Robotic cleaning appliance, sensor processing unit and surface type detection method

Abstract

A robotic cleaning device includes an acoustic transducer and a processor coupled to a housing. The acoustic transducer transmits an acoustic signal toward a surface within its ringing distance and receives a corresponding returned signal. After the acoustic signal stops, the processor samples the ringing signal generated by the acoustic transducer during an early portion before the corresponding returned signal is reflected back to the acoustic transducer and during a later portion including the corresponding returned signal. The processor uses the sampled early portion to estimate a blank ringing signal, which represents the performance of the acoustic transducer in the absence of a returned signal. The processor compares the estimated blank ringing signal with the later portion of the ringing signal and generates a metric based on the comparison. The processor uses the metric to determine the surface type from a plurality of surface types.

Description

[0001] Cross-references to related applications

[0002] This application claims priority and benefit to co-pending U.S. Patent Application No. 17 / 226,447 entitled "Surface Detection," filed April 9, 2021, by Mitchell H. Kline et al., Attorney No. IVS-982, which has been assigned to the assignee of this application, the disclosure of which is hereby incorporated herein by reference in its entirety. Technical Field

[0003] This application relates to ultrasonic sensors and ultrasonic sensing, and more specifically to using ultrasonic sensing to detect the type of surface on which a device operates. Background Technology

[0004] There are various devices that move or operate on floors or other surfaces such as walls, windows, roofs, tables, countertops, sidewalks, roads, etc. These surfaces can be indoor surfaces, outdoor surfaces, or some combination thereof. One or more examples of such devices can be semi-autonomous, meaning that when operating on one or more surfaces, some functions of the device are controlled by a human while others are automated. One or more examples of such devices can be robots, meaning that when operating on one or more surfaces, some or all of the device's functions can operate autonomously under the control of one or more processors. Some examples of these devices may include, but are not limited to: remotely controlled vehicles, telepresence robots, electric scooters, electric wheelchairs, wheeled delivery robots, drones that operate near or are about to land on or take off from surfaces, wheeled delivery vehicles, floor vacuum cleaners, and robotic cleaning appliances (including robotic floor cleaners and / or robotic floor vacuum cleaners). Summary of the Invention

[0005] In some aspects, the technology described herein relates to a robotic cleaning appliance, comprising: a housing, a surface treatment item coupled to the housing; and an acoustic transducer coupled to the housing, the acoustic transducer being configured to transmit acoustic signals toward a surface below the robotic cleaning appliance and to receive corresponding returned signals reflected from the surface, wherein the surface is within a ringing distance associated with the acoustic transducer; and a processor coupled to the housing, the processor being configured to: sample multiple points of the ringing signal generated by the acoustic transducer after the active transmission of the acoustic signal has ceased, wherein the sampling occurs... During an early portion of the ringing signal before the corresponding returned signal from the surface has been reflected back and received by the acoustic transducer, and the sampling also occurs during a later portion of the ringing signal including the corresponding returned signal; the early portion of the sampling is used to estimate a blank ringing signal for the acoustic transducer, wherein the estimated blank ringing signal represents the performance of the acoustic transducer when no returned signal is received; the estimated blank ringing signal is compared with the later portion of the ringing signal; a metric is generated based on the comparison; and the metric is used to determine the type of the surface from a plurality of surface types.

[0006] In some aspects, the technology described herein relates to a sensor processing unit comprising: an acoustic transducer configured to transmit an acoustic signal toward a surface and receive a corresponding returned signal reflected from the surface, wherein the surface is within a ringing distance associated with the acoustic transducer; and a sensor processor coupled to the acoustic transducer, the sensor processor being configured to: after the active transmission of the acoustic signal ceases, sample multiple points of the ringing signal generated by the acoustic transducer, wherein the sampling occurs from the corresponding returned signal from the surface. The sampling occurs during an early portion of the ringing signal before it is reflected back and received by the acoustic transducer, and also during a later portion of the ringing signal including the corresponding returned signal; the early portion of the signal is used with multiple sampling points to estimate a blank ringing signal for the acoustic transducer, wherein the estimated blank ringing signal represents the performance of the acoustic transducer without receiving any returned signal; the estimated blank ringing signal is compared with the later portion of the ringing signal; a metric is generated based on the comparison; and the metric is used to determine the type of the surface from a plurality of surface types.

[0007] In some aspects, the technology described herein relates to a surface type detection method, comprising: receiving a returned signal from an acoustic transducer by a processor coupled to the acoustic transducer, wherein the acoustic transducer is configured to send an acoustic signal to a surface and receive a returned signal reflected from the surface, wherein the returned signal corresponds to the sent acoustic signal, and wherein the surface is within a ringing distance associated with the acoustic transducer; in response to the cessation of active transmission of the acoustic signal, sampling by the processor at multiple points of the ringing signal generated by the acoustic transducer, wherein the sampling occurs after the returned signal from the surface has been reflected. The sampling occurs during an early portion of the ringing signal before it is received by the acoustic transducer, and also during a later portion of the ringing signal including the returned signal; the processor uses the early portion of multiple sampling points to estimate a blank ringing signal for the acoustic transducer, wherein the estimated blank ringing signal represents the performance of the acoustic transducer when no returned signal is received; the processor compares the estimated blank ringing signal with the later portion of the ringing signal; the processor generates a metric based on the comparison; and the processor uses the metric to determine the type of the surface from a plurality of surface types. Attached Figure Description

[0008] The accompanying drawings, which are incorporated into and form part of the description of the specific embodiments, illustrate various embodiments of the subject matter and, together with the description of the embodiments, serve to explain the principles of the subject matter discussed below. Unless otherwise specified, the drawings mentioned in this brief description should be understood as not being drawn to scale. Throughout this document, the same items are labeled with the same item numbers.

[0009] Figure 1A and Figure 1B Example block diagrams are shown of some aspects of a device that moves or operates around on a surface according to various embodiments.

[0010] Figure 2 A top front perspective view of an example of a system according to various embodiments is shown, the system including a device that moves or operates on a surface and a base station for the device.

[0011] Figures 3A-3C Various embodiments are shown. Figure 2 A side view of an example system, which includes a device that moves or operates around on a surface and a base station for the device.

[0012] Figure 4A A side view of an example surface type detection sensor assembly, according to various embodiments, is shown. It can be used on a device to transmit acoustic signals and receive returned acoustic signals.

[0013] Figure 4B Various embodiments are shown. Figure 4A A top view of the surface type detection sensor assembly.

[0014] Figure 4C A side cross-sectional view of the configuration of a surface type detection sensor assembly according to some embodiments is shown.

[0015] Figure 5 A top-plan view of the span of several different floor surface types according to various embodiments is shown.

[0016] Figure 6 A block diagram illustrating the signal path of a surface type detection sensor according to various embodiments is shown.

[0017] Figure 7A A graph showing the amplitude of the sampled returned signal from the surface type detection sensor, according to an embodiment, is shown.

[0018] Figure 7B A graph is shown illustrating curve fitting of the returned signal sampled from an early portion of the ringdown signal from a surface type detection sensor, according to an embodiment, and used to estimate the blank ringdown signal of the surface type detection sensor.

[0019] Figure 7C A graph showing the phases of two ringing signals from a surface type detection sensor, according to an embodiment, is shown.

[0020] Figure 8 A block diagram is shown for determining the curve length in the complex domain using ringing signals and blank ringing signals for a surface type detection sensor according to an embodiment.

[0021] Figure 9 The blank ringing signal and the measured ringing signal curve length in the complex domain of the surface type detection sensor according to an embodiment are shown.

[0022] Figure 10 A diagram illustrating how classifiers, according to various embodiments, are trained to detect surface types based on input is shown.

[0023] Figure 11 A diagram is shown of a trained classifier for detecting surface types based on input, according to various embodiments.

[0024] Figure 12A-12B A flowchart of an example method for surface type detection according to various embodiments is shown. Detailed Implementation

[0025] Reference will now be made in detail to various embodiments of the subject matter, examples of which are illustrated in the accompanying drawings. While various embodiments are discussed herein, it will be understood that they are not intended to limit these embodiments. Rather, the proposed embodiments are intended to cover alternatives, modifications, and equivalents that may be included within the spirit and scope of the various embodiments as defined by the appended claims. Furthermore, numerous specific details are set forth in the description of the embodiments to provide a thorough understanding of embodiments of the subject matter. However, embodiments may be practiced without these specific details. In other instances, well-known methods, processes, components, and circuits have not been described in detail so as not to unnecessarily obscure aspects of the described embodiments.

[0026] Discussion Overview

[0027] Various devices that move or operate on floors or other surfaces can benefit from the ability to determine information about the surface on which they move or operate. Surfaces can include floors, building exteriors, windows, soil, gravel, fabric, roofs, roads, sidewalks, paths, water (or other liquids), etc. Surfaces can be flat and smooth or rough and textured. In some embodiments, the surface can be fixed while the device operates on or moves on it. In other embodiments, the surface is movable while the device remains fixed or moves relative to the moving surface. A wide variety of “surface types” exist on which such devices can operate or move. These “surface types” can be divided into two main groups, hard surfaces and soft surfaces, according to some aspects. In some embodiments, subgroups may exist within each of these two main groups. In some embodiments, additional or alternative “main groups” may exist, such as wet, dry, clean, dirty, etc.

[0028] For illustrative purposes and not for limitation, consider floor surfaces that can be divided into hard floors and soft floors. Hard floors include, but are not limited to, floor surfaces such as: tile, wood, linoleum, laminate flooring, metal, cemented stone, concrete, stone, etc. Soft floors include, but are not limited to: carpet, sculpted carpet, low-pile carpet, cut-pile carpet, high-pile carpet, and other types of carpet, etc. Although flooring has been described as falling into two main categories, hard and soft, floor surfaces can be divided into any number of categories ranging from hard to soft (e.g., hard, medium-hard, medium-soft, soft, etc.).

[0029] During operation on a surface or as the device moves across the surface, it may be advantageous for the device to detect information about the surface, such as whether it is wet (and in some cases, how wet it is on a moisture scale), dirty (and in some cases, how dirty it is and / or with what kind of dirt), dry, hard (and in some cases, how hard it is on a hardness scale), soft (and in some cases, how soft it is on a softness scale), etc. As will be described herein, such detection of surface type may include sending acoustic signals to the surface and processing the returned acoustic signals to determine the surface type. These acoustic signals may include signals in one or more of the infrasound, acoustic, and ultrasonic ranges. The returned signals include directly returned signals (transmitted directly to the receiver / transducer and bounced) and secondary returned signals (multipath reflections before reaching the receiver / transducer). When an acoustic transducer is driven with a drive waveform to oscillate and send an acoustic signal, it will continue to oscillate for a period of time after the drive waveform stops. When the driving waveform stops, active transmission from the acoustic transducer ceases, but the transducer will continue to oscillate for a period of time when it rings (e.g., when the driven oscillation decays). This period of time associated with the decay or ringing of the driven oscillation can be referred to as the ringing time of the acoustic transducer. In some embodiments, the ringing time can be similar in length to the period during which the acoustic transducer is actively driven immediately before ringing begins. Because the transducer continues to passively oscillate during the ringing time, these passive oscillations can make it difficult to distinguish the returned signal received during the ringing time, effectively blinding the acoustic transducer to objects that provide the returned signal arriving at the acoustic transducer during the ringing period.

[0030] For example, if the sensor is close to a surface (such as a floor), the round-trip time of the returned signal (especially the primary returned signal) may be very small, causing it to arrive during the later part of the ringing period. In such a case, the sensed surface is within the transducer's ringing distance because the returned signal is received while ringing is still occurring. Because of this, the detection of the rapidly received returned signal from the surface within the ringing distance may be impeded or otherwise affected by the transducer's ringing. Even if the returned signal is very strong, it may be masked by the larger amplitude and / or interference caused by the ringing of the acoustic transducer used as a surface type detection sensor. In other words, the acoustic transducer in the ringing may still vibrate with sufficient amplitude to effectively mask the relatively small returned signal it receives during a portion of its ringing period.

[0031] Some embodiments described herein depict techniques for identifying a returned signal from a ringing signal during the ringing period of an acoustic transducer. For example, a “void ringing signal” can be estimated by sampling points of the ringing signal very early in the ringing period before the returned signal arrives and then extrapolating those points using an attenuation factor. This estimated void ringing signal represents what the pure ringing signal from the acoustic transducer would look like without any returned signal arriving and affecting it with constructive or destructive interference. This estimated void ringing signal can then be compared to the actual ringing signal that occurs, and this comparison can be used to generate a metric that can be used to determine the type of surface providing the returned signal received during the ringing period of the acoustic transducer. The metric can be the result of background subtraction, magnitude comparison, curve length comparison, threshold comparison, and / or other comparisons. This technique for comparing a portion of the actually received ringing signal with the estimated void ringing signal can be used with an amplitude signal or with one or more demodulated components (e.g., in-phase or quadrature signals in the complex domain).

[0032] The discussion begins with a description of notation and naming. It then moves to a description of block diagrams of example components of some example devices that move or operate on surfaces. Example depictions of devices in the form of robotic cleaning implements are discussed. Example depictions of surface type detection sensor components are described. The signal path of the acoustic transducer is described. A graph depicting the estimation of a blank ringing signal from an amplitude signal and comparing the two to form a metric is discussed. A block diagram of generating a blank ringing signal in the complex domain and comparing it to a portion of the actual ringing signal in the complex domain is discussed. A graph depicting the comparison of the ringing signal and the blank ringing signal through curve lengths in the complex domain is described, and the metric is discussed. The use of the generated metric and other information to train a classifier and perform machine learning is discussed. Finally, the operation of the robotic cleaning implement and its components (including the acoustic transducer and processor) is discussed in conjunction with a description of an example method for surface type detection.

[0033] Symbols and naming

[0034] Some parts of the following detailed description are presented using other symbolic representations of programs, logic blocks, procedures, modules, and operations on data bits within computer memory. These descriptions and representations are means used by those skilled in the art of data processing to most effectively communicate the substance of their work to others skilled in the art. In this application, programs, logic blocks, procedures, modules, etc., are contemplated as one or more inconsistent programs or instructions that result in desired results. These programs are those that require physical manipulation of physical quantities. Typically, although not essential, these quantities take the form of electrical or magnetic signals that can be stored, transmitted, combined, compared, and otherwise manipulated in electronic devices / components.

[0035] However, it should be remembered that all these and similar terms will be associated with appropriate physical quantities and are merely convenient labels applied to those quantities. Unless otherwise specifically stated, as will be apparent from the following discussion, it should be understood that throughout the description of the embodiments, the discussion uses terms such as “transmit,” “receive,” “sample,” “estimate,” “compare,” “generate,” “store,” “provide,” “classify,” “utilize,” and “determine” to refer to the actions and processes of electronic devices or components (such as: host processors, sensor processing units, sensor processors, digital signal processors or other processors, memories, surface type detection sensors (e.g., acoustic transducers), robotic cleaning appliances, devices configured to operate on or move around on surfaces, some combinations thereof, etc.). Electronic devices / components manipulate and convert data represented as physical (electronic and / or magnetic) quantities within registers and memories into other data similarly represented as stored, transmitted, processed, or displayed physical quantities within the component.

[0036] The embodiments described herein can be discussed in the general context of processor-executable instructions (such as program modules or logic) residing on some form of non-transitory processor-readable medium, which are executed by one or more computers, processors, or other devices. Typically, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. The functionality of program modules can be combined or distributed as needed in various embodiments.

[0037] In the accompanying drawings, a single block may be described as performing one or more functions; however, in practice, the one or more functions performed by that block may be performed in a single component or across multiple components, and / or may be performed using hardware, software, or a combination of hardware and software. To clearly illustrate this interchangeability between hardware and software, various exemplary components, blocks, modules, circuits, and steps have been described in general terms of their functionality. Whether these functionalities are implemented as hardware or software depends on the specific application and the design constraints imposed on the system as a whole. Those skilled in the art may implement the described functionalities in different ways for each specific application, but such implementation decisions should not be construed as departing from the scope of this disclosure. Furthermore, the example electronic devices described herein may include components other than those shown, including well-known components.

[0038] Unless specifically described as implemented in a particular manner, the techniques described herein can be implemented in hardware, or a combination of hardware and firmware and / or software. Any feature described as a module or component may also be implemented together in an integrated logic device, or separately as a discrete but interoperable logic device. If implemented in software, the techniques may be implemented at least in part by a non-transitory computer-readable storage medium comprising computer / processor-readable instructions that, when executed, cause a processor and / or other component of a computer or electronic device to perform one or more of the methods described herein. The non-transitory processor-readable data storage medium may form part of a computer program product, which may include packaging material.

[0039] Non-transitory processor-readable storage media (also known as non-transitory computer-readable storage media) may include random access memory (RAM) such as synchronous dynamic random access memory (SDRAM), read-only memory (ROM), non-volatile random access memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), flash memory, and other known storage media. Alternatively or additionally, this technology may be implemented at least in part by a processor-readable communication medium that carries or conveys code in the form of instructions or data structures and can be accessed, read, and / or executed by a computer or other processor.

[0040] The various illustrative logic blocks, modules, circuits, and instructions described in conjunction with the embodiments disclosed herein can be executed by one or more processors (such as one or more host processors or one or more cores thereof), digital signal processors (DSPs), general-purpose microprocessors, application-specific integrated circuits (ASICs), application-specific instruction set processors (ASIPs), field-programmable gate arrays (FPGAs), or other equivalent integrated or discrete logic circuit systems. As used herein, the term "processor" can refer to any of the foregoing structures or any other structure suitable for implementing the techniques described herein. Additionally, in some aspects, the functionality described herein may be provided within dedicated software or hardware modules configured as described herein. Moreover, the technology can be entirely implemented within one or more circuit or logic elements. A general-purpose processor may be a microprocessor, but alternatively, it may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, such as multiple microprocessors, one or more microprocessors combined with an ASIC or DSP, or any other such configuration or suitable combination of processors.

[0041] In the various example embodiments discussed herein, the chip is defined as comprising at least one substrate typically formed of a semiconductor material. A single chip may be formed, for example, from multiple substrates, wherein the substrates are mechanically bonded to maintain functionality. Multiple chips (or multiple chips) comprise at least two substrates, wherein the two substrates are electrically connected but not mechanically bonded.

[0042] A package provides electrical connections between bonding pads and metal leads on a chip (or, for example, a multi-chip module) that can be soldered onto a printed circuit board (PCB). A package typically includes a substrate and a cover. An integrated circuit (IC) substrate can refer to a silicon substrate with circuitry (typically CMOS circuitry), but other circuitry is possible and anticipated. A MEMS substrate provides mechanical support for one or more MEMS structures. The MEMS structure layers are attached to the MEMS substrate. MEMS substrates are also referred to as handle substrates or handle wafers. In some embodiments, the handle substrate acts as a cap for the MEMS structure.

[0043] Some embodiments may include, for example, a surface type detection sensor 150. This sensor can be any suitable acoustic sensor operating within any suitable acoustic range. For example, in some embodiments, the surface type detection sensor can be an ultrasonic sensor utilizing a MEMS ultrasonic transducer. In some embodiments, the surface type detection sensor may include a digital signal processor (DSP), which may be configured as part of an ASIC that may be integrated into the same package as the transducer. An example of such an ultrasonic sensor that can be used with various embodiments (not limited thereto) is the CH101 ultrasonic distance sensor from Chirp Microsystems, a TDK Group company in Berkeley, California. The CH101 is merely one example of an ultrasonic sensor, and other types and / or brands of ultrasonic sensors may be similarly utilized.

[0044] Some embodiments may include, for example, one or more motion sensors. For instance, an embodiment with accelerometers, gyroscopes, and magnetometers or other compass technologies may be referred to as a 9-axis device, each providing measurements along three axes orthogonal to each other. In another embodiment, a three-axis accelerometer and a three-axis gyroscope may be used to form a 6-axis device. Other embodiments may include, for example, accelerometers, gyroscopes, compasses, and pressure sensors, and may be referred to as a 10-axis device. Other embodiments may not include all sensors or may provide measurements along one or more axes. Some or all of the sensors may be MEMS sensors. Some or all of the sensors may be integrated with a sensor processor in a sensor processing unit and arranged in a single semiconductor package.

[0045] In some embodiments, for example, one or more sensors may be formed on a first substrate. Various embodiments may include, for example, solid-state sensors and / or any other type of sensor. Electronic circuitry in the sensor processing unit may, for example, receive measurement outputs from one or more sensors. In various embodiments, the electronic circuitry processes sensor data. The electronic circuitry may, for example, be implemented on a second silicon substrate. In some embodiments, the first substrate may be vertically stacked, attached, and electrically connected to the second substrate in a single semiconductor chip, while in other embodiments, the first substrate may be laterally arranged and electrically connected to the second substrate in a single semiconductor package (such as a single integrated circuit).

[0046] Example devices that move around on the surface or operate on the surface.

[0047] Figure 1A and Figure 1BSome example components of device 100 that moves around or operates on a surface are shown. Some examples of device 100 may include, but are not limited to: remotely controlled vehicles, telepresence robots, electric scooters, electric wheelchairs, wheeled delivery robots, drones that operate near or are about to land on or take off from a surface, wheeled delivery vehicles, floor vacuum cleaners, and robotic cleaning appliances (including: robotic floor cleaners, robotic floor vacuum cleaners, or combinations thereof).

[0048] Figure 1A A block diagram of the components of an example device 100A that moves or operates on a surface according to various aspects of this disclosure is shown. As shown, the example device 100A includes a communication interface 105, a host processor 110, a host memory 111, and at least one surface type detection sensor 150. In some embodiments, the device 100 may additionally include one or more of a transceiver 113, one or more motion sensors 160, one or more drive wheel controllers 170, and one or more surface treatment controllers 180 (which can control cleaning tools and / or any surface treatment items that can determine actions based on surface type). Some embodiments may include sensors for detecting motion, position, surface type, or environmental background (e.g., nearby objects and / or obstacles, whether the surface is hard or soft, whether the surface is carpeted or not, whether the surface is clean or dirty, whether the surface is wet or dry, etc.); some examples of these sensors may include, but are not limited to, infrared sensors, cameras, microphones, and Global Navigation Satellite System sensors (i.e., Global Positioning System receivers). Figure 1A As described herein, the components included are communicatively coupled to each other, for example, via communication interface 105.

[0049] The host processor 110 may be configured, for example, to perform various calculations and operations related to the general functions of the device 100 (e.g., sending commands to move, turn, avoid obstacles, and operate / control tools). The host processor 110 may be one or more microprocessors, central processing units (CPUs), DSPs, general-purpose microprocessors, ASICs, ASIPs, FPGAs, or other processors that run software programs or applications, which may be stored in the host memory 111 and associated with the general and conventional functions and capabilities of the device 100.

[0050] Communication interface 105 can be any suitable bus or interface, such as a Fast Peripheral Component Interconnect (PCIe) bus, Universal Serial Bus (USB), Universal Asynchronous Receiver / Transmitter (UART) serial bus, a suitable Advanced Microcontroller Bus Architecture (AMBA) interface, an Internal Integrated Circuit (I2C) bus, a Serial Digital Input / Output (SDIO) bus, or other equivalents, and may include multiple communication interfaces. Communication interface 105 can facilitate communication between SPU 120 and one or more of the following: host processor 110, host memory 111, transceiver 113, surface type detection sensor 150, one or more motion sensors 160, one or more drive wheel controllers 170, and / or one or more surface treatment controllers 180.

[0051] The host memory 111 may include programs, modules, applications, or other data used by the host processor 110. In some embodiments, the host memory 111 may also store information received from or provided to the sensor processing unit 120 (see, for example...). Figure 1B The host memory 111 can be any suitable type of memory, including but not limited to electronic memory (e.g., read-only memory (ROM), random access memory (RAM) or other electronic memory).

[0052] Transceiver 113 (when included) may be one or more wired or wireless transceivers that facilitate the reception of data from an external transmitting source and the transmission of data from device 100 to an external receiver. Examples of external transmitting sources / receivers may be base stations to which device 100 returns for charging, maintenance, docking, etc. By way of example and not limitation, in various embodiments, transceiver 113 may include one or more of the following: cellular transceivers, wireless LAN transceivers (e.g., transceivers conforming to one or more IEEE 802.11 specifications for wireless LAN communications), wireless personal area network transceivers (e.g., transceivers conforming to one or more IEEE 802.15 specifications (or similar) for wireless personal area network communications), and wired serial transceivers (e.g., Universal Serial Bus for wired communications).

[0053] Surface type detection sensor 150 may be an acoustic transducer. In some embodiments, surface type detection sensor 150 is an ultrasonic transducer (i.e., an acoustic transducer operating in the ultrasonic frequency range). In some embodiments, where surface type detection sensor 150 operates in the ultrasonic range, it may operate in the range between 50 kHz and 500 kHz or in the range between 150 kHz and 200 kHz. Of course, other ultrasonic ranges are contemplated and available. Surface type detection sensor 150 is configured to send an acoustic signal to a surface and receive an acoustic return signal. The sent acoustic signal may include signals in one or more of the infrasonic, acoustic, and ultrasonic ranges. The returned signal includes a direct return signal (which is sent, encounters the surface, and is reflected directly from the surface to the receiver) and a secondary return signal (which is reflected by multiple paths before reaching the receiver). In some embodiments, surface type detection sensor 150 may be a surface type detection sensor assembly 350 (e.g., see...). Figures 4A-4C As part of the surface type detection sensor assembly 350, the surface type detection sensor assembly 350 may include an acoustic interface 450 (e.g., see...). Figures 4A-4C Acoustic interface 450 (e.g., tube, cavity, angle, some combination thereof, etc.) is used to direct the transmitted acoustic wave signal toward the surface and to direct the acoustic wave return signal back to the surface type detection sensor 150.

[0054] Motion sensor 160 (when included) can be implemented as a MEMS-based motion sensor, including inertial sensors (such as a gyroscope 161 or an accelerometer 163) or electromagnetic sensors (such as a Hall effect or Lorentz field magnetometer 165). In some embodiments, at least a portion of motion sensor 160 may also be based, for example, on sensor technologies other than MEMS technology (e.g., CMOS technology, etc.). As desired, one or more motion sensors of motion sensor 160 can be configured to provide raw data output of measurements along three orthogonal axes or any equivalent structure.

[0055] The drive wheel controller 170 or other motion control mechanism (when included) may include a motor controller, switches and / or logic that operate on command to: drive one or more wheels or other motion mechanisms (e.g., box tracks, propellers), change the rotational speed of the drive wheels or other motion mechanisms, moderate the allowable slip or rotation of the drive wheels, move the drive wheels or other motion mechanisms in a desired direction, stop the drive wheels or other motion mechanisms, and / or steer the device 100 using the drive wheels (such as via differential speed or rotation) or other motion mechanisms.

[0056] One or more surface treatment controllers 180 (when included) may include motor controllers, switches and / or logic to turn on, off and / or adjust the operation and / or orientation of one or more surface treatment items (such as determining one or more cleaning tools and / or one or more other items to take action based on surface type).

[0057] Figure 1B A block diagram of the components of an example device 100B, which moves around or operates on a surface according to various aspects of this disclosure, is shown. Device 100B is similar to device 100A except that it includes a sensor processing unit (SPU) 120, in which a surface-type detection sensor 150 is disposed. SPU 120 (when included) includes: a sensor processor 130; internal memory 140; and at least one surface-type detection sensor 150. In some embodiments, SPU 120 may additionally include one or more motion sensors 160 (e.g., gyroscope 161, accelerometer 163, magnetometer 165) and / or one or more other sensors, such as light sensors, infrared sensors, GNSS sensors, microphones, etc. In various embodiments, SPU 120 or a portion thereof (e.g., sensor processor 130) is communicatively coupled to a host processor 110, host memory 111, and other components of device 100 via a communication interface 105 or other well-known means. SPU 120 may also include one or more communication interfaces (not shown) that are similar to communication interface 105 and are used for communication between one or more components within SPU 120.

[0058] Sensor processor 130 may be one or more microprocessors, CPUs, DSPs, general-purpose microprocessors, ASICs, ASIPs, FPGAs, or other processors running software programs that may be stored in memory, such as internal memory 140 (or elsewhere), and associated with the functionality of SPU 120. In some embodiments, one or more functions described as being performed by sensor processor 130 may be shared with or executed wholly or partially by another processor of device 100 (e.g., host processor 110).

[0059] Internal memory 140 can be any suitable type of memory, including but not limited to electronic memory (e.g., read-only memory (ROM), random access memory (RAM), or other electronic memory). Internal memory 140 can store algorithms, routines, or other instructions for instructing sensor processor 130 to process data output by one or more of motion sensors 160 and / or one or more of surface type detection sensors 150. In some embodiments, internal memory 140 may store one or more modules that can be implemented on sensor processor 130 to perform specific functions. Some examples of modules may include, but are not limited to, statistical processing modules, motion processing modules, surface type detection modules, and / or decision modules.

[0060] The surface type detection sensor 150 may be an acoustic transducer that operates in the aforementioned manner and with any acoustic range. In some embodiments, the surface type detection sensor 150 is an ultrasonic transducer, such as a PMUT (Piezoelectric Micromachining Ultrasonic Transducer). The surface type detection sensor 150 may be a MEMS device and may be very small, for example, having a facing surface smaller than 4 mm by 4 mm by 1.5 mm. Depending on the application and available space, the ultrasonic sensor may be large or small. In some embodiments, the surface type detection sensor 150 may be a SoC (System-on-Chip) including a DSP. In some embodiments, the SoC package of the surface type detection sensor 150 includes a sensor processing unit 120 and includes a sensor processor 130 and internal memory 140. In some embodiments, the surface type detection sensor 150 may be part of a surface type detection sensor assembly (e.g., see...). Figures 4A-4C The surface type detection sensor assembly may include a tube for directing the transmitted acoustic wave signal toward the surface and directing the acoustic wave return signal back to the surface type detection sensor 150.

[0061] Motion sensor 160 (when included) can be implemented as a MEMS-based motion sensor, including inertial sensors (such as gyroscope 161 or accelerometer 163) or electromagnetic sensors (such as Hall effect or Lorentz field magnetometer 165). In some embodiments, at least a portion of motion sensor 160 may also be based, for example, on sensor technologies other than MEMS technology (e.g., CMOS technology, etc.). As desired, one or more motion sensors 160 can be configured to provide raw data output of measurements along three orthogonal axes or any equivalent structure. One or more motion sensors 160 are communicatively coupled to sensor processor 130 via a communication interface, bus, or other well-known communication means.

