# US 2022/0091257 A1 · Autonomous moving object with radar sensor

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2022/0091257 A1
- Document ID: US20220091257A1
- Status: Published application
- Kind: A1
- Application number: 17/310,102
- Filed: 2020-01-24
- Published: 2022-03-24
- Earