binary image may have one or more distinct (e.g., not contiguous) shapes.
  • binary image 506 comprises one white shape 514.
  • Binary image 504 comprises white shapes 516, 518, 520.
  • binary images 502, 504, 506 as initialization object 150 can be advantageous because each are distinct images that likely would not otherwise appear in a surrounding. Also, because binary images 502, 504, 506, are binary (e.g., black and white), they may be distinct from natural colors that would appear. The blocky white shapes and binary colors of images 502, 504, 506 also make them easily detectable using a low resolution RGB camera as camera 306. Also, binary images 502, 504, 506 lack rotational symmetry such that the rotation of initialization object 150 can also be detected.
  • binary images 502, 504, 506 can allow robot 102 to detect any one of binary images 502, 504, 506 with low hardware costs (e.g., cheaper components) and lower processing costs (e.g., involves relatively little computing power). Moreover, binary images 502, 504, 506 can be low cost to make and have many combinations, many of which can be randomly generated by a computer. Having many combinations can allow different binary images to be placed at different locations associated with different navigable routes.
  • Initialization object 150 can have a plurality of sides and corners, including right side, left side, top side, bottom side, top left corner, top right corner, bottom left corner, bottom right corner, or any other relative designation of any side, corner, and/or edge.
  • binary image 502 can have left side 530, right side 534, top side 532, and bottom side 536 as it is illustrated.
  • Binary image 502 can also top left corner 540, top right corner 544, bottom left corner 538, and bottom right corner 542.
  • initialization object 150 can include other images besides binary images (e.g., binary images 502, 504, 506).
  • initialization object 150 can be substantially similar to binary images (e.g., binary images 502, 504, 506), except using different colors besides black and/or white, or a plurality of colors, including colors viewable in the visible spectrum and/or colors not in the visible spectrum (e.g., viewable by machines or under machine-aid using UV, IR, etc.).
  • colors not viewable in the visible spectrum can reduce and/or prevent undesirable viewing by humans of the initialization object.
  • initialization object 150 can be an image of a scene, object, person, etc.
  • initialization object 150 can be any image, sign, object, and/or any object that can be recognized by a computer.
  • FIG. 6 illustrates a process flow diagram of an example method 600 where robot 102 can be initialized to autonomously navigate a learned route. For example, in portions 602, 604, 606 in teaching phase 614, robot 102 can learn route 116 demonstrated by user 904. Subsequently, in portions 608, 610, 612 in autonomous navigation phase 616, robot 102 can autonomously navigate route 116 as routes 106, 126 or substantially similar routes.
  • Portion 602 can include robot 102 detecting initialization object 150. This first detection of initialization object 150 can begin the initialization process for robot 102.
  • the appearance of initialization object 150 using exteroceptive sensor 300 can be stored in memory 302 in order for robot 102 to identify initialization object 150.
  • initialization object 150 is a binary image (e.g., one of binary images 502, 504, 506)
  • the binary image can be stored in memory 302.
  • sensor 300 can sense its surrounding (e.g., within the angles of view of sensor 300) and controller 304 can process the data obtained from sensor 300 and analyze that data to detect initialization object 150.
  • controller 304 can analyze each taken image (e.g., images taken by sensor 300 (e.g., camera 306)), and analyze those taken images for initialization object 150.
  • the analysis can comprise image processing.
  • controller 304 can take a taken image from camera 306 and perform a sequence of image processing techniques to make the binary image detectable.
  • library 324 can comprise one or more images of initialization object 150 and/or substantially similar objects that can be used to identify initialization object 150 in a taken image.
  • library 324 can be used in a supervised or unsupervised machine learning algorithm for controller 304 to learn to identify initialization object 150 in taken images.
  • the location of initialization object 150 and/or substantially similar objects can be identified in one or more images in library 324 (e.g., labelled by a user (e.g., hand-labelled) or automatically, such as with a computer program that is configured to generate/simulate library images of initialization object 150 and/or a substantially similar object and/or label those library images).
  • these images in library 324 can include images of initialization object 150 and/or substantially similar objects in different lighting conditions, angles, sizes (e.g., distances), clarity (e.g., blurred, obstructed/occluded, partially off frame, etc.), colors, surroundings, etc.
  • controller 304 can first be trained to identify initialization object 150 and/or substantially similar objects in many different situations, and then use that training to identify initialization object 150 in a given taken image.
  • controller 304 can compare taken images to labelled images in library 324. Where there is a substantial match between the taken image and one or more images in library 324, controller 304 can then identify initialization object 150 based at least in part on the one or more images of initialization object 150 and/or substantially similar objects labelled in library 324.
  • the amount of match between a taken image and one or more images in library 324 can be determined by comparison.
  • the difference between the taken image and one or more images from library 324 can be taken. This difference can be represented at least in part in a differential image.
  • the differential image can then be evaluated to determine the degree of similarity between the taken image and the one or more images form library 324.
  • controller 304 can take the sum (e.g., arithmetic sum, sum of the absolute values, etc.) of a plurality of pixels of the differential image and compare that sum to a predetermined threshold, where the predetermined threshold can be determined based at least in part on the robustness and tolerance of false positives.
  • controller 304 can determine that there has not been a match because the difference is too great.
  • an image comparator can be used that executes methods of comparing images to determine their similarities, and in some cases, the percentage similarity. These comparators can analyze the taken image and the one or more images from library 324 pixel-by-pixel, region-by-region, and/or based on shapes or distinct features of the images to look for similarities and/or differences.
  • library 324 of images of initialization object 150 and/or substantially similar objects can be used along with other image processing techniques to search each taken image for initialization object 150.
  • controller 304 can perform border following and/or contour tracing to extract boundaries in a taken image.
  • the borders of initialization object 150 can be particularly easy to detect where initialization object 150 contrasts from its surrounding.
  • initialization object 150 has unique patterns or colors from its surrounding, initialization object 150 can contrast from its surroundings and its border can be readily identified.
  • the black and white colors of a binary image e.g., binary images 502, 504, 506) can stand out and contrast with surroundings in many environments.
  • controller 304 can perform segmentation on a taken image.
  • initialization object 150 is a binary image
  • adaptive thresholding can segment the binary images.
  • segmentation can reduce false positives and reduce noise in the taken image.
  • the taken images, after and/or before segmentation can also be further cleaned up with morphological image processing (e.g., erosion and dilation to remove noise), filters, etc.
  • shape of initialization object 150 is distinct (e.g., square and/or rectangular like binary images 502, 504, 506), objects that are not substantially the same shape (or distorted/transformed versions of the shape as might be seen from a variety of angles of imaging) can be removed from the taken image.
  • initialization object 150 is square and/or rectangular
  • objects in the taken image that are not concave contours and/or objects with four distinct corners may be ignored and/or removed from the processing.
  • one or more images of library 324 can be used to detect initialization object 150 in the taken image by comparison.
  • the exact pattern of white shapes e.g., white shapes 516, 518, 520
  • controller 304 can be compared by controller 304 to one or more images of library 324. In this way, the presence and location of initialization object 150 can be identified in the taken image.
  • the plurality of initialization objects can each be indicative of different initial locations for learning routes and autonomously navigating routes.
  • different initialization objects can be used on different floors of a building, wherein robot 102 travels different routes on different floors based at least in part on detection of those different initialization objects.
  • different initialization objects can be on the same floor, but represent at least in part of different initial locations and different routes, accordingly, that robot 102 can travel.
  • it can be advantageous for the different initialization objects to have distinct appearances to avoid misidentification.
  • the pattern of white shapes can be different between each initialization object.
  • detection of initialization object 150 can be paired with detections by other sensors.
  • using detections by other sensors can account for situations where initialization object 150 is moved and/or where the environment has changed significantly. This use of other sensors can provide for signal redundancy and/or enhanced safety.
  • another sensor can be a laser scan lidar configured to image a scene.
  • the image can be a 3D image or a 2D image compressed from the data obtained from the scan lidar.
  • One or more scan lidar images can be associated with a valid detection of initialization object 150.
  • robot 102 can both determine if initialization object 150 has been detected and if the scan lidar image is substantially similar to the one or more scan lidar images associated with detection of initialization object 150. In this way, robot 102 (e.g., controller 304 of robot 102) can verify initialization object 150 using the scan lidar image.
  • Portion 604 can include robot 102 determining its position (e.g., where position includes robot's location, displacement and/or orientation, and can sometimes be interchangeable with the term pose as used herein) to initialization object 150 detected in portion 602.
  • FIG. 7 illustrates an overhead view of example robot 102 positioned at an example angle to example initialization object 150, where example sensor 300 measures an example distance 724 and example angle 718 from example point 794 on example sensor 300 to example point 792 of example initialization object 150.
  • Point 796 can be another point positioned on robot 102, such as the center, center of mass, position on another sensor (e.g., a point on proprioceptive sensor 308), or any other point on robot 102.
  • Point 792 can be a reference point on initialization object 150.
  • point 792 can be the center (e.g., the midpoint on the vertical and horizontal directions of initialization object 150), corner point (e.g., top left corner (e.g., top left corner 540 of binary image 502), top right corner (e.g., top right corner 544 of binary image 502), bottom left corner (e.g., bottom left corner 538 of binary image 502), bottom right corner (e.g., bottom right corner 542 of binary image 502), edge midpoint (e.g., midpoint of the right side (e.g., right side 534 of binary image 502), midpoint of the left side (e.g., left side 530 of binary image 502), midpoint of the bottom side (e.g., bottom side 536 of binary image 502), midpoint of the top side (e.g., top side 532 of binary image 502)), or any position disposed on and/or in initialization object 150.
  • Point 794 can be
  • robot 102 can include a body with a plurality of sides, such as front side 702, right side 708, left side 706, and back side 704.
  • Robot 102 can also have a top side 764 and a bottom side (not pictured).
  • a person having ordinary skill in the art should appreciate that robot 102 can have other sides as well, corresponding to the surfaces of robot 102, which can vary by shape (e.g., rectangular, pyramidal, humanoid, or any designed shape).
  • front side 702 can be positioned on the forward-facing side of robot 102, where the forward-facing side is forward in the direction of forward movement of robot 102.
  • Back side 704 can be positioned on the backward -facing side of robot 102, where the backward-facing side is the side facing in substantially the opposite direction of the forward facing side, such as front side 702.
  • Right side 708 can be the right-hand side relative to front side 702
  • left side 706 can be the left-hand side relative to front side 702.
  • robot 102 can have additional sensors, such as additional exteroceptive sensors 700A - 700C, which can be substantially similar to exteroceptive sensor 300 or any other sensor, and additional proprioceptive sensor 710, which can be substantially similar to proprioceptive sensor 308 or any other sensor.
  • robot 102 can utilize one or more sensors (e.g., sensors 300, 700A - 700C) to detect and/or initialize robot 102 to initialization object 150.
  • sensors e.g., sensors 300, 700A - 700C
  • a sensor positioned on and/or around those sides can be used to detect initialization object 150 and determine robot 102's position relative to initialization object 150.
  • initialization object 150 can be positioned on a ceiling.
  • Robot 102 can measure a distance 724 between point 794 and point 792, which can be an absolute distance measurement using standard units, such as inches, feet, meters, or any other unit of measurement (e.g., measurements in the metric, US, or other system of measurement).