[0062] Example System

[0063] Figure 2 A top front perspective view of an example of a system 200 according to various embodiments is shown. System 200 includes one embodiment of a device 100 that moves or operates across a surface, and also includes a base station 202 for the device 100. The device 100 includes a housing 201 to which one or more items may be coupled. The base station 202 provides a location where the device 100 can be positioned, parked, or docked when not moving or operating across a surface. In some embodiments, the base station 202 (which may also be referred to as docking 202) can provide information / instructions to and / or receive information from the device 100, such as via physical and / or wireless communication coupled to a transceiver 113. In some embodiments, the base station 202 can provide a charger for the device 100, such that when the device 100 is coupled to or properly oriented to the base station 202, the base station 202 charges the device 100 via physical or wireless (e.g., inductive) electrical coupling. In some embodiments, base station 202 may provide one or more sample surfaces 203 (e.g., 203-1 and 203-2), which can be used to test or calibrate surface type detection sensors, other electronic devices, and / or computing resources used by device 100 to detect the type of surface on which device 100 moves or operates. For example, in some embodiments, one or more samples of soft surface 203-1 and hard surface 203-2 may be provided as test targets at a known location of device 100, which device 100 can sense and detect as it positions itself within and / or leaves base station 202. In some embodiments, device 100 may also test the detection of a transition from one surface type (e.g., soft surface 203-1) to another surface type (e.g., hard surface 203-2) when two or more different types of sample surfaces 203 are positioned adjacent to each other. In some instances, any portion of the surface of base station 202 may be used similarly as sample surfaces 203.

[0064] Figures 3A-3C Various embodiments are shown. Figure 2 A side front view of an example system 200, which includes a device 100 that moves or operates across a surface and a base station 202 for the device 100. Figures 3A-3C As shown, device 100 and base station 202 are arranged on surface 300, and device 100 moves and operates on surface 300. By way of example and not limitation, device 100 is depicted as a robotic cleaning appliance. Device 100 includes a surface type detection sensor assembly 350 coupled to housing 201 or any suitable portion of device 100. Surface type detection sensor assembly 350 includes surface type detection sensor 150 (as will be combined with...) Figure 4A(Described). Although the surface type detection sensor assembly 350 is shown on the bottom of device 100 and pointing towards surface 300 (e.g., towards the floor), in other embodiments, the surface type detection sensor assembly 350 may be additionally or alternatively arranged on the top, side, or other parts for detecting vertical surfaces (e.g., walls) and / or overhanging surfaces. Device 100 may include one or more wheels 303 and / or 305, which may be driven and / or controlled by one or more drive wheel controllers 170. Device 100 may include one or more surface treatment items, such as surface treatment item 304 that may be driven and / or controlled by one or more surface treatment controllers 180. The depiction of surface treatment item 304 is merely an example and is not intended to limit the types of surface treatment items, tools, or tools represented in the illustration. Not limited thereto, surface treatment item 304 may include one or more of the following or combinations thereof: a suction tool or suction opening relative to the floor surface or other surface of floor 300; a rotating tool (e.g., a roller brush or broom); a sweeping tool (e.g., a broom); a wiping tool (e.g., a cloth / cloth-covered surface); a brushing tool (e.g., a fixed or movable brush head); a dusting tool; a mopping tool; and a spraying tool configured to spray a cleaning agent or other liquid.

[0065] exist Figure 3A In this process, device 100 travels along direction 301 on surface 300 and prepares to dock with base station 202. Figure 3B Meanwhile, device 100 continues to move along direction 301 on surface 300 and begins docking with base station 202. Figure 3C Meanwhile, device 100 is still moving in direction 301 and has almost completed its docking with base station 202.

[0066] See Figure 3A The acoustic signal transmitted by the surface type detection sensor 150 can be sent from an opening in the surface type detection sensor assembly 350 toward the surface 300, and the corresponding returned signal received from the surface 300 can be used by the device 100 to detect the surface type of the surface 300. In embodiments where the surface type detection sensor assembly 350 is located on another part of the device 100 (e.g., the side or top), the surface can be a horizontal surface (such as a wall) or an overhang (such as the underside of a coffee table or chair).

[0067] See Figure 3BAn acoustic signal transmitted by the surface type detection sensor 150 can be sent from an opening in the surface type detection sensor assembly 350 toward the sample surface 203-1, and the corresponding returned signal received from the sample surface 203-1 can be used by the device 100 to detect the surface type of the sample surface 203-1. Because the location and surface type of the sample surface 203-1 are known to the device 100, this detection operation can be used to calibrate the detection of these types of surfaces (e.g., soft surfaces), to store the received returned signal as a paradigm to which other signals can be compared, to store the reflectivity measure associated with the sample surface 203-1 as a paradigm for comparison with other reflectivity measures, and to detect malfunctions or variations in the operation of the surface type detection sensor (such as changes over time, due to the accumulation of debris and / or dust particles, due to temperature changes, aging, damage, humidity, installation angle, etc.). In this way, one or more sample surfaces 203 (or other known surfaces) can be used for the calibration of the surface type detection sensor assembly 350.

[0068] See Figure 3C The acoustic signal transmitted by the surface type detection sensor 150 can be sent from an opening in the surface type detection sensor assembly 350 toward the sample surface 203-2, and the corresponding returned signal received from the sample surface 203-2 can be used by the device 100 to detect the surface type of the sample surface 203-2. Since the location and surface type of the sample surface 203-2 are known to the device 100, this detection operation can be used to calibrate the detection of these types of surfaces (e.g., hard surfaces), store the received returned signal as a paradigm to which other signals can be compared, store the reflectivity metric associated with the sample surface 203-2 as a paradigm for comparison with other reflectivity metrics, and detect malfunctions or changes in the operation of the surface type detection sensor (such as changes over time, due to debris accumulation, due to temperature variations, etc.). Similarly, when multiple sample surfaces are adjacent or closely spaced, the transition from one sample surface 203-1 to the next sample surface 203-1 can be used to test the ability to detect such surface type changes.

[0069] Regarding calibration. In some embodiments, the amount of acoustic signal transmitted by the surface type detection sensor 150 can vary between sensors, can be altered due to environmental conditions (humidity, temperature, presence of debris in the sensor, etc.), and / or can also vary for individual sensors during the lifespan of the surface type detection sensor 150. Therefore, a calibration procedure can be used to improve the accuracy of surface type determination. By knowing the amount of transmitted signal and / or the percentage of transmitted signal received as a return signal from a particular surface type, the returned signal reflected from any surface type can be characterized more accurately. Calibration can be performed by measuring the reflection of the returned signal from a known surface. For example, sensor 150 can be calibrated when device 100 is in docking / charging base station 202, where the surface type beneath device 100 is historically known due to repeated measurements during docking and undocking with base station 202. Base station 202 may have one or more dedicated surface sample surface segments 203 (203-1, 203-2) that can be additionally or alternatively used for calibration. Similarly, measuring the reflection of signals returned from a known surface can be used in machine learning, such as training classifiers to detect different surface types based on certain inputs.

[0070] The device 100 may engage, disengage, deploy, redeploy, adjust the height, speed, or otherwise adjust the surface treatment item 304 based wholly or partially on surface type detection performed using the surface type detection sensor 150. For example, in response to such detection of the type of flooring (e.g., hard flooring or soft flooring) or any other characteristic of the flooring, the device 100 may take one or more actions to adjust aspects of its operation. By way of example and not limitation, in various embodiments, device 100 or a portion thereof may: adjust the moving speed of device 100; adjust the drive motor of device 100; modulate the allowable slip or rotation of the drive wheel of device 100; adjust or otherwise modulate the speed of the suction motor of device 100, or otherwise modulate the suction of the surface treatment item; adjust the height of the suction surface treatment item and / or suction opening relative to the surface (e.g., surface 300); adjust the height of the rotating surface treatment item (e.g., roller brush) or other surface treatment item of device 100 relative to the surface; adjust the rotational speed of the rotating surface treatment item of device 100; activate device 100. Rotary surface treatment project of device 100; deactivate rotary surface treatment project of device 100; use liquid / spray cleaner tool of device 100; activate brush tool of device 100; deactivate brush tool of device 100; stop use of liquid / spray cleaner tool of device 100; adopt mop / wiping tool of device 100; stop deployment of mop / wiping tool of device 100; adopt polishing tool of device 100; stop deployment of polishing tool of device 100; activate alarm or signal (audible, visible or some combination) of device 100 or associated with device 100; and / or deactivate alarm or signal of device 100 or associated with device 100. These and other actions may be performed by device 100 via instructions provided by a self-processor to: surface treatment controller 180, drive wheel controller 170 and / or one or more other components of device 100.

[0071] In some embodiments, surface detection and / or surface type detection can facilitate or assist in detecting whether the device is operating horizontally, horizontally, stuck, or where one of its wheel / surface contact points is not in contact with the surface. For example, if device 100 partially moves over a drop, such as over the edge of a staircase or roof, the surface type detection sensor 150 may fail to detect the surface or surface type because the surface is out of range, or may detect that the surface has substantially descended or become substantially closer since the previous measurement if it has become sticky or horizontal. Appropriate action may be initiated by device 100 to remedy situations where device 100 is detected as horizontally, stuck, or operating where one of its wheel / surface contact points is not in contact with the surface.

[0072] although Figures 3A-3C Only a single surface type detection sensor assembly 350 is shown, but some embodiments may include multiple transducers that can operate at different frequencies. The properties of the surface type can affect different frequencies in different ways, and therefore using multiple frequencies can facilitate more accurate detection / determination of the surface type.

[0073] Example surface type detection sensor component

[0074] Figure 4A A side front view of an example of a surface type detection sensor assembly 350, which can be used on device 100 to transmit and receive acoustic signals according to various embodiments, is shown. The surface type detection sensor assembly 350 includes at least a surface type detection sensor 150. In some embodiments of the surface type detection sensor assembly 350, a housing 450 (shown in dashed lines) may be included and coupled to the surface type detection sensor 150, and the housing 450 encloses and / or defines an acoustic interface such as a tube, cavity, corner, or some combination thereof. In other embodiments, the surface type detection sensor assembly 350 includes the surface type detection sensor 150 but does not include the housing 450. In some embodiments, the surface type detection sensor assembly 350 is coupled to device housing 201 (or other parts of device 100) such that it transmits acoustic signals in the direction in which it is expected to encounter a surface or operate on a surface (e.g., downward toward a floor surface, laterally toward a wall or where an object may obstruct the path of device 100, etc.). For example, in a floor vacuum cleaner or robotic floor cleaning embodiment of device 100, the surface type detection sensor assembly 350 may be arranged on the bottom of device 100 or configured to sense from the bottom of device 100 outwards, for example... Figure 3AThe manner shown ensures that when device 100 is in operation, the transmitted acoustic signals are directed toward the floor surface 300. Similarly, in a drone embodiment of device 100 (not described), acoustic signals can be transmitted in the direction of a surface where the drone can land. Figure 1A and 1B As described above, the surface type detection sensor 150 may also be coupled to a host processor 110 or a sensor processor. In some embodiments, either or both of the host processor 110 or the sensor processor operate to process and make determinations based on the received returned signals.

[0075] Figure 4B Various embodiments are shown. Figure 4A Top view of the surface type detection sensor assembly 350. Section line AA marked with... Figure 4C The cross-sectional view shown in the figure is associated with the position and orientation.

[0076] Figure 4C A side cross-sectional view of the configuration of a surface type detection sensor assembly 350A according to some embodiments is shown. It can be seen that the surface type detection sensor 150 includes an acoustic transducer 401 (e.g., an ultrasonic transducer), which is covered by a cap 402 and separated from the housing 450A by a membrane 403. The housing 450A encloses and / or defines an acoustic interface tube 405A (which may also be referred to as a “corner,” “acoustic tube,” or simply a “tube”). The acoustic tube 405A facilitates the travel of acoustic signals transmitted from the surface type detection sensor 150 into the opening 407A of the acoustic interface tube 405A, through the length of the acoustic interface tube 405A, and out of the opening 406A toward the surface 300, wherein the direction of travel is indicated by a directional arrow 410. Similarly, acoustic tube 405A facilitates the travel of acoustic signals from surface 300 back into opening 406A of acoustic interface tube 405A, through the length of acoustic interface tube 405A, and out of opening 407A to be received by surface type detection sensor 150, wherein the direction of travel is indicated by directional arrow 411. Acoustic interface tube 405A has a diameter 404A that is constant over the entire length of tube 405A. It should be understood that the dimensions are presented by way of example only, and in other embodiments, for example, the diameter 404A may be larger or smaller, or vary along the length (e.g., becoming wider near opening 406A). In some embodiments, the dimensions of acoustic interface tube 405A may be selected to limit the acoustic field of surface type detection sensor 150.

[0077] In one example embodiment, the received returned signals (such as those indicated by arrow 411) are digitized. Digitization may be performed by the surface-mount type detection sensor 150 on the DSP board, by the sensor processor 130, by the host processor 110, or by another processor.

[0078] Examples of multiple floor surface types

[0079] Figure 5 A top plan view of a span 500 of flooring materials having multiple different floor surface types is shown. For example, span 500 includes hardwood flooring 510, which is separated from carpeted flooring 530 by a metal edging 520. Hardwood flooring 510 is an example of a hard surface. Most of the hardwood flooring 510 is clean flooring, but area 511 represents wet areas (such as puddles), and area 512 represents dirty areas (such as areas with track mud). In some embodiments, areas 511 and 512 can be distinguished from clean and dry portions of the hardwood flooring 510 by analyzing the returned signals received by surface type detection sensor 150. Carpeted flooring 530 is an example of a soft surface. Most of the carpeted flooring 530 is clean carpeted flooring, but area 531 represents wet areas (such as water-soaked areas of carpet), and area 532 represents dirty areas (such as areas of carpet with track mud). In some embodiments, regions 531 and 532 can be distinguished from the clean and dry portions of the carpeted floor 530 by analyzing the returned signals received by the surface type detection sensor 150. As previously discussed, there can be multiple floor surface types, and Figure 5 Only a limited selection of them is shown. Arrow 501 indicates the direction of travel from left to right relative to span 500. For example, a robotic cleaning appliance (such as device 100) may travel in direction 501 and encounter one or more of the different floor types shown during its movement.

[0080] Blank ringing signal estimation and comparison

[0081] Figure 6 A block diagram of the signal path 600 of the acoustic transducer embodiment of the surface type detection sensor 150 according to various embodiments is shown. Starting from the upper left, a drive waveform 601 is applied to the acoustic transducer 150. In some embodiments, the drive waveform may be a square wave, such as square wave 610; however, other oscillating waveforms may be used. The drive waveform 601 causes the membrane (e.g., membrane 403) of the acoustic transducer 150 to modulate / vibrate back and forth during a period Tx (as shown in transducer modulation 620) and generate an acoustic signal transmitted from the acoustic transducer 150. When the drive waveform 601 stops, active transmission stops and reception using the acoustic transducer 150 begins. A portion of the received signal (during time Rx of 620) occurs during the ringing time (T... RINGDOWNDuring the Rx period, the oscillating motion induced in the membrane 403 by the driving waveform decays to a degree that the received signal can be easily distinguished, as they induce themselves to modulate into the membrane 403. The amplitude of the received signal (in the Rx period) can be analyzed and / or they can undergo demodulation 603 to obtain access to the demodulated output 630 (such as in-phase (I) data 631 and quadrature (Q) data 632 in the complex domain). After filtering 604 with a finite impulse response (FIR) filter, a cascaded integrated combination (CIC) filter, and / or other suitable filters, the returned signal, represented as amplitude signal 641, the filtered I signal 642, and / or the filtered Q signal 643, are provided as output 640. By way of example, in Figure 7A and Figure 7B The amplitude signal 641 is discussed in more detail and with a larger scale diagram.

[0082] Figure 7A A graph 700A, according to an embodiment, illustrates the amplitude of a sampled return signal from a surface type detection sensor 150. The sample descends along the line indicated by amplitude signal 641, and the time to receive the sample is equal to the distance to the object in meters, which will be associated with the sample reception time (if it is a corresponding return signal reflected from the object / surface). The peak of amplitude signal 641 marks the end of active transmission of the acoustic signal when the drive waveform 601 is no longer applied to the acoustic transducer 150. Ringing decay begins immediately after this peak. Lines 745 and 746 distinguish the region of sampling points for the data in the early portion 703A of the ringing signal component of signal 641. For example, this region may cover sampling points 9 to 15 (i.e., samples 9 to 15). More or fewer samples can be taken in the early portion 703A, provided there are enough sampling points to model / fit the curve of the signal for that time period. Regarding the sample between lines 745 and 746, due to the distance and required round-trip flight time, it is physically impractical or impossible for the returned signal to be reflected from the floor or other surface and received back at the acoustic transducer 150. Therefore, the signal sampled between lines 745 and 746 includes only the ringing signal and not the received returned signal that has already been reflected from the floor or other surface. If the floor, surface, or other object is close and reflects the signal, the sample acquired during the later portion 704A of the ringing component of signal 641 will typically include some of the received returned signal. These returned signals will cause constructive or destructive interference in signal 641. Line 747 illustrates the time period of interest for the ringing signal occurring between lines 746 and 747.

[0083] although Figure 7AOnly amplitude signal 641 is shown, but it should be understood that it can be similarly shown as I data signal 642 and / or Q data signal 643, and either or both of the I data and Q data can be sampled between lines 745 and 746 (early part) or after line 746 (late part).

[0084] Figure 7B A graph 700B is shown illustrating the sampling points of the returned signal, fitted to the early portion of the ringing signal from the surface type detection sensor according to an embodiment, and used to estimate the blank ringing signal of the surface type detection sensor. In some embodiments, a processor (e.g., sensor processor 130, host processor 110, etc.) determines the characteristics of signal lines 641 connecting, for example, samples 9-15. The processor then applies attenuation to model or estimate the blank ringing signal. Figure 7B In the diagram, portion 703B of curve 741 is fitted to the sampled data of curve 641 during the earlier portion 703A, while portion 704B is a projection or estimated extension of the decay slope of portion 741 extending from line 746 to line 747. One example of a feature is the slope of the line; another is the amplitude. An example of decay is exponential decay. The curve fit can be any suitable high-order fit / model, which is then extrapolated to later portions of the signal. Figure 7B As shown, the estimated blank ringing signal simulates the later portion 704B of the ringing signal 741, where no returned signal arrives after line 746, and this later portion 704B begins after points 9-17 (e.g., the early portion of the ringing signal 703A). In other words, the later portion 704B is estimated based on the characteristics of the early portion by applying a selected attenuation rate to curve 703B formed / modeled from measurement points in the early portion. The blank ringing signal 741 is an estimate or model of what the ringing component of the later portion (after line 746) of the amplitude signal 641 looks like without any constructive or destructive interference caused by the received returned signal. Additionally, it should be noted that the early portion 703B is also modeled / fitted based on data already selected from a time period that should be free from most or all constructive or destructive interference caused by the received returned signal.

[0085] In some embodiments, a processor (e.g., sensor processor 130, host processor 110, etc.) compares the estimated blank ringing signal 741 or a portion thereof with a similar portion of the actually measured ringing signal component of signal 641. In one example, it can be... Figure 7A Parts of 703A and 704A are respectively with Figure 7B A comparison is made between parts 703B and 704B. In another example, Figure 7APart of 704A can be used with Figure 7B The comparison is performed on part 704B. This comparison can use an estimated blank ringing signal 741 to compensate for the actual measured ringing signal component of signal 641. One such mechanism for comparison / compensation is background subtraction, where the estimated blank ringing signal is subtracted from the actual ringing signal, resulting in a difference. The resulting difference (i.e., the compensated signal) can be used to derive one or more metrics representing the magnitude of the returned signal received in the comparison portion of the actually returned signal 741. For example, any data point or set of data points from the compensated ringing signal can be used as a metric. For example, the metric can be based on one or more maximum values ​​of the compensated ringing signal. In other examples, the sum, energy, integral, or surface area of ​​the compensated signal can be used to determine the metric. In this way, for example, a metric based on the compensated ringing signal can be used to determine the type of surface from which the returned signal is reflected and received as part of a later portion of the ringing signal. For example, this metric can be compared with a threshold to determine whether the returned signal comes from a certain type of surface (e.g., hard floor (if above a certain magnitude) or soft floor (if not above a certain magnitude)). If the type of the surface is known a priori, then a metric can be provided to the classifier as labeled data for machine learning.