  • distance 724 can be measured in relative and/or non-absolute units, such as ticks, pixels, percentage of range of a sensor, and the like.
  • distance 724 can be expressed in x- and y- coordinates relative to an origin, such as point 792, point 794, point 796 or any other determined location.
  • the x-coordinate can be the distance to the origin to a first axis
  • the y-coordinate can be the distance to the origin in a second axis, the second axis orthogonal to the first forming a rectangular coordinate system.
  • distance 724 can be the difference between the x-coordinates and y-coordinates of point 792 and point 794.
  • distance 724 can be expressed in three dimensions, including the aforementioned x- and y- coordinates, and a z-coordinate, where the z- coordinate can be the distance to the origin in a third axis.
  • Robot 102 can also approximate its orientation.
  • robot 102 can approximate the relative angle 718 between point 794 and point 792.
  • Angle 718 can be measured in degrees, radians, or any unit.
  • angle 718 can be measured relative to a 2D plane, such as a horizontal plane (e.g., the aforementioned rectangular coordinate system of distance 724 or other measurements).
  • additional angles can be measured, such as one or more of roll, yaw, and, pitch of point 792 to point 794.
  • the determination by robot 102 of position relative to initialization object 150 can include determining a distance from a point on robot 102 to a point on initialization object 150 as well as an orientation between those points.
  • robot 102 in order to determine the position of point 794 relative to point 792, robot 102 can identify properties/characteristics of sensor 300 that detected initialization object 150. In some implementations, these properties/characteristics can be represented as calibration parameters detailing aspects of sensor 300.
  • calibration parameters can include focal length, distortion, angle of view, distortion effects, aspect ratio, height above ground, etc.
  • These calibration parameters can be known by a user based on specifications of camera 306 and/or the set-up/positioning of camera 306.
  • lens 314 and camera 306 can be manufactured to specifications that can be known to robot 102 (and/or any of its components and/or subcomponents) or a user.
  • a user 904 of robot 102 can measure such calibration parameters, such as by measuring the height camera 306 is from the ground.
  • one or more of the properties/characteristics of camera 306 can be estimated from one or more images taken by camera 306. For example, there can be known relationships between images taken by camera 306 and the properties/characteristics of camera 306.
  • FIG. 8 illustrates an example initialization object 150 being represented in an example image 800.
  • camera 306 can take image 800 of a scene including objects such as initialization object 150.
  • Initialization object 150 (as illustrated as an example sign with a binary image on wall 350 in FIG. 8), is in three-dimensional space having coordinates (x, y, z) in coordinate system 804.
  • the constants c x and Cy can be the displacements, from the optical axis, of the center of coordinates on the image 800 in the ximage and yimage directions, respectively. Knowing these relationships can allow calibration of camera 306, such as computer- implemented calibration, so that distorted images can be corrected.
  • Image 800 represents a rectangular portion of field of view 320 in the illustrated non-limiting example.
  • image 800 is for illustrative purposes, and image 800 could comprise other views having different characteristics (e.g., angles, lighting, distortions, obstructions/occlusions, partial pictures, etc.) than as illustrated.
  • the estimation of position of point 794 (and/or point 796) to point 792 can then utilize pose estimation systems and methods.
  • pose estimation systems and methods can include analytic or geometric methods and/or learning-based methods.
  • the location of each 2D point in an image 800 can be treated as a mapping from the 3D scene (e.g., the scene in field of view 320) to the 2D image.
  • initialization object 150 can be projected onto the image 800.
  • the calibration parameters of camera 306 and the geometry of initialization object 150 can be also known, a function relating the 3D points in the scene including initialization object 150 to a 2D image 800 can be found. There are many such functions, some of which are known in the art.
  • a projective camera model can take the form P ⁇ CS where, P is a matrix representative of 2D points (e.g., pixels) in an image, S is a matrix representative of 3D points in a scene, C is the camera matrix, and ⁇ indicates at least in part that the left and right hand sides of the equation are equal up to a non-zero scalar multiplication.
  • C can comprise, among other things, values based at least in part on the various calibration parameters of camera 306, such as one or more of focal length, tangential and radial distortion coefficients, optical center, aspect ratio, etc.
  • C itself can be represented as the product of multiple matrices.
  • A is a camera intrinsic matrix (e.g., containing characteristics of camera 306) and [R t] includes extrinsic parameters (expressed in matrix form) such as the rotation and translation relating the scene coordinates (e.g., coordinate system 804) to the 2D image coordinates (e.g., coordinate system 802).
  • the rotation R and transition t can give the position of initialization object 150.
  • estimating [R t] based at least in part on known camera intrinsics and locations of initialization object 150 in an image 800 can allow robot 102 to find the position of robot 102 relative to initialization object 150.
  • A can be a 3x3 matrix expressed as:
  • oc and ⁇ can be scale factors in the image image and yimage axes respectively, / can be the focal length, ⁇ can be representative of the skewness of the axes of the image coordinates.
  • R can be a 3 x 3 rotation matrix and t can be a 3 -dimensional translation vector.
  • t can be the position of the origin of the scene coordinate system (e.g., coordinate system 804) expressed in coordinates of a 3D camera- centered coordinate system, where the aperture of lens 314 of camera 306 can be the origin.
  • the translation t can be transformed into scene coordinates.
  • M can be a column vector describing the location of the aperture of lens 314 in scene coordinates.
  • controller 304 can thus solve expressions of the form P ⁇ CS for one or more of Rc, M, R, t, or any other parameter, thus finding the position of camera 306 in scene coordinates (e.g., where the coordinates can be centered at a point, such as one of points 792, 794, 796 or another point on initialization object 150, robot 102, or elsewhere in the environment). Controller 304 can solve for the position using known mathematical methods, including Direct Linear Transformation ("DLT”) and others. Distortion (e.g., radial and/or translational) of lens 314 can also be factored in non-linear models, where the non-linear models can be solved by nonlinear estimation techniques.
  • DLT Direct Linear Transformation
  • a point e.g., point 792, point 794, or point 796
  • the scene coordinate system e.g., coordinate system 804
  • learning-based methods can be used by robot 102 to find the position of one or more of points 794, 796 to point 792, as illustrated in FIG. 7.
  • library 324 can comprise example images of initialization object 150 and/or substantially similar objects depicting at least in part initialization object 150 and/or substantially similar objects at various positions (e.g., distances in a coordinate system and/or orientations/angles).
  • Library 324 can then be used in a supervised or unsupervised machine learning algorithm for controller 304 to learn to identify/associate the position of one or more of points 794, 796 to point 792 from taken images (e.g., image 800).
  • the images of library 324 can be identified (e.g., labelled by a user (e.g., hand-labelled) or automatically, such as with a computer program that is configured to generate/simulate library images of initialization object 150 and/or a substantially similar object and/or label those library images) with their respective corresponding measured positions of one or more of points 794, 796 to point 792.
  • library 324 can also include images of initialization object 150 and/or substantially similar objects in different lighting conditions, angles, sizes (e.g., distances), clarity (e.g., blurred, obstructed/occluded, partially off frame, etc.), colors, surroundings, etc.
  • controller 304 can first be trained to identify the position of one or more of points 794, 796 to point 792 based at least in part on the corresponding images of initialization object 150 and/or substantially similar objects in many different situations in library 324. Robot 102 can then use that training to identify the position of one or more of points 794, 796 to initialization object 150 in a taken image.
  • controller 304 can process each taken image
  • controller 304 can identify the position of one or more of points 794, 796 to point 792 from the taken image as substantially the same as the position associated with the matched image or plurality of images from library 324. In some cases, where the taken image does not substantially match an image in library 324, controller 304 can estimate the position of one or more of points 794, 796 relative to point 792 based at least in part on similar images in library 324 that may not be exact matches.
  • controller 304 may average (and/or perform other statistical estimations based at least in part upon) a plurality of positions corresponding to similar images in library 324 to provide an estimate of the position of one or more of points 794, 796 to point 792 on the taken image.
  • robot 102 can determine the position of one or more of points 794, 796 relative to point 792.
  • the position of point 796 relative to point 792 can also be found.
  • the positioning of point 794 to point 796 can be measured (e.g., in x, y, z coordinates and/or the angles), the position of point 796 to point 792 can also be found.
  • the position of point 796 relative to point 792 can be calculated based at least in part on the position of point 794 to point 792 using arithmetic, such as by adding the angle and/or the difference in X-, y-, and/or z- coordinates between point 796 and point 794 to the position of point 794 relative to point 792.
  • the origin of the scene coordinate system can be taken to be point 792, point 794, point 796, or any other point.
  • identifying point 792 as the original can provide a potentially stationary object as the origin, which can supply consistency between the origin for routes 106, 116, 126, and/or any other route navigated from an initialization location relative to initialization object 150.
  • robot 102 can also initialize its odometry in odometry unit 326. In some cases, robot 102 can initialize proprioceptive sensor 308. For example, knowing the initial position of one or more of points 794, 796 to point 792 can allow robot 102 to know its initial location. It can then track other positions to which it moves relative to the initial location and/or point 792. This can allow robot 102 to determine one or more of its travelled positions on demonstrated route 116, as well as when it is autonomously navigating, such as on routes 106, 126.
  • portion 606 can include robot 102 learning a route.
  • a user 904 can initiate route learning, such as by instructing robot 102 to learn a route through example interface 900.
  • FIG. 9A illustrates an example interface 900 for receiving an input from a user to begin learning (e.g., from teaching by a user 904) a route or to choose a route.
  • a user 904 can select input 972 on interface 900 on user interface 318, where input 972 can allow a user to select a recorded route of robot 102.
  • robot 102 can automatically begin learning a route by virtue of detecting initialization object 150. For example, where robot 102 has no routes stored in memory 302, it can automatically begin learning whatever actions and/or route user 904 performs, and the positions and/or trajectories of those actions relative to initialization object 150 and/or any other point described in this disclosure.
  • robot 102 can initialize route learning.
  • Robot 102 can begin learning a route from the initial location where robot 102 detected initialization object 150.
  • FIG. 9B illustrates a side view of user 904 controlling robot 102 while teaching a route.
  • User 904 can be a janitor, custodian, and/or any person or robot who can use robot 102.
  • robot 102 can be a floor cleaner configured to clean the floor of a store, warehouse, office building, home, storage facility, etc.
  • robot 102 can have brush 908 and squeegee 912 configured to clean the floor beneath and/or around (e.g., within the reach of brush 908 and squeegee 912) robot 102.
  • User 904 can demonstrate route 116 (illustrated in FIG. IB) to robot 102.
  • robot 102 can also learn actions and/or actuator instructions associated with positions (e.g., relative to a reference point, such as one or more of points 792 or any other point) and/or trajectories (e.g., while moving in a certain direction or in a particular sequence along route 106) along route 116, such as actuation of brush 908 to clean the floor.
  • Robot 102 can be configured in any number of ways for control by user 904. As illustrated, user 904 can walk behind robot 102 and/or steer robot 102 using steering wheel 910 as robot 102 moves forward under motorized power. In other implementations, robot 102 can be a ride-on floor cleaner (not pictured) where user 904 can ride on a seat or standing platform of robot 102 and control it.
  • user 904 can control robot 102 remotely with a remote control, such as a radio remote, mobile device, joystick, and/or any other apparatus for navigation known in the art.
  • This control can include turning left, turning right, moving forward (e.g., using a gas pedal or telling robot 102 to go in a forward direction), moving backwards (e.g., using a reverse pedal or telling robot 102 to go in a backward direction), turn on/off, raise/lower brush, turn on/off water, etc.