[0086] Combination Figures 7A-7B The technique described and illustrated with amplitude signal 641 can be similarly used with a filtered in-phase signal 642 or a filtered quadrature signal 643 to generate an estimated blank ringing signal portion of the signal, and then compared with the actually received ringing signal portion of the same signal. In this way, comparisons can be made, and metrics can be generated from one or more of the amplitude signal, the filtered in-phase signal, and the filtered quadrature signal from the acoustic transducer 150. The metrics can be used to determine the surface type from which the returned signal is reflected and received as part of the later portion of the ringing signal.

[0087] Figure 7CA graph 700C, according to an embodiment, illustrates the phase of two ringing signals from a surface type detection sensor 150. The y-axis of graph 700C represents the phase of plotted ringing signals 748 and 749. Plotted ringing signal 748 represents an example phase diagram of a blank ringing signal, starting at line 745, where constructive or destructive interference from the returned signal has no effect on the phase (or a very small effect). Plotted ringing signal 749 represents an example phase diagram of a ringing signal starting at line 745, where a returned signal is received from a hard reflective surface and the returned signal affects the phase. It can be seen that, due to the interference of the reflected signal and the ringing signal, plotted ringing signal 749 exhibits a much larger phase change than plotted ringing signal 748. The abrupt change in plotted ringing signal 749 is due to the plotting of the phase within the range of -180 degrees to 180 degrees.

[0088] Figure 8 A block diagram 800 illustrates a method for determining the curve lengths of a ringing signal and a blank ringing signal in a complex domain for a surface type detection sensor, according to an embodiment. In some embodiments, a processor (e.g., sensor processor 130, host processor 110, etc.) performs... Figure 8 The actions shown.

[0089] exist Figure 8 Starting from the upper left, full ringing I and Q data (IQ data 810) from the acoustic transducer 150 are provided as input to the ringing curve length estimator 815. See also Figure 6 and Figure 7A The IQ data 810 (in other words, the complete data starting from the ringing) will include data from... Figure 7A The sampling of I and Q data points (early ringing portions) within the window defined by lines 745 and 746, as well as data in subsequent ringing points occurring after line 746, is performed. The I and Q portions of the IQ data 810 are processed and used to generate a curve of the I data over time in the complex domain relative to the Q data. The length of this curve is calculated, and the output is the actual ringing curve length 816. The actual ringing curve length 816 and its associated ringing curve are provided to comparison box 830. This technique generates one curve per frame, where a frame is an array of sampled IQ data, meaning one transmit and receive cycle to produce the curve shown in the figure. Intra-frame sampling time information is independent of the curve length metric, but the metric can change between frames. That is, the inter-frame sampling rate determines the detection rate.

[0090] See also Figure 8Starting from the lower left, early ringing I and Q data 820 are provided as input to the blank ringing curve length estimator 828. The early ringing IQ data is IQ data obtained from samples of points in the early portion of the ringing before the acoustic transducer 150 receives a return signal (e.g., occurring when...). Figure 7A The line between 745 and 746 Figure 6 (I data 642 and Q data 643). This early data can be processed in a similar way, as regarding... Figure 7B The estimation of the blank ringing signal 704B is described. For example, one or more features of a line connecting the set of points in the in-phase data of I and Q data 820 can be subsequently attenuated to estimate the blank ringing signal of the in-phase data. Similarly, one or more features of a line connecting the set of points in the quadrature data of I and Q data 820 can be subsequently attenuated to estimate the blank ringing signal of the in-phase data. The blank ringing curve length estimator 828 extends the I and Q blank ringing curves starting from the earlier ringing IQ data 820 and forms estimated blank ringing signals (blank in-phase ringing signal and blank quadrature ringing signal) in the complex domain for the corresponding range covered by IQ data 810. On a polar plot, the ideal version of these blank rings will form a straight line or an approximate straight line. The estimated blank ringing signal curve length 829 of the I and Q blank ringing signals is calculated and used to generate a curve of I data over time in the complex domain relative to the Q data. The length of this curve is calculated, and the output is the estimated blank ringing curve length 829. The estimated blank ringing curve length 829 and its associated ringing curve are provided to the comparison box 830.

[0091] On each sampling frame of the IQ, the curve length can be calculated. As already described, the empty curve length is estimated from an early point in the IQ frame. The signal curve length is calculated from the entire dataset of measurement data within the frame. The final metric for each frame is the absolute or relative difference between these curve lengths (either method is valid). A metric is calculated for each frame in this manner. In some embodiments, where this occurs between 10 and 100 frames per second, this also results in 10-100 curve length comparisons per second.

[0092] In box 830, the actual ringing curve length 816 can be compared with the estimated blank ringing curve length 829, for example, by subtraction or division, to form a metric 840. This metric can be used to determine the type of surface from which the reflected signal is received as part of the later portion of the ringing signal. Similarly, the actual I and Q ringing signals can be compared with the estimated blank ringing signal, for example by background subtraction, to determine additional metrics at points along the time period associated with the ringing signal. These additional metrics can be used to determine the type of surface from which the reflected signal is received as part of the later portion of the ringing signal. One or more metrics can be compared with one or more thresholds or one or more reference metrics to determine the characteristics of the surface. For example, a higher curve length may correspond to a harder surface. In other words, the difference in curve length between the measured ringing signal and the estimated blank ringing signal represents the reflectivity of the surface, where a more reflective surface has a higher reflectivity than a less reflective surface.

[0093] Figure 9 The diagram illustrates the curve lengths in a complex domain for a blank ringing signal 974 and a measured ringing signal 964 from a surface type detection sensor 150 according to an embodiment. Angles represent phase, while signal intensity is represented outwards from the origin (2000-10000). Both curves / signals 974 and 964 begin with the same maximum amplitude at point 901, and the amplitude decays over time as the signal moves toward the origin of the polar plot. For example, curve 964 may represent a visual depiction of IQ data, while the sum 815 represents the length of the phase and amplitude data of the IQ ringing signal plotted relative to each other on a parametric curve (e.g., pole plot 900) to produce the curve shown by curve 964. In this example, curve 964 represents an actual measured ringing signal from a hard or reflective surface. The length of curve 964 may then be mathematically measured by a processor (e.g., processor 110 or sensor processor 130). Similarly, in this example embodiment, curve 974 is a visual depiction of the sum 820, where the phase and amplitude data of the IQ ringing signal are plotted relative to each other on a parametric curve (e.g., polar coordinate curve 900) to produce the curve shown in curve 974. Curve 974 represents the estimated blank ringing signal. The length of curve 974 can then be mathematically measured by a processor (such as host processor 110 or sensor processor 130).

[0094] As can be seen, curve 964 has more phase variation and is longer than curve 974, which is closer to a straight line. The lengths of curves 964 and 974 can be mathematically compared by the processor to generate a metric that can determine the floor type. Any metric indicating how much curve 964 deviates from a straight line can be used as a metric to determine the floor type. For example, the difference between the curves can be compared to a threshold to determine how hard or soft the surface is, and thus the type of surface (e.g., carpet, epoxy, wood, cementite, concrete, tile, etc.). This can be repeated multiple times per second, as can be discussed.

[0095] Figure 10 A diagram illustrates a classifier 1000 trained 1050 to detect surface types based on input, according to various embodiments. Machine learning training utilizes labeled data (in this case, floor type annotation 1040). The annotation for floor type 1040 can be binary, such as hard (e.g., for wood, tile, concrete, or other hard surfaces) or soft (e.g., for carpet, blanket, or other soft surfaces). In other instances, the annotation for floor type 1040 can have greater nuances, such as damp (and in some cases, more wet on a moisture scale), dirty (and in some cases, more dirty and / or dirty with what), dry, hard (and in some cases, more hard on a hardness scale), soft (and in some cases, more soft on a softness scale), etc. Other inputs provided in conjunction with the floor type annotation 1040 may include: sensor parameters 1010; features 1020; and mechanical parameters 1030. Feature 1020 may include aspects such as the length of the blank ringing curve, a comparison between the actual blank ringing curve length and the estimated blank ringing curve length, and the area between the blank ringing curve and a straight line; sensor parameters 1010 may include parameters such as the frequency and / or amplitude of the drive waveform 601 used to drive the acoustic transducer 150. Mechanical parameters may include aspects such as the height of the acoustic transducer 150 above the sensed floor or surface; and the angle of the acoustic transducer 150 relative to the sensed floor or surface.

[0096] Figure 11 Illustrations show a trained classifier 1000 used to detect surface types based on inputs according to various embodiments. For example, after training on inputs associated with known floor types (e.g., associated with floor type annotation 1040), inputs 1010, 1020, and / or 1030 can be provided to classifier 1000 without floor type annotation 1040, and classifier 1000 can determine floor type 1160 among a plurality of floor surface types (e.g., hard, soft, etc.) classified via machine learning using those inputs.

[0097] Example operation method

[0098] Figure 12A-12B The process of the method shown in flowchart 1200 will be referred to Figure 1A-11 The flowchart 1200 describes elements and / or components of one or more of the diagrams. It should be appreciated that in some embodiments, processes may be performed in a different order than those described in the flowchart, some of the described processes may not be performed, and / or one or more additional processes to those described processes may be performed. Flowchart 1200 includes programs in various embodiments that are executed by one or more processors (e.g., processor 130, host processor 110, DSP, etc.) under the control of computer-readable and computer-executable instructions stored on a non-transitory computer-readable storage medium (e.g., host memory 111, internal memory 140, etc.). It should also be understood that one or more processes described in flowchart 1200 may be implemented in hardware or in a combination of hardware and firmware and / or software.

[0099] For illustrative purposes only, Figure 1- Figure 3C Device 100 is a robotic cleaning appliance that includes a surface type detection sensor 150 in the form of an acoustic transducer capable of operating within the ultrasonic frequency range. It can represent any type of robotic cleaning equipment, such as equipment that cleans floors by sweeping, wiping, mopping, polishing, or vacuuming. However, it should be understood that this robotic cleaning appliance may take other forms and have features and components different from those depicted and described.

[0100] In some embodiments, the surface type detection sensor 150 is a floor type detection sensor, which is typically used to detect the presence or absence of the floor 300 beneath the robotic cleaning appliance 100 and further to collect returned signals from which the type of the floor (e.g., hard floor, soft floor, etc.) can be detected. In some embodiments, the surface type detection sensor 150 is a wall type detection sensor, which is typically used to detect the presence or absence of a wall laterally adjacent to the robotic cleaning appliance 100 and further to collect returned signals from which the type of floor (e.g., hard wall, soft wall, etc.) can be detected. In some embodiments, the device 100 takes certain actions based on the type of the floor surface detected by the surface type detection sensor 150. For example, when the device 100 is a robotic vehicle, the propulsion force can be adjusted based on the type of surface on which the robotic vehicle operates.

[0101] Figure 12A-12B A flowchart 1200 illustrates an example method for surface type detection according to various embodiments.

[0102] See Figure 12A At process 1210 of flowchart 1200, in various embodiments, a processor coupled to an acoustic transducer receives a returned signal from the acoustic transducer. The acoustic transducer is configured to send acoustic signals (with signals generated by...) toward a surface (such as a surface) Figure 4C Arrow 410 indicates the direction of travel), and receives the returned signal reflected from the surface (with the signal reflected by the arrow). Figure 4C (Arrow 411 indicates the direction of travel). The returned signal corresponds to the transmitted acoustic signal, and the surface is within the ringing distance associated with the acoustic transducer. Surface type detector 150 is an example of an acoustic transducer. That is, the surfaces are so close that some of the returned signal begins to be received while the acoustic transducer is still ringing. Host processor 110, sensor processor 130, and / or other suitable and available processors can direct the transmission of the acoustic signal and can receive and perform processing on the returned signal initially received at the acoustic transducer.

[0103] See also Figure 12A At process 1220 of flowchart 1200, in various embodiments, in response to the cessation of active transmission of the acoustic signal, multiple points of the ringing signal generated by the acoustic transducer are sampled. Sampling occurs during the early portion of the ringing signal before the returned signal from the surface is reflected back and received by the acoustic transducer. Sampling also occurs during the later portion of the ringing signal, including the returned signal. See also Figure 6 and Figure 7B In one embodiment, early partial sampling may occur during transducer ringing, such as between lines 745 and 746, before the returned signal begins to be received at acoustic transducer 150. In the same embodiment, later partial sampling may occur after line 746 and while ringing is still occurring. Each sample of the returned signal has an amplitude that can be measured and then digitally represented as a signal that can be processed to determine an amplitude signal, an in-phase signal (if demodulated), and / or a quadrature-phase signal (if demodulated). Host processor 110, sensor processor 130, and / or other suitable and available processors may perform sampling.

[0104] See also Figure 12AAt process 1230 in flowchart 1200, in various embodiments, an early portion of multiple sampling points is used to estimate the blank ringing signal of the acoustic transducer. The estimated blank ringing signal represents the performance of the acoustic transducer in the absence of any returned signal. It should be understood that the estimated blank ringing signal can be estimated once, such as at the factory; more frequently, such as at startup, when entering / exiting a docking / charging station, or according to a recursive schedule; or in real-time / near real-time as needed. The host processor 110, sensor processor 130, and / or other suitable and available processors can be relative to... Figure 7A , Figure 7B and / or Figure 8 The estimation is performed in the manner discussed.

[0105] See also Figure 12A At process 1240 of flowchart 1200, in various embodiments, the estimated blank ringing signal is compared with a later portion of the ringing signal. This comparison may involve, for example, using background subtraction techniques (see, for example, [link to relevant documentation]). Figure 7B To subtract two, this background subtraction technique is based on comparing one or more signal curves between the actual ringing and the corresponding estimated blank ringing signal; comparing the length of the signal curve used for the ringing signal with the estimated blank ringing curve length in the complex domain (see, for example) Figure 8 and Figure 9 (or one or more comparisons). The host processor 110, sensor processor 130, and / or other suitable and available processors can perform one or more comparisons.

[0106] See also Figure 12A At process 1250 in flowchart 1200, in various embodiments, a metric is generated based on comparison. The metric can be a product of comparisons, background subtraction results, differences between curve lengths, etc. In some embodiments, the metric is based on a comparison of the characteristics of the estimated blank ringing signal and the same characteristics of the later portion of the ringing signal, wherein the characteristics are based on at least one of in-phase and quadrature components. For example, the metric can be based on a comparison of the signal curves in the complex domain and / or a comparison of the corresponding curve lengths (in the complex domain) of the estimated blank ringing signal and the later portion of the ringing signal. In some embodiments, the metric is based on a comparison of the corresponding amplitudes of the estimated blank ringing signal and the later portion of the ringing signal. The host processor 110, sensor processor 130, and / or other suitable and available processors can generate the metric.

[0107] See also Figure 12AIn process 1260 of flowchart 1200, in various embodiments, the metric is used to determine the type of the surface from multiple surface types. In some embodiments where background subtraction is used to generate the metric, the difference resulting from the subtraction is a metric and can be compared with a threshold to determine the surface type. For example, if the magnitude of the metric meets or exceeds the threshold, then it can be determined that a surface of one type (e.g., hard floor) has been detected. If the magnitude of the metric is below the threshold, then it can be determined that another type of surface (e.g., soft floor) has been detected. In other embodiments, where curve length, such as the difference in curve length, is used as a metric, this metric can be provided as input to a floor type classifier (e.g., classifier 1000), which has been trained to determine the type of the surface from multiple surface types. It should be understood that any metric can be provided to a floor type classifier (e.g., classifier 1000) that has been trained using that type of metric. Host processor 110, sensor processor 130, and / or other suitable and available processors can make decisions, can act as classifiers, and / or can provide metrics to individual classifiers.

[0108] See Figure 12B At process 1270 in flowchart 1200, in various embodiments, the estimated blank ringing signal can be stored for future use with subsequent transmissions of acoustic signals by the acoustic transducer. That is, the estimated blank ringing signal may not be estimated each time it is used. Instead, it can be estimated once, then stored and used more than once. In one embodiment, the estimated blank ringing signal may be estimated once, for example, in a factory or manufacturing environment, and then stored for future use. In another embodiment, the estimated blank ringing signal may be estimated at startup and then stored for future use until shutdown. In one embodiment, the estimated blank ringing signal may be estimated upon leaving the docking / charging station and stored for future use until shutdown or re-entry into the docking / charging station. In some embodiments, the estimated blank ringing signal may be estimated at certain intervals, such as every minute during operation, every 5 minutes during operation, monthly, or at some other interval or multiple intervals, and then stored for future use. The host processor 110, sensor processor 130 and / or other suitable and available processors may store the estimated blank ringing signal in memory (e.g., host memory 111, internal memory 140 or other storage device).

[0109] Brief description of each aspect

[0110] In one aspect, the robotic cleaning appliance includes a housing, a surface treatment item coupled to the housing, an acoustic transducer coupled to the housing, and a processor coupled to the acoustic transducer and the housing. The acoustic transducer is configured to transmit an acoustic signal toward a surface below the robotic cleaning appliance and receive a corresponding reflected signal from the surface, wherein the surface is within a ringing distance associated with the acoustic transducer. In some embodiments, the acoustic signal is an ultrasonic signal. In some aspects, the processor is configured to sample multiple points of the ringing signal generated by the acoustic transducer after active transmission of the acoustic signal has ceased. Sampling occurs during an early portion of the ringing signal before the corresponding reflected signal from the surface is reflected back and received by the acoustic transducer. Sampling also occurs during a later portion of the ringing signal, which includes the corresponding reflected signal that has already been reflected back from the surface and received by the acoustic transducer. Using the early portions of the multiple sampling points, the processor estimates a blank ringing signal for the acoustic transducer. The estimated blank ringing signal represents the performance of the acoustic transducer when no return signal is received. The processor compares the estimated blank ringing signal with a later portion of the ringing signal and generates a metric based on this comparison. The metric is used to determine the surface type from multiple surface types. For example, in some aspects, the metric may be provided as input to a floor type classifier trained to determine the surface type from multiple surface types.

[0111] In some aspects of robotic cleaning appliances, the processor can store estimated blank ringing signals for future use with subsequent transmissions of acoustic signals from the acoustic sensor. The storage device can be within the processor's memory or in other memory accessible to the processor.

[0112] Regarding the generation of the metric, in some aspects of the robotic cleaning appliance, the processor can generate the metric based on a comparison of the characteristics of the estimated blank ringing signal with the same characteristics of the later portion of the ringing signal, wherein the characteristics are based on at least one of in-phase and quadrature components.

[0113] Regarding the generation of the metric, in some aspects of this robotic cleaning appliance, the processor can generate the metric based on a comparison of the estimated blank ringing signal with the corresponding curve length of the later portion of the ringing signal.

[0114] Regarding the generation of metrics, in some aspects of robotic cleaning appliances, the processor can generate metrics based on a comparison of the estimated blank ringing signal with the corresponding amplitude of the later part of the ringing signal.

[0115] In one aspect, the sensor processing unit includes an acoustic transducer coupled to a sensor processor. The acoustic transducer is configured to transmit an acoustic signal to a surface and receive a corresponding reflected signal from the surface, wherein the surface is within a ringing distance associated with the acoustic transducer. The acoustic signal can be in the ultrasonic range. The sensor processor is configured to sample multiple points of the ringing signal generated by the acoustic transducer after the active transmission of the acoustic signal has ceased. Sampling occurs during an early portion of the ringing signal before the corresponding reflected signal from the surface is received by the acoustic transducer. Sampling is also performed in a later portion of the ringing signal, which includes the corresponding reflected signal. In some aspects, the sensor processor uses the early portion of the multiple sampling points to estimate a blank ringing signal for the acoustic sensor, wherein the estimated blank ringing signal represents the performance of the acoustic sensor in the absence of any received reflected signal. The sensor processor compares the estimated blank ringing signal with the later portion of the ringing signal and generates a metric based on the comparison. The sensor processor utilizes the metric to determine the surface type from multiple surface types. For example, in some aspects, the metric can be provided as input to a floor type classifier trained to determine the type of a surface from multiple surface types. The floor type classifier can be embodied in the sensor processing unit or can be external to the sensor processing unit.

[0116] In some aspects of the sensor processing unit, the sensor processor can store estimated blank ringing signals for future use with subsequent transmissions of acoustic signals from the acoustic sensor. The storage device can be located within the sensor processing unit's memory or in other memory accessible to the sensor processor.

[0117] Regarding the generation of metrics, in some aspects of the sensor processing unit, the sensor processor can generate metrics based on a comparison of the characteristics of the estimated blank ringing signal with the same characteristics of the later part of the ringing signal, wherein the characteristics are based on at least one of in-phase components and quadrature components.