  • user 904 may control one or more actuators, which can control instruments of robot 102, such as such as turning on/off water, spraying water, turning on/off vacuums, moving vacuum hose positions, gesticulating an arm, raising/lowering a lift, turning camera 306, actuating sensors 300, 308, 700A - 700C, etc.
  • Robot 102 can also learn the control of such instruments relative to its route 116 (e.g., robot 102's position and/or trajectories along route 116). For example, where robot 102 is a floor cleaner learning route 116, robot 102 can also learn when along route 116 to actuate these various instruments on robot 102, such as brush 908 and/or squeegee 916.
  • robot 102 can perform those actions and/or actuator instructions when it is passes by a position, and/or in some cases, when it passes by that position in the same direction and/or at the same relative time as a learned route. Accordingly, in these implementations, robot 102 would not perform those actions and/or actuator instructions each time it passes a position (e.g., where it loops around and passes the same physical location multiple times), but only perform such actions and/or such actuator instructions when it passes by the location either in a particular direction or at particular instance(s) in the route.
  • Robot 102 can then store route 116, and any associated actions along route
  • Robot 102 can also use one or more of sensors 300, 700A - 700C to record sensed objects (e.g., object 906) in the surrounding of robot 102.
  • sensed objects e.g., object 906
  • one or more sensors 300, 700 A - 700C can emit energy waves 960 A-C to detect objects around robot 102.
  • robot 102 can both use its odometry (e.g., with proprioceptive sensor 308 of odometry unit 326) and its exteroceptive sensors (e.g., exteroceptive sensors 300, 700A-C) to sense its surrounding while learning route 116.
  • robot 102 can learn any controls in relationship to its position and/or trajectory with respect to point 792.
  • robot 102 can learn to turn at certain positions, and/or when it goes to certain positions in certain directions and/or at certain points relative to the learned route.
  • Robot 102 can learn to actuate brush 908 when at certain positions, and/or when it goes to certain positions in certain directions and/or at certain points relative to the learned route.
  • Robot 102 can learn to associate any sort of action with particular positions and/or trajectories relative to point 792 or any other point.
  • Sensors 300, 700 A - 700C can detect objects which can both allow robot 102 to map the location of those objects on a map stored in memory 302 and/or avoid objects it comes across.
  • Robot 102 can map environment 100 and/or route 116 using Simultaneous Localization and Mapping ("SLAM”), augmented reality, scale-invariant feature transform (“SIFT”), and/or other systems and methods.
  • SLAM Simultaneous Localization
  • end location 114 can be determined by detection of initialization object 150 or a different initialization object.
  • the position of robot 102 at end location 114 can be verified at end location 114.
  • robot 102 can compare robot 102's odometry to the verified position and determine the drift of the odometry. This can facilitate identification and correction of errors as robot 102 maps the environment and/or routes.
  • robot 102 detects initialization object 150 at both the initial location and end location 114 robot 102's route may be substantially a closed loop.
  • portion 608 includes detecting initialization object 150 again.
  • This second detection can coincide with a second placement of robot 102 into an initial location relative to initialization object 150.
  • This detection can occur at a later point in time after portion 606, such as substantially right after the demonstration of portion 606, or at some later time, such as hours later, days later, weeks later, or whenever the user 904 desires to clean the floor.
  • robot 102 can detect initialization object 150 in a substantially similar way to how robot 102 detected initialization object 150 with respect to portion 602.
  • portion 610 can include robot 102 determining robot 102's position with respect to initialization object 150 (e.g., point 792) detected in portion 608 substantially similar to the way robot 102 determined its position to initialization object 150 detected in portion 602.
  • initialization object 150 e.g., point 792
  • Portion 612 can include robot 102 selecting a recorded route to navigate autonomously.
  • the selection of the recorded route (e.g., route 116) by robot 102 can be based at least in part on user input.
  • a user 904 can select input 972 on user interface 900 (illustrated in FIG. 9A) on user interface 318, where input 972 can allow a user to select a recorded route of robot 102.
  • interface 1000 illustrated in FIG. 10, can appear.
  • FIG. 10 illustrates example interface 1000, which can be used for route selection.
  • Interface 1000 can present a plurality of routes for selection displayed as selectable inputs 1002A - 1002F.
  • a user 904 may select one of selectable inputs 1002 A - 1002F via touch, in the case user interface 318 includes a touch screen, and/or any other input mechanism of user interface 318.
  • input 1002F can correspond with route 116 and/or any other route learned in portion 606.
  • robot 102 can then select and recall the corresponding learned route based at least in part on the selection of user 904.
  • robot 102 detects initialization object 150, it can filter the routes in memory to present for selection only those route(s) that are associated with initialization object 150. This ability can advantageously avoid confusion and prevent user 904 from selecting a route that is not navigable from a particular position relative to initialization object 150.
  • robot 102 can automatically select a recorded route based on the initialization object it detected in portion 608.
  • initialization object 150 can be associated with only demonstrated route 116.
  • FIG. 11 illustrates example environment 100 with a plurality of example initialization objects 150, 1150A - 1150C.
  • initialization object 150 can be associated with only demonstrated route 116.
  • robot 102 can have other initialization objects 1150A-C, positioned in one or more locations in environment 100, each associated with other demonstrated routes, and each initialization objects 150, 1150A - 1150C, distinguishable by unique characteristics from the others.
  • having a plurality of initialization objects 150, 1150A - 1150C can allow a user to demonstrate, and allow robot 102 to move autonomously through, a variety of routes.
  • robot 102 can more quickly begin autonomous navigation with minimal additional user input.
  • Additional initialization objects can be advantageous when there are multiple floors and/or environments. Each floor and/or environment can have its own initialization objects, distinguishable by unique characteristics from the others. In some cases, each of the plurality of initialization objects can themselves be associated with a plurality of user selectable routes.
  • robot 102 can then autonomously navigate that route. For example, robot 102 can navigate route 106 and/or route 126 autonomously.
  • a user can select routes based at least in part on predetermined actions and/or predetermined sequences of actions of robot 102, such as actions at least in part under user control.
  • user 904 can input a sequence of actions (e.g., turn left, turn right, reverse, move forward, etc.) that is associated with a route. When robot 102 receives such a sequence of actions it can recall a particular route and navigate that route.
  • user 904 can control robot 102 in a particular way, such as in a figure eight, square, and/or in the shape of one or more letters or numbers (e.g., "B", “R”, “7”, and/or any other letter or number).
  • This control can be associated at least in part with a route.
  • robot 102 detects that robot 102 has been controlled in the shape of such one or more letters or numbers, it can recall a route associated with that controlled shape and navigate that route.
  • FIG. 12 illustrates an example method 1200 for operating an example robot 102.
  • Portion 1202 includes taking a first image of an initialization object while the robot is in a first location.
  • Portion 1204 includes determining a first starting position of the robot relative to the initialization object based at least in part on the first image while the robot is in the first location.
  • Portion 1206 includes learning a route by user demonstration, beginning from the first location, where the learned route associates actions of the robot with positions of the robot relative to the initialization object, wherein the positions are determined at least in part from the first starting position.
  • Portion 1208 includes taking a second image of the initialization object while the robot is in a second location.
  • Portion 1210 includes determining a second starting position of the robot relative to the initialization object based at least in part on the second image while the robot is in the second location.
  • Portion 1212 includes navigating the robot autonomously beginning from the second location through at least a portion of the learned route while instructing the robot to perform one or more associated actions based at least in part with positions of the robot relative to the initialization object, wherein the positions are determined at least in part from the second starting position.
  • computer and/or computing device can include, but are not limited to, personal computers (“PCs”) and minicomputers, whether desktop, laptop, or otherwise, mainframe computers, workstations, servers, personal digital assistants ("PDAs”), handheld computers, embedded computers, programmable logic devices, personal communicators, tablet computers, mobile devices, portable navigation aids, J2ME equipped devices, cellular telephones, smart phones, personal integrated communication or entertainment devices, and/or any other device capable of executing a set of instructions and processing an incoming data signal.
  • PCs personal computers
  • PDAs personal digital assistants
  • handheld computers handheld computers
  • embedded computers embedded computers
  • programmable logic devices personal communicators
  • tablet computers tablet computers
  • mobile devices portable navigation aids
  • J2ME equipped devices portable navigation aids
  • J2ME equipped devices J2ME equipped devices
  • cellular telephones smart phones
  • personal integrated communication or entertainment devices personal integrated communication or entertainment devices
  • Such computer program and/or software may be rendered in any programming language or environment including, for example, C/C++, C#, Fortran, COBOL, MATLABTM, PASCAL, Python, assembly language, markup languages (e.g., HTML, SGML, XML, VoXML), and the like, as well as object-oriented environments such as the Common Object Request Broker Architecture ("CORBA”), JAVATM (including J2ME, Java Beans, etc.), Binary Runtime Environment (e.g., BREW), and the like.
  • CORBA Common Object Request Broker Architecture
  • JAVATM including J2ME, Java Beans, etc.
  • BREW Binary Runtime Environment
  • connection, link, transmission channel, delay line, and/or wireless can include a causal link between any two or more entities (whether physical or logical/virtual), which enables information exchange between the entities.
  • the term “including” should be read to mean “including, without limitation,” “including but not limited to,” or the like; the term “comprising” as used herein is synonymous with “including,” “containing,” or “characterized by,” and is inclusive or open-ended and does not exclude additional, unrecited elements or method steps; the term “having” should be interpreted as “having at least;” the term “such as” should be interpreted as “such as, without limitation;” the term 'includes” should be interpreted as “includes but is not limited to;” the term “example” is used to provide exemplary instances of the item in discussion, not an exhaustive or limiting list thereof, and should be interpreted as “example, but without limitation;” adjectives such as "known,” “normal,” “standard,” and terms of similar meaning should not be construed as limiting the item described to a given time period or to an item available as of a given time, but instead should be read to encompass known, normal, or standard technologies that may be available
  • a group of items linked with the conjunction "and” should not be read as requiring that each and every one of those items be present in the grouping, but rather should be read as “and/or” unless expressly stated otherwise.
  • a group of items linked with the conjunction “or” should not be read as requiring mutual exclusivity among that group, but rather should be read as “and/or” unless expressly stated otherwise.
  • the terms “about” or “approximate” and the like are synonymous and are used to indicate that the value modified by the term has an understood range associated with it, where the range can be ⁇ 20%, ⁇ 15%, ⁇ 10%, ⁇ 5%, or ⁇ 1%).
  • a result e.g., measurement value
  • close can mean, for example, the result is within 80% of the value, within 90% of the value, within 95% of the value, or within 99% of the value.
  • defined or “determined” can include “predefined” or “predetermined” and/or otherwise determined values, conditions, thresholds, measurements, and the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • General Physics & Mathematics (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Mechanical Engineering (AREA)
  • Robotics (AREA)
  • Electromagnetism (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Business, Economics & Management (AREA)
  • Game Theory and Decision Science (AREA)
  • Health & Medical Sciences (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Manipulator (AREA)
  • Numerical Control (AREA)
PCT/US2017/032285 2016-05-11 2017-05-11 Systems and methods for initializing a robot to autonomously travel a trained route Ceased WO2017197197A1 (en)

Priority Applications (5)

Application Number Priority Date Filing Date Title
KR1020187035946A KR102444658B1 (ko) 2016-05-11 2017-05-11 훈련된 경로를 자율주행하도록 로봇을 초기화하기 위한 시스템 및 방법
JP2019511832A JP6942177B2 (ja) 2016-05-11 2017-05-11 ロボットが学習した経路を自律的に走行するように初期化するためのシステムおよび方法
EP17796891.4A EP3455686A4 (en) 2016-05-11 2017-05-11 SYSTEMS AND METHODS FOR INITIALIZING A ROBOT FOR THE AUTONOMOUS DRIVING OF A TRAINED ROUTE
CA3023557A CA3023557A1 (en) 2016-05-11 2017-05-11 Systems and methods for initializing a robot to autonomously travel a trained route
CN201780041844.2A CN109416538A (zh) 2016-05-11 2017-05-11 用于对机器人进行初始化以沿着训练路线自主行进的系统和方法

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US15/152,436 2016-05-11
US15/152,436 US10241514B2 (en) 2016-05-11 2016-05-11 Systems and methods for initializing a robot to autonomously travel a trained route

Publications (1)

Publication Number Publication Date
WO2017197197A1 true WO2017197197A1 (en) 2017-11-16

Family

ID=60266841

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2017/032285 Ceased WO2017197197A1 (en) 2016-05-11 2017-05-11 Systems and methods for initializing a robot to autonomously travel a trained route

Country Status (7)

Country Link
US (3) US10241514B2 (enExample)
EP (1) EP3455686A4 (enExample)
JP (1) JP6942177B2 (enExample)
KR (1) KR102444658B1 (enExample)
CN (1) CN109416538A (enExample)
CA (1) CA3023557A1 (enExample)
WO (1) WO2017197197A1 (enExample)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2019179422A (ja) * 2018-03-30 2019-10-17 日立造船株式会社 飛行体、情報記録方法、飛行体の制御方法、および制御プログラム
JP2021509986A (ja) * 2018-01-02 2021-04-08 ゼネラル・エレクトリック・カンパニイ 人間による実演に基づく、ロボットによる産業タスクの学習のためのシステム及び方法
EP3872528A4 (en) * 2018-10-22 2021-12-29 Ecovacs Robotics Co., Ltd. Travel control method, device, and storage medium
CN114578379A (zh) * 2021-10-12 2022-06-03 苏州趣洁电器有限公司 识别柔性障碍物的方法、装置及自行走设备
DE102021204149A1 (de) 2021-04-27 2022-10-27 BSH Hausgeräte GmbH Objekterkennung für ein Hausgerät