[0118] Regarding the generation of metrics, in some aspects of the sensor processing unit, the sensor processor can generate metrics based on a comparison of the estimated blank ringing signal with the corresponding curve length of the later portion of the ringing signal.

[0119] Regarding the generation of metrics, in some aspects of the sensor processing unit, the sensor processor can generate metrics based on a comparison of the estimated blank ringing signal with the corresponding amplitude of the later portion of the ringing signal.

[0120] In one aspect of the surface type detection method, the method is implemented by a processor coupled to an acoustic transducer. The acoustic transducer is configured to send an acoustic signal to the surface and receive a returned signal reflected from the surface. The returned signal corresponds to the sent acoustic signal (which may be an ultrasonic signal), and the surface is within a ringing distance associated with the acoustic transducer. The method includes the processor receiving the returned signal from the acoustic transducer, which is configured to send an acoustic signal to the surface and receive a returned signal reflected from the surface, wherein the returned signal corresponds to the sent acoustic signal, and wherein the surface is within a ringing distance associated with the acoustic transducer. In some aspects, the method further includes sampling a plurality of points of the ringing signal generated by the acoustic transducer by the processor. Sampling occurs after and can be responsive to the cessation of active transmission of the acoustic signal. Sampling occurs during an early portion of the ringing signal before the returned signal from the surface is reflected back and received by the acoustic transducer. Sampling also occurs during a later portion of the ringing signal, including the returned signal. In some aspects, the method also includes estimating a blank ringing signal of the acoustic transducer by a processor. The estimation is performed using an early portion of multiple sampling points, and the estimated blank ringing signal represents the performance of the acoustic transducer in the absence of any returned signal. In some aspects, the method includes having the processor compare the estimated blank ringing signal with a later portion of the ringing signal. In some aspects, the method includes having the processor generate a metric based on the comparison. In some aspects, the method includes having the processor utilize the metric to determine the type of a surface from multiple surface types. For example, in some aspects, the metric may be provided as input to a floor type classifier trained to determine the type of a surface from multiple surface types.

[0121] In some aspects of this method, the method further includes storing the estimated blank ringing signal for future use with subsequent transmissions of acoustic signals from the acoustic sensor. The storage device may be located in a memory locally coupled to the processor, or in a memory or storage device located remotely from the processor.

[0122] Regarding the generation of metrics, in some aspects of the method, the method may involve generating metrics based on a comparison of features of the estimated blank ringing signal with the same features in the later part of the ringing signal, wherein the features are based on at least one of in-phase and quadrature components.

[0123] Regarding the generation of metrics, in some aspects of this method, the method may involve generating metrics based on a comparison of the length of the corresponding curve of the estimated blank ringing signal with the later portion of the ringing signal.

[0124] Regarding the generation of metrics, in some aspects of this method, the method may involve generating metrics based on a comparison of the estimated blank ringing signal and the corresponding amplitude of the later portion of the ringing signal.

[0125] in conclusion

[0126] The examples set forth herein are presented to best explain and describe particular applications, and thereby enable those skilled in the art to make and use embodiments of the described examples. However, those skilled in the art will recognize that the foregoing descriptions and examples have been given for illustrative and exemplary purposes only. The set forth descriptions are not intended to be exhaustive or to limit the embodiments to the precise forms disclosed. Rather, the specific features and actions described above are disclosed as exemplary forms for implementing the claims.

[0127] Throughout this document, references to "an embodiment," "a particular embodiment," "an embodiment," "various embodiments," "some embodiments," or similar terms mean that a specific feature, structure, or characteristic described in connection with that embodiment is included in at least one embodiment. Therefore, the appearance of such phrases in various places throughout this specification does not necessarily refer to the same embodiment. Furthermore, a particular feature, structure, or characteristic of any embodiment may be combined in any suitable manner with one or more other features, structures, or characteristics of one or more other embodiments, without limitation.

Claims

1. A robotic cleaning device, comprising: The housing, and the surface treatment items are coupled to the housing; as well as An acoustic transducer is coupled to the housing and configured to send acoustic signals toward a surface below the robotic cleaning device and receive corresponding returned signals reflected from the surface, wherein the surface is within a ringing distance associated with the acoustic transducer. A processor, coupled to the housing, and configured to: After the active transmission of the acoustic signal stops, multiple points of the ringing signal generated by the acoustic transducer are sampled, wherein the sampling occurs during an early portion of the ringing signal before the corresponding returned signal from the surface has been reflected back and received by the acoustic transducer, and the sampling also occurs during a later portion of the ringing signal including the corresponding returned signal. The early portion of the acoustic transducer is used with multiple sampling points to estimate the blank ringing signal, wherein the estimated blank ringing signal represents the performance of the acoustic transducer when no return signal is received. The estimated blank ringing signal is compared with the later portion of the ringing signal; Metrics are generated based on comparison. as well as The metric is used to determine the type of the surface from a plurality of surface types.

2. The robotic cleaning device of claim 1, wherein the processor is further configured to: The estimated blank ringing signal is stored for future use with subsequent transmissions of acoustic signals from the acoustic transducer.

3. The robotic cleaning appliance of claim 1, wherein the processor is configured to generate metrics based on comparisons, the processor is configured to: The metric is generated by comparing the features of the estimated blank ringing signal with the same features in the later portion of the ringing signal, wherein the features are based on at least one of in-phase and quadrature components.

4. The robotic cleaning appliance of claim 3, wherein the processor is configured to generate metrics based on comparisons, the processor is configured to: The metric is generated based on a comparison of the estimated blank ringing signal with the corresponding curve length of the later portion of the ringing signal.

5. The robotic cleaning appliance of claim 1, wherein the processor is configured to generate metrics based on comparisons, the processor is configured to: The metric is generated by comparing the estimated blank ringing signal with the corresponding amplitude of the later portion of the ringing signal.

6. The robotic cleaning apparatus of claim 1, wherein the processor is configured to determine the type of the surface from a plurality of surface types using the metric, comprising the processor being configured to: The metric is provided as input to a floor type classifier, which is trained to determine the type of the surface from a plurality of surface types.

7. The robotic cleaning device as claimed in claim 1, wherein the acoustic signal is within the ultrasonic frequency range.

8. A sensor processing unit, comprising: An acoustic transducer configured to send an acoustic signal toward a surface and receive a corresponding returned signal reflected from the surface, wherein the surface is within a ringing distance associated with the acoustic transducer. as well as A sensor processor, coupled to the acoustic transducer, and configured to: After the active transmission of the acoustic signal stops, multiple points of the ringing signal generated by the acoustic transducer are sampled, wherein the sampling occurs during an early portion of the ringing signal before the corresponding returned signal from the surface has been reflected back and received by the acoustic transducer, and the sampling also occurs during a later portion of the ringing signal including the corresponding returned signal. The early portion of the acoustic transducer is used with multiple sampling points to estimate the blank ringing signal, wherein the estimated blank ringing signal represents the performance of the acoustic transducer when no return signal is received. The estimated blank ringing signal is compared with the later portion of the ringing signal; Metrics are generated based on comparison. as well as The metric is used to determine the type of the surface from a plurality of surface types.

9. The sensor processing unit of claim 8, wherein the sensor processor is further configured to: The estimated blank ringing signal is stored for future use with subsequent transmissions of acoustic signals from the acoustic transducer.

10. The sensor processing unit of claim 8, wherein the sensor processor is configured to generate a metric based on comparison, the processor is configured to: The metric is generated by comparing the features of the estimated blank ringing signal with the same features in the later part of the ringing signal, wherein the features are based on at least one of the in-phase component and the quadrature component.

11. The sensor processing unit of claim 10, wherein the sensor processor is configured to generate a metric based on a comparison, the processor is configured to: The metric is generated based on a comparison of the estimated blank ringing signal with the corresponding curve length of the later portion of the ringing signal.

12. The sensor processing unit of claim 8, wherein the sensor processor is configured to generate a metric based on a comparison, the processor is configured to: The metric is generated by comparing the estimated blank ringing signal with the corresponding amplitude of the later portion of the ringing signal.

13. The sensor processing unit of claim 8, wherein the processor is configured to determine the type of the surface from a plurality of surface types using the metric, the processor is configured to: The metric is provided as input to a floor type classifier, which is trained to determine the type of the surface from a plurality of surface types.

14. The sensor processing unit of claim 8, wherein the acoustic signal is in the ultrasonic frequency range.

15. A surface type detection method, comprising: A processor coupled to an acoustic transducer receives a returned signal from the acoustic transducer, wherein the acoustic transducer is configured to send an acoustic signal to a surface and receive a returned signal reflected from the surface, wherein the returned signal corresponds to the sent acoustic signal, and wherein the surface is within a ringing distance associated with the acoustic transducer. In response to the cessation of active transmission of the acoustic signal, the processor samples multiple points of the ringing signal generated by the acoustic transducer, wherein the sampling occurs during an early portion of the ringing signal before the returned signal from the surface has been reflected back and received by the acoustic transducer, and the sampling also occurs during a later portion of the ringing signal including the returned signal. The processor uses early portions of multiple sampling points to estimate the blank ringing signal of the acoustic transducer, wherein the estimated blank ringing signal represents the performance of the acoustic transducer when no return signal is received; The processor compares the estimated blank ringing signal with the later portion of the ringing signal; The processor generates the metric based on comparison; as well as The processor uses the metric to determine the type of the surface from a plurality of surface types.

16. The method of claim 15, further comprising: The processor stores the estimated blank ringing signal for future use with subsequent transmissions of acoustic signals from the acoustic transducer.

17. The method of claim 15, wherein generating the metric by the processor based on comparison comprises: The metric is generated by the processor based on a comparison of the features of the estimated blank ringing signal with the same features in the later portion of the ringing signal, wherein the features are based on at least one of in-phase and quadrature components.

18. The method of claim 17, wherein generating the metric by the processor based on comparison comprises: The metric is generated by the processor based on a comparison of the estimated blank ringing signal with the corresponding curve length of the later portion of the ringing signal.

19. The method of claim 15, wherein generating the metric by the processor based on comparison comprises: The processor generates the metric based on a comparison of the estimated blank ringing signal with the corresponding amplitude of the later portion of the ringing signal.

20. The method of claim 15, wherein determining the type of the surface from a plurality of surface types using the metric comprises: The processor provides the metric as input to a floor type classifier, which is trained to determine the type of the surface from a plurality of surface types.

Images