Families Citing this family (59)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12075957B2 (en) 2014-03-24 2024-09-03 Intelligent Cleaning Equipment Holdings Co. Ltd. Floor cleaning machines having intelligent systems, associated sub-assemblies incorporating intelligent systems, and associated methods of use
KR101697857B1 (ko) * 2015-04-08 2017-01-18 엘지전자 주식회사 이동 로봇 및 그의 위치인식방법
US10782686B2 (en) * 2016-01-28 2020-09-22 Savioke, Inc. Systems and methods for operating robots including the handling of delivery operations that cannot be completed
US10282849B2 (en) 2016-06-17 2019-05-07 Brain Corporation Systems and methods for predictive/reconstructive visual object tracker
KR102526083B1 (ko) * 2016-08-30 2023-04-27 엘지전자 주식회사 이동 단말기 및 그의 동작 방법
US10274325B2 (en) 2016-11-01 2019-04-30 Brain Corporation Systems and methods for robotic mapping
US10001780B2 (en) * 2016-11-02 2018-06-19 Brain Corporation Systems and methods for dynamic route planning in autonomous navigation
US10723018B2 (en) 2016-11-28 2020-07-28 Brain Corporation Systems and methods for remote operating and/or monitoring of a robot
US10852730B2 (en) 2017-02-08 2020-12-01 Brain Corporation Systems and methods for robotic mobile platforms
JP6638699B2 (ja) * 2017-05-30 2020-01-29 トヨタ自動車株式会社 誤操作判定装置
US10926725B2 (en) 2018-10-10 2021-02-23 Stocked Robotics, Inc. Non-destructive kit mounting system for driverless industrial vehicles
WO2019107164A1 (ja) * 2017-11-28 2019-06-06 Thk株式会社 画像処理装置、移動ロボットの制御システム、移動ロボットの制御方法
JP2019102047A (ja) * 2017-11-28 2019-06-24 Thk株式会社 画像処理装置、移動ロボットの制御システム、移動ロボットの制御方法
WO2019126234A1 (en) * 2017-12-19 2019-06-27 Learning Resources, Inc. Toy robot
KR102015030B1 (ko) * 2018-01-09 2019-08-27 엘지전자 주식회사 이동 로봇 및 이동 로봇의 제어방법
JP6960061B2 (ja) * 2018-01-10 2021-11-05 シムビ ロボティクス, インコーポレイテッドSimbe Robotics, Inc. 溢れおよび危険物を検出して対応する方法
US10635844B1 (en) * 2018-02-27 2020-04-28 The Mathworks, Inc. Methods and systems for simulating vision sensor detection at medium fidelity
CN108381509B (zh) * 2018-03-19 2021-03-02 京东方科技集团股份有限公司 智能抓取装置及其控制方法、智能抓取控制系统
CN111182174B (zh) * 2018-11-09 2020-12-04 珠海格力电器股份有限公司 一种扫地机器人补光的方法及装置
WO2020102946A1 (zh) * 2018-11-19 2020-05-28 珊口（深圳）智能科技有限公司 构建地图、定位、导航、控制方法及系统、移动机器人
US11066088B2 (en) * 2018-12-13 2021-07-20 Xidian University Detection and positioning method for train water injection port
CN109528103B (zh) * 2019-01-22 2023-09-15 苏州锐科兰电器有限公司 拖地机
JP7309371B2 (ja) * 2019-01-25 2023-07-18 株式会社ソニー・インタラクティブエンタテインメント ロボット制御システム
JP7417356B2 (ja) 2019-01-25 2024-01-18 株式会社ソニー・インタラクティブエンタテインメント ロボット制御システム
JP7190919B2 (ja) * 2019-01-25 2022-12-16 株式会社ソニー・インタラクティブエンタテインメント 画像解析システム
US11130235B2 (en) * 2019-02-15 2021-09-28 Boston Dynamics, Inc. Work mode and travel mode for mobile robots
WO2020189462A1 (ja) * 2019-03-15 2020-09-24 ヤマハ発動機株式会社 既定ルート走行車両
KR102201002B1 (ko) * 2019-03-26 2021-01-12 엘지전자 주식회사 로봇 청소기
US11554495B2 (en) 2019-05-30 2023-01-17 Lg Electronics Inc. Method of localization using multi sensor and robot implementing same
WO2021006364A1 (ko) * 2019-07-05 2021-01-14 엘지전자 주식회사 지능형 로봇 청소기의 주행 방법
US11580724B2 (en) * 2019-07-23 2023-02-14 Toyota Research Institute, Inc. Virtual teach and repeat mobile manipulation system
CN110365908B (zh) * 2019-08-01 2020-06-19 北京鲲鹏神通科技有限公司 一种视觉控制装置
DE102019125469A1 (de) * 2019-09-23 2021-03-25 Hako Gmbh Verfahren zum Reinigen einer Bodenfläche mit einer Bodenreinigungsmaschine
US12358145B2 (en) 2020-03-18 2025-07-15 Cognex Corporation System and method for three-dimensional calibration of a vision system
CN111265139A (zh) * 2020-03-19 2020-06-12 广东蓝水花智能电子有限公司 一种智能马桶座圈控制方法
US11532205B2 (en) 2020-03-27 2022-12-20 Aristocrat Technologies, Inc. Gaming service automation system with graphical user interface
US20210341300A1 (en) * 2020-05-04 2021-11-04 Here Global B.V. Method, apparatus, and system for providing a personally relevant navigation route comparison
CN111681451B (zh) * 2020-06-12 2022-10-14 苏州博众智能机器人有限公司 一种停车数据处理方法、系统和停车场
US20210397198A1 (en) * 2020-06-18 2021-12-23 Ford Global Technologies, Llc Enhanced vehicle operation
CN113408382A (zh) * 2020-07-13 2021-09-17 追创科技(苏州)有限公司 自移动设备的控制方法、装置、存储介质及自移动设备
US11654566B2 (en) 2020-08-12 2023-05-23 General Electric Company Robotic activity decomposition
US11897134B2 (en) 2020-08-12 2024-02-13 General Electric Company Configuring a simulator for robotic machine learning
EP3968051B1 (en) * 2020-09-15 2024-10-30 Infineon Technologies AG Guiding system for a robot, base station including such a guiding system, and method for guiding a robot
USD1006884S1 (en) 2020-09-25 2023-12-05 Aristocrat Technologies, Inc. Gaming services robot
AU2021385603A1 (en) * 2020-11-30 2023-06-29 Clutterbot Inc. Clutter-clearing robotic system
DE112022001174B4 (de) * 2021-04-27 2025-07-10 Nvidia Corporation Strategieebenen für die maschinensteuerung
EP4374237B1 (de) * 2021-07-23 2025-04-16 Alfred Kärcher SE & Co. KG Bodenreinigungssystem, bodenreinigungsgerät und verfahren zum betreiben eines bodenreinigungssystems oder eines bodenreinigungsgerätes
CN118871869A (zh) * 2022-01-21 2024-10-29 波士顿动力公司 用于记录机器人任务的系统和方法
WO2023144523A1 (en) * 2022-01-25 2023-08-03 Gama Healthcare Ltd Methods for controlling the movement of a robot for disinfection
JP2023121298A (ja) * 2022-02-21 2023-08-31 沖電気工業株式会社 制御装置、制御方法、及びプログラム
WO2023173209A1 (en) * 2022-03-14 2023-09-21 Anram Holdings Retrofit kits for enabling autonomous operation in autoscrubbers
USD1060896S1 (en) 2022-05-05 2025-02-04 Intelligent Cleaning Equipment Holdings Co. Ltd. Cleaning machine
USD1054640S1 (en) 2022-05-05 2024-12-17 Intelligent Cleaning Equipment Holdings Co. Ltd. Side brush
WO2024016101A1 (en) 2022-07-18 2024-01-25 Intelligent Cleaning Equipment Holdings Co., Ltd. Systems and methods for robotic navigation, teaching and mapping
US12485528B2 (en) 2022-09-29 2025-12-02 Aristocrat Technologies, Inc. Mobile kiosk service robot and base assembly for same
US20240265565A1 (en) * 2023-02-06 2024-08-08 Walmart Apollo, Llc Systems and methods for analyzing depth in images obtained in product storage facilities to detect outlier items
US20240292990A1 (en) * 2023-03-01 2024-09-05 Clutterbot, Inc. Robot vacuum system with obstruction control
USD1075211S1 (en) 2024-01-09 2025-05-13 Quasi Inc. Autonomous mobile delivery transporter
US12310545B1 (en) * 2024-08-29 2025-05-27 Clutterbot, Inc. General purpose tidying robot

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5341540A (en) * 1989-06-07 1994-08-30 Onet, S.A. Process and autonomous apparatus for the automatic cleaning of ground areas through the performance of programmed tasks
US20090228166A1 (en) * 2006-01-18 2009-09-10 I-Guide, Llc Robotic Vehicle Controller
US20160121487A1 (en) * 2014-11-03 2016-05-05 Qualcomm Incorporated Communicating Configurable Instruction Sets to Robots for Controlling Robot Behavior
US20160287044A1 (en) * 2013-09-18 2016-10-06 Murata Machinery, Ltd. Autonomously traveling floor washer, cleaning schedule data structure, storage medium, method for creating cleaning schedule, and program

Family Cites Families (196)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4278142A (en) * 1978-05-08 1981-07-14 Agency Of Industrial Science And Technology Automatic guidance system for vehicles
US5280179A (en) 1979-04-30 1994-01-18 Sensor Adaptive Machines Incorporated Method and apparatus utilizing an orientation code for automatically guiding a robot
JPS59121405A (ja) * 1982-12-14 1984-07-13 Honda Motor Co Ltd 移動ロボツトの制御装置
US4638445A (en) 1984-06-08 1987-01-20 Mattaboni Paul J Autonomous mobile robot
US5121497A (en) 1986-03-10 1992-06-09 International Business Machines Corporation Automatic generation of executable computer code which commands another program to perform a task and operator modification of the generated executable computer code
US4763276A (en) 1986-03-21 1988-08-09 Actel Partnership Methods for refining original robot command signals
US4852018A (en) 1987-01-07 1989-07-25 Trustees Of Boston University Massively parellel real-time network architectures for robots capable of self-calibrating their operating parameters through associative learning
US4790402A (en) * 1987-09-28 1988-12-13 Tennant Company Automated guided vehicle
EP0366350A3 (en) * 1988-10-25 1991-05-22 Tennant Company Guiding an unmanned vehicle by reference to overhead features
US5155684A (en) * 1988-10-25 1992-10-13 Tennant Company Guiding an unmanned vehicle by reference to overhead features
ATE150879T1 (de) 1989-10-17 1997-04-15 Perkin Elmer Corp Robotische schnittstelle
US5646845A (en) 1990-02-05 1997-07-08 Caterpillar Inc. System and method for controlling an autonomously navigated vehicle
US8352400B2 (en) 1991-12-23 2013-01-08 Hoffberg Steven M Adaptive pattern recognition based controller apparatus and method and human-factored interface therefore
US5673367A (en) 1992-10-01 1997-09-30 Buckley; Theresa M. Method for neural network control of motion using real-time environmental feedback
CA2081519C (en) 1992-10-27 2000-09-05 The University Of Toronto Parametric control device
KR0161031B1 (ko) 1993-09-09 1998-12-15 김광호 로보트의 위치오차보정장치
US5602761A (en) 1993-12-30 1997-02-11 Caterpillar Inc. Machine performance monitoring and fault classification using an exponentially weighted moving average scheme
WO1997010081A1 (en) 1995-09-11 1997-03-20 Kabushiki Kaisha Yaskawa Denki Robot controller
US6169981B1 (en) 1996-06-04 2001-01-02 Paul J. Werbos 3-brain architecture for an intelligent decision and control system
US6366293B1 (en) 1998-09-29 2002-04-02 Rockwell Software Inc. Method and apparatus for manipulating and displaying graphical objects in a computer display device
US6243622B1 (en) 1998-10-16 2001-06-05 Xerox Corporation Touchable user interface using self movable robotic modules
EP1037134A2 (en) 1999-03-16 2000-09-20 Matsushita Electric Industrial Co., Ltd. Virtual space control data receiving apparatus and method
US6124694A (en) 1999-03-18 2000-09-26 Bancroft; Allen J. Wide area navigation for a robot scrubber
EP1092458A1 (en) 1999-04-30 2001-04-18 Sony Corporation Electronic pet system, network system, robot, and storage medium
JP3537362B2 (ja) 1999-10-12 2004-06-14 ファナック株式会社 ロボットシステム用グラフィック表示装置
WO2001067749A2 (en) 2000-03-07 2001-09-13 Sarnoff Corporation Camera pose estimation
EP1195231A4 (en) 2000-03-31 2006-01-25 Sony Corp ROBOT, METHOD FOR CONTROLLING A ROBOTIC ACTION, METHOD AND DEVICE FOR DETECTING EXTERNAL FORCE
US8543519B2 (en) 2000-08-07 2013-09-24 Health Discovery Corporation System and method for remote melanoma screening
JP4765155B2 (ja) 2000-09-28 2011-09-07 ソニー株式会社 オーサリング・システム及びオーサリング方法、並びに記憶媒体
US6678413B1 (en) 2000-11-24 2004-01-13 Yiqing Liang System and method for object identification and behavior characterization using video analysis
US6442451B1 (en) 2000-12-28 2002-08-27 Robotic Workspace Technologies, Inc. Versatile robot control system
JP2002239960A (ja) 2001-02-21 2002-08-28 Sony Corp ロボット装置の動作制御方法、プログラム、記録媒体及びロボット装置
US20020175894A1 (en) 2001-03-06 2002-11-28 Vince Grillo Hand-supported mouse for computer input
US6917925B2 (en) 2001-03-30 2005-07-12 Intelligent Inference Systems Corporation Convergent actor critic-based fuzzy reinforcement learning apparatus and method
JP2002301674A (ja) 2001-04-03 2002-10-15 Sony Corp 脚式移動ロボット及びその運動教示方法、並びに記憶媒体
DE60118317T2 (de) 2001-04-30 2006-12-14 Sony France S.A. Autonom Roboter
US6584375B2 (en) 2001-05-04 2003-06-24 Intellibot, Llc System for a retail environment
US6636781B1 (en) 2001-05-22 2003-10-21 University Of Southern California Distributed control and coordination of autonomous agents in a dynamic, reconfigurable system
JP3760186B2 (ja) 2001-06-07 2006-03-29 独立行政法人科学技術振興機構 二脚歩行式移動装置及びその歩行制御装置並びに歩行制御方法
JP4188607B2 (ja) 2001-06-27 2008-11-26 本田技研工業株式会社 二足歩行移動体の床反力推定方法及び二足歩行移動体の関節モーメント推定方法
WO2003007129A2 (en) 2001-07-13 2003-01-23 Broks Automation, Inc. Trajectory planning and motion control strategies for a planar three-degree-of-freedom robotic arm
US6710346B2 (en) 2001-08-02 2004-03-23 International Business Machines Corporation Active infrared presence sensor
CA2458979C (en) 2001-08-31 2010-09-28 George Danko Coordinated joint motion control system
US6812846B2 (en) 2001-09-28 2004-11-02 Koninklijke Philips Electronics N.V. Spill detector based on machine-imaging
US7243334B1 (en) 2002-01-16 2007-07-10 Prelude Systems, Inc. System and method for generating user interface code
JP3790816B2 (ja) 2002-02-12 2006-06-28 国立大学法人 東京大学 人型リンク系の運動生成方法
US8010180B2 (en) 2002-03-06 2011-08-30 Mako Surgical Corp. Haptic guidance system and method
US20040134336A1 (en) 2002-04-22 2004-07-15 Neal Solomon System, methods and apparatus for aggregating groups of mobile robotic vehicles
US7505604B2 (en) 2002-05-20 2009-03-17 Simmonds Precision Prodcuts, Inc. Method for detection and recognition of fog presence within an aircraft compartment using video images
US7834754B2 (en) 2002-07-19 2010-11-16 Ut-Battelle, Llc Method and system for monitoring environmental conditions
WO2004018158A2 (en) 2002-08-21 2004-03-04 Neal Solomon Organizing groups of self-configurable mobile robotic agents
KR100493159B1 (ko) * 2002-10-01 2005-06-02 삼성전자주식회사 이동체의 효율적 자기 위치 인식을 위한 랜드마크 및 이를이용한 자기 위치 인식 장치 및 방법
JP4079792B2 (ja) * 2003-02-06 2008-04-23 松下電器産業株式会社 ロボットの教示方法と教示機能付きロボット
AU2003900861A0 (en) 2003-02-26 2003-03-13 Silverbrook Research Pty Ltd Methods,systems and apparatus (NPS042)
JP3950805B2 (ja) 2003-02-27 2007-08-01 ファナック株式会社 教示位置修正装置
US7313279B2 (en) 2003-07-08 2007-12-25 Computer Associates Think, Inc. Hierarchical determination of feature relevancy
SE0301531L (sv) 2003-05-22 2004-11-23 Abb Ab A Control method for a robot
US7212651B2 (en) 2003-06-17 2007-05-01 Mitsubishi Electric Research Laboratories, Inc. Detecting pedestrians using patterns of motion and appearance in videos
US7769487B2 (en) 2003-07-24 2010-08-03 Northeastern University Process and architecture of robotic system to mimic animal behavior in the natural environment
WO2005028166A1 (ja) 2003-09-22 2005-03-31 Matsushita Electric Industrial Co., Ltd. 弾性体アクチュエータの制御装置及び制御方法
US7342589B2 (en) 2003-09-25 2008-03-11 Rockwell Automation Technologies, Inc. System and method for managing graphical data
JP4592276B2 (ja) 2003-10-24 2010-12-01 ソニー株式会社 ロボット装置のためのモーション編集装置及びモーション編集方法、並びにコンピュータ・プログラム
EP1724653B1 (en) 2004-02-25 2014-05-07 The Ritsumeikan Trust Control system of floating mobile body
JP4661074B2 (ja) 2004-04-07 2011-03-30 ソニー株式会社 情報処理システム、情報処理方法、並びにロボット装置
EP1622072B1 (en) 2004-07-27 2010-07-07 Sony France S.A. An automated action-selection system and method and application thereof for training prediction machines and for driving the development of self-developing devices
SE0402672D0 (sv) 2004-11-02 2004-11-02 Viktor Kaznov Ball robot
US7211979B2 (en) 2005-04-13 2007-05-01 The Broad Of Trustees Of The Leland Stanford Junior University Torque-position transformer for task control of position controlled robots
US7765029B2 (en) 2005-09-13 2010-07-27 Neurosciences Research Foundation, Inc. Hybrid control device
JP4876511B2 (ja) 2005-09-29 2012-02-15 株式会社日立製作所 ロジック抽出支援装置
EP2050544B1 (en) 2005-09-30 2011-08-31 iRobot Corporation Robot system with wireless communication by TCP/IP transmissions
US7668605B2 (en) 2005-10-26 2010-02-23 Rockwell Automation Technologies, Inc. Wireless industrial control user interface
KR101300493B1 (ko) 2005-12-02 2013-09-02 아이로보트 코퍼레이션 커버리지 로봇 이동성
US7741802B2 (en) 2005-12-20 2010-06-22 Intuitive Surgical Operations, Inc. Medical robotic system with programmably controlled constraints on error dynamics
KR101146119B1 (ko) * 2005-12-21 2012-05-16 재단법인 포항산업과학연구원 실내 이동 로봇의 위치 인식 방법 및 장치
US7953526B2 (en) * 2006-01-18 2011-05-31 I-Guide Robotics, Inc. Robotic vehicle controller
US8224018B2 (en) 2006-01-23 2012-07-17 Digimarc Corporation Sensing data from physical objects
US7576639B2 (en) 2006-03-14 2009-08-18 Mobileye Technologies, Ltd. Systems and methods for detecting pedestrians in the vicinity of a powered industrial vehicle
US8924021B2 (en) 2006-04-27 2014-12-30 Honda Motor Co., Ltd. Control of robots from human motion descriptors
JP4320363B2 (ja) 2006-05-25 2009-08-26 豪洋 石崎 作業ロボット
KR100791382B1 (ko) 2006-06-01 2008-01-07 삼성전자주식회사 로봇의 이동 경로에 따라 소정 영역의 특성에 관한 정보를수집하고 분류하는 방법 및 상기 영역 특성에 따라제어되는 로봇, 상기 영역 특성을 이용한 ｕｉ 구성 방법및 장치
US20080059015A1 (en) 2006-06-09 2008-03-06 Whittaker William L Software architecture for high-speed traversal of prescribed routes
US8843244B2 (en) 2006-10-06 2014-09-23 Irobot Corporation Autonomous behaviors for a remove vehicle
JP4699426B2 (ja) 2006-08-08 2011-06-08 パナソニック株式会社 障害物回避方法と障害物回避移動装置
US8174568B2 (en) 2006-12-01 2012-05-08 Sri International Unified framework for precise vision-aided navigation
JP4267027B2 (ja) 2006-12-07 2009-05-27 ファナック株式会社 ロボット制御装置
EP2479627A3 (en) 2007-03-29 2013-03-06 iRobot Corporation Robot operator control unit configuration system and method
US20080274683A1 (en) 2007-05-04 2008-11-06 Current Energy Controls, Lp Autonomous Ventilation System
US8255092B2 (en) 2007-05-14 2012-08-28 Irobot Corporation Autonomous behaviors for a remote vehicle
GB2452508A (en) * 2007-09-05 2009-03-11 Sony Corp Generating a three-dimensional representation of a sports game
JP5213023B2 (ja) 2008-01-15 2013-06-19 本田技研工業株式会社 ロボット
CN101952087B (zh) 2008-02-06 2013-01-16 松下电器产业株式会社 机器人、机器人的控制装置及控制方法
JP5181704B2 (ja) 2008-02-07 2013-04-10 日本電気株式会社 データ処理装置、姿勢推定システム、姿勢推定方法およびプログラム
US8175992B2 (en) 2008-03-17 2012-05-08 Intelliscience Corporation Methods and systems for compound feature creation, processing, and identification in conjunction with a data analysis and feature recognition system wherein hit weights are summed
KR101261148B1 (ko) 2008-04-02 2013-05-06 아이로보트 코퍼레이션 로보틱스 시스템
JP4715863B2 (ja) 2008-05-01 2011-07-06 ソニー株式会社 アクチュエータ制御装置及びアクチュエータ制御方法、アクチュエータ、ロボット装置、並びにコンピュータ・プログラム
JP5215740B2 (ja) 2008-06-09 2013-06-19 株式会社日立製作所 移動ロボットシステム
US20100174384A1 (en) 2008-09-04 2010-07-08 Iwalk, Inc. Hybrid terrain-adaptive lower-extremity systems
US20110282169A1 (en) 2008-10-29 2011-11-17 The Regents Of The University Of Colorado, A Body Corporate Long Term Active Learning from Large Continually Changing Data Sets
US20100114372A1 (en) 2008-10-30 2010-05-06 Intellibot Robotics Llc Method of cleaning a surface using an automatic cleaning device
JP5242342B2 (ja) 2008-10-31 2013-07-24 株式会社東芝 ロボット制御装置
US8428781B2 (en) 2008-11-17 2013-04-23 Energid Technologies, Inc. Systems and methods of coordination control for robot manipulation
EP2209018A1 (en) 2009-01-15 2010-07-21 Nederlandse Organisatie voor toegepast-natuurwetenschappelijk Onderzoek TNO A method for estimating an object motion characteristic from a radar signal, a computer system and a computer program product
US8423182B2 (en) 2009-03-09 2013-04-16 Intuitive Surgical Operations, Inc. Adaptable integrated energy control system for electrosurgical tools in robotic surgical systems
US8120301B2 (en) 2009-03-09 2012-02-21 Intuitive Surgical Operations, Inc. Ergonomic surgeon control console in robotic surgical systems
US8364314B2 (en) 2009-04-30 2013-01-29 GM Global Technology Operations LLC Method and apparatus for automatic control of a humanoid robot
JP4676544B2 (ja) 2009-05-29 2011-04-27 ファナック株式会社 工作機械に対してワークの供給及び取り出しを行うロボットを制御するロボット制御装置
US8694449B2 (en) 2009-05-29 2014-04-08 Board Of Trustees Of Michigan State University Neuromorphic spatiotemporal where-what machines
US8774970B2 (en) 2009-06-11 2014-07-08 S.C. Johnson & Son, Inc. Trainable multi-mode floor cleaning device
US8706297B2 (en) 2009-06-18 2014-04-22 Michael Todd Letsky Method for establishing a desired area of confinement for an autonomous robot and autonomous robot implementing a control system for executing the same
WO2011001569A1 (ja) 2009-07-02 2011-01-06 パナソニック株式会社 ロボット、ロボットアームの制御装置、及びロボットアームの制御プログラム
EP2284769B1 (en) 2009-07-16 2013-01-02 European Space Agency Method and apparatus for analyzing time series data
US20110026770A1 (en) 2009-07-31 2011-02-03 Jonathan David Brookshire Person Following Using Histograms of Oriented Gradients
US8250901B2 (en) 2009-09-22 2012-08-28 GM Global Technology Operations LLC System and method for calibrating a rotary absolute position sensor
US20110082735A1 (en) * 2009-10-06 2011-04-07 Qualcomm Incorporated Systems and methods for merchandising transactions via image matching in a content delivery system
TW201113815A (en) 2009-10-09 2011-04-16 Primax Electronics Ltd QR code processing method and apparatus thereof
US8679260B2 (en) 2009-11-11 2014-03-25 Intellibot Robotics Llc Methods and systems for movement of an automatic cleaning device using video signal
US8423225B2 (en) 2009-11-11 2013-04-16 Intellibot Robotics Llc Methods and systems for movement of robotic device using video signal
US8521328B2 (en) 2009-12-10 2013-08-27 The Boeing Company Control system for robotic vehicles
TW201123031A (en) 2009-12-24 2011-07-01 Univ Nat Taiwan Science Tech Robot and method for recognizing human faces and gestures thereof
JP5506618B2 (ja) 2009-12-28 2014-05-28 本田技研工業株式会社 ロボットの制御装置
JP5506617B2 (ja) 2009-12-28 2014-05-28 本田技研工業株式会社 ロボットの制御装置
US8668638B2 (en) 2010-02-11 2014-03-11 Intuitive Surgical Operations, Inc. Method and system for automatically maintaining an operator selected roll orientation at a distal tip of a robotic endoscope
KR101169674B1 (ko) 2010-03-11 2012-08-06 한국과학기술연구원 원격현전 로봇, 이를 포함하는 원격현전 시스템 및 이의 제어 방법
US8660355B2 (en) 2010-03-19 2014-02-25 Digimarc Corporation Methods and systems for determining image processing operations relevant to particular imagery
US9405975B2 (en) 2010-03-26 2016-08-02 Brain Corporation Apparatus and methods for pulse-code invariant object recognition
US9122994B2 (en) 2010-03-26 2015-09-01 Brain Corporation Apparatus and methods for temporally proximate object recognition
US9311593B2 (en) 2010-03-26 2016-04-12 Brain Corporation Apparatus and methods for polychronous encoding and multiplexing in neuronal prosthetic devices
US8336420B2 (en) 2010-06-02 2012-12-25 Disney Enterprises, Inc. Three-axis robotic joint using four-bar linkages to drive differential side gears
FR2963132A1 (fr) 2010-07-23 2012-01-27 Aldebaran Robotics Robot humanoide dote d'une interface de dialogue naturel, methode d'utilisation et de programmation de ladite interface
US20120045068A1 (en) 2010-08-20 2012-02-23 Korea Institute Of Science And Technology Self-fault detection system and method for microphone array and audio-based device
WO2012040554A2 (en) 2010-09-23 2012-03-29 Stryker Corporation Video monitoring system
KR20120035519A (ko) 2010-10-05 2012-04-16 삼성전자주식회사 먼지 유입 감지 유닛 및 이를 구비하는 로봇 청소기
US20120143495A1 (en) 2010-10-14 2012-06-07 The University Of North Texas Methods and systems for indoor navigation
US9015093B1 (en) 2010-10-26 2015-04-21 Michael Lamport Commons Intelligent control with hierarchical stacked neural networks
US8726095B2 (en) 2010-12-02 2014-05-13 Dell Products L.P. System and method for proactive management of an information handling system with in-situ measurement of end user actions
JP5185358B2 (ja) 2010-12-13 2013-04-17 株式会社東芝 行動履歴検索装置
JP5074640B2 (ja) 2010-12-17 2012-11-14 パナソニック株式会社 弾性体アクチュエータ駆動機構の制御装置及び制御方法、並びに、制御プログラム
US8380652B1 (en) 2011-05-06 2013-02-19 Google Inc. Methods and systems for autonomous robotic decision making
US8639644B1 (en) 2011-05-06 2014-01-28 Google Inc. Shared robot knowledge base for use with cloud computing system
US8908923B2 (en) * 2011-05-13 2014-12-09 International Business Machines Corporation Interior location identification
US9566710B2 (en) 2011-06-02 2017-02-14 Brain Corporation Apparatus and methods for operating robotic devices using selective state space training
US9189891B2 (en) 2011-08-16 2015-11-17 Google Inc. Systems and methods for navigating a camera
US9015092B2 (en) 2012-06-04 2015-04-21 Brain Corporation Dynamically reconfigurable stochastic learning apparatus and methods
US8798840B2 (en) 2011-09-30 2014-08-05 Irobot Corporation Adaptive mapping with spatial summaries of sensor data
US20130096719A1 (en) 2011-10-13 2013-04-18 The U.S.A. As Represented By The Administrator Of The National Aeronautics And Space Administration Method for dynamic optimization of a robot control interface
JP6305673B2 (ja) 2011-11-07 2018-04-04 セイコーエプソン株式会社 ロボット制御システム、ロボットシステム及びロボット
JP5399593B2 (ja) 2011-11-10 2014-01-29 パナソニック株式会社 ロボット、ロボットの制御装置、制御方法、及び制御プログラム
KR101305819B1 (ko) 2012-01-04 2013-09-06 현대자동차주식회사 착용식 로봇의 사용자 조작 의도 토크 추출방법
US8958911B2 (en) 2012-02-29 2015-02-17 Irobot Corporation Mobile robot
JP5895628B2 (ja) 2012-03-15 2016-03-30 株式会社ジェイテクト ロボットの制御方法及びロボット制御装置、並びにロボット制御システム
US9221177B2 (en) 2012-04-18 2015-12-29 Massachusetts Institute Of Technology Neuromuscular model-based sensing and control paradigm for a robotic leg
US9031779B2 (en) 2012-05-30 2015-05-12 Toyota Motor Engineering & Manufacturing North America, Inc. System and method for hazard detection and sharing
US9208432B2 (en) 2012-06-01 2015-12-08 Brain Corporation Neural network learning and collaboration apparatus and methods
KR101954144B1 (ko) * 2012-06-08 2019-03-05 엘지전자 주식회사 로봇 청소기와, 이의 제어 방법, 및 이를 포함한 로봇 청소 시스템
WO2013188579A1 (en) 2012-06-12 2013-12-19 Trx Systems, Inc. Wi-fi enhanced tracking algorithms
US8996174B2 (en) 2012-06-21 2015-03-31 Rethink Robotics, Inc. User interfaces for robot training
US20130346347A1 (en) 2012-06-22 2013-12-26 Google Inc. Method to Predict a Communicative Action that is Most Likely to be Executed Given a Context
BR112014033122A2 (pt) 2012-07-04 2017-06-27 Indra Sist S A sistema integrado de detecção prematura de hidrocarbonetos em um meio aquoso
US8977582B2 (en) 2012-07-12 2015-03-10 Brain Corporation Spiking neuron network sensory processing apparatus and methods
US9367798B2 (en) 2012-09-20 2016-06-14 Brain Corporation Spiking neuron network adaptive control apparatus and methods
US8793205B1 (en) 2012-09-20 2014-07-29 Brain Corporation Robotic learning and evolution apparatus
DE102012109004A1 (de) 2012-09-24 2014-03-27 RobArt GmbH Roboter und Verfahren zur autonomen Inspektion oder Bearbeitung von Bodenflächen
US9020637B2 (en) 2012-11-02 2015-04-28 Irobot Corporation Simultaneous localization and mapping for a mobile robot
US8972061B2 (en) 2012-11-02 2015-03-03 Irobot Corporation Autonomous coverage robot
US20140187519A1 (en) 2012-12-27 2014-07-03 The Board Of Trustees Of The Leland Stanford Junior University Biomarkers for predicting major adverse events
EP2752726B1 (de) 2013-01-08 2015-05-27 Cleanfix Reinigungssysteme AG Bodenbehandlungsmaschine und Verfahren zum Behandeln von Bodenflächen
JP6132659B2 (ja) 2013-02-27 2017-05-24 シャープ株式会社 周囲環境認識装置、それを用いた自律移動システムおよび周囲環境認識方法
US20140330496A1 (en) * 2013-02-28 2014-11-06 Austin Crouse Trainable robotic apparatus, system and method
US8958937B2 (en) 2013-03-12 2015-02-17 Intellibot Robotics Llc Cleaning machine with collision prevention
US9764468B2 (en) 2013-03-15 2017-09-19 Brain Corporation Adaptive predictor apparatus and methods
CN105050775B (zh) 2013-03-15 2017-08-25 直观外科手术操作公司 软件可配置的操纵器自由度
US9008840B1 (en) 2013-04-19 2015-04-14 Brain Corporation Apparatus and methods for reinforcement-guided supervised learning
US9292015B2 (en) 2013-05-23 2016-03-22 Fluor Technologies Corporation Universal construction robotics interface
US20140358828A1 (en) 2013-05-29 2014-12-04 Purepredictive, Inc. Machine learning generated action plan
US9242372B2 (en) 2013-05-31 2016-01-26 Brain Corporation Adaptive robotic interface apparatus and methods
TWI585561B (zh) 2013-06-03 2017-06-01 新智控私人有限公司 機器人裝置的離機式導航方法與裝置
KR101323705B1 (ko) * 2013-06-05 2013-11-11 한경대학교 산학협력단 무인 화물 이송로봇을 이용한 무인 화물 이송시스템
US9384443B2 (en) 2013-06-14 2016-07-05 Brain Corporation Robotic training apparatus and methods
US9792546B2 (en) 2013-06-14 2017-10-17 Brain Corporation Hierarchical robotic controller apparatus and methods
US20150032258A1 (en) 2013-07-29 2015-01-29 Brain Corporation Apparatus and methods for controlling of robotic devices
SG2013071808A (en) 2013-09-24 2015-04-29 Ctrlworks Pte Ltd Offboard navigation apparatus capable of being coupled to a movable platform
US9579789B2 (en) 2013-09-27 2017-02-28 Brain Corporation Apparatus and methods for training of robotic control arbitration
US9296101B2 (en) 2013-09-27 2016-03-29 Brain Corporation Robotic control arbitration apparatus and methods
US9315192B1 (en) 2013-09-30 2016-04-19 Google Inc. Methods and systems for pedestrian avoidance using LIDAR
US9144907B2 (en) 2013-10-24 2015-09-29 Harris Corporation Control synchronization for high-latency teleoperation
US9463571B2 (en) * 2013-11-01 2016-10-11 Brian Corporation Apparatus and methods for online training of robots
US10612939B2 (en) 2014-01-02 2020-04-07 Microsoft Technology Licensing, Llc Ground truth estimation for autonomous navigation
US10078712B2 (en) 2014-01-14 2018-09-18 Energid Technologies Corporation Digital proxy simulation of robotic hardware
US9567077B2 (en) 2014-02-14 2017-02-14 Accenture Global Services Limited Unmanned vehicle (UV) control system
US9346167B2 (en) 2014-04-29 2016-05-24 Brain Corporation Trainable convolutional network apparatus and methods for operating a robotic vehicle
US10255319B2 (en) 2014-05-02 2019-04-09 Google Llc Searchable index
US20150339589A1 (en) 2014-05-21 2015-11-26 Brain Corporation Apparatus and methods for training robots utilizing gaze-based saliency maps
US9475195B2 (en) 2014-09-12 2016-10-25 Toyota Jidosha Kabushiki Kaisha Anticipatory robot navigation
WO2016168944A1 (en) 2015-04-24 2016-10-27 Avidbots Corp. Apparatus and methods for semi-autonomous cleaning of surfaces
US20170139551A1 (en) * 2015-11-16 2017-05-18 Bank Of America Corporation System for determining user interfaces to display based on user location
US20170329347A1 (en) * 2016-05-11 2017-11-16 Brain Corporation Systems and methods for training a robot to autonomously travel a route

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5341540A (en) * 1989-06-07 1994-08-30 Onet, S.A. Process and autonomous apparatus for the automatic cleaning of ground areas through the performance of programmed tasks
US20090228166A1 (en) * 2006-01-18 2009-09-10 I-Guide, Llc Robotic Vehicle Controller
US20160287044A1 (en) * 2013-09-18 2016-10-06 Murata Machinery, Ltd. Autonomously traveling floor washer, cleaning schedule data structure, storage medium, method for creating cleaning schedule, and program
US20160121487A1 (en) * 2014-11-03 2016-05-05 Qualcomm Incorporated Communicating Configurable Instruction Sets to Robots for Controlling Robot Behavior

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2021509986A (ja) * 2018-01-02 2021-04-08 ゼネラル・エレクトリック・カンパニイ 人間による実演に基づく、ロボットによる産業タスクの学習のためのシステム及び方法
JP7391023B2 (ja) 2018-01-02 2023-12-04 ゼネラル・エレクトリック・カンパニイ 人間による実演に基づく、ロボットによる産業タスクの学習のためのシステム及び方法
JP2019179422A (ja) * 2018-03-30 2019-10-17 日立造船株式会社 飛行体、情報記録方法、飛行体の制御方法、および制御プログラム
JP7173746B2 (ja) 2018-03-30 2022-11-16 日立造船株式会社 飛行体、情報記録方法、飛行体の制御方法、および制御プログラム
EP3872528A4 (en) * 2018-10-22 2021-12-29 Ecovacs Robotics Co., Ltd. Travel control method, device, and storage medium
US11487297B2 (en) 2018-10-22 2022-11-01 Ecovacs Robotics Co., Ltd. Method of travel control, device and storage medium
US12253858B2 (en) 2018-10-22 2025-03-18 Ecovacs Robotics Co., Ltd. Method of travel control, device and storage medium
DE102021204149A1 (de) 2021-04-27 2022-10-27 BSH Hausgeräte GmbH Objekterkennung für ein Hausgerät
CN114578379A (zh) * 2021-10-12 2022-06-03 苏州趣洁电器有限公司 识别柔性障碍物的方法、装置及自行走设备

Also Published As

Publication number Publication date
JP2019515407A (ja) 2019-06-06
EP3455686A4 (en) 2020-02-19
US20170329333A1 (en) 2017-11-16
US10241514B2 (en) 2019-03-26
CA3023557A1 (en) 2017-11-16
US20190171210A1 (en) 2019-06-06
US11803185B2 (en) 2023-10-31
US20220083058A1 (en) 2022-03-17
JP6942177B2 (ja) 2021-09-29
KR102444658B1 (ko) 2022-09-20
KR20190030197A (ko) 2019-03-21
CN109416538A (zh) 2019-03-01
EP3455686A1 (en) 2019-03-20

Similar Documents

Publication Publication Date Title
US11803185B2 (en) Systems and methods for initializing a robot to autonomously travel a trained route
US12393196B2 (en) Systems and methods for training a robot to autonomously travel a route
Van den Bergh et al. Real-time 3D hand gesture interaction with a robot for understanding directions from humans
Davison et al. MonoSLAM: Real-time single camera SLAM
US10762386B2 (en) Method of determining a similarity transformation between first and second coordinates of 3D features
Wilson Depth-sensing video cameras for 3d tangible tabletop interaction
US12477093B2 (en) Wide viewing angle stereo camera apparatus and depth image processing method using the same
EP1926004A2 (en) Method, apparatus, and medium for controlling mobile device based on image of real space including the mobile device
Frank et al. Toward mobile mixed-reality interaction with multi-robot systems
Tölgyessy et al. Foundations of visual linear human–robot interaction via pointing gesture navigation
CN110603122A (zh) 用于交互式学习应用的自动个性化反馈
US12053892B2 (en) Systems, apparatuses and methods for removing false positives from sensor detection
CN116295401A (zh) 一种室内纯视觉机器人障碍物感知方法
Gulalkari et al. Object following control of six-legged robot using Kinect camera
JP7354528B2 (ja) 自律移動装置、自律移動装置のレンズの汚れ検出方法及びプログラム
CN119839883B (zh) 智能家居机器人及其任务执行方法和存储介质
Magrabi Simulation of collision avoidance by navigation assistance using stereo vision
Dönmez et al. Multi-Camera Vision-Based Mobile Robot Control and Path Planning
WO2024226946A2 (en) Systems and methods for feature detection de-duplication and panoramic image generation
Schwab Object avoidance and wall following using the Kinect
Niemantsverdriet Using a 3-dimensional model for navigation and correction of odometry errors in indoor robotics

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 3023557

Country of ref document: CA

ENP Entry into the national phase

Ref document number: 2019511832

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17796891

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 20187035946

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2017796891

Country of ref document: EP

Effective date: 20181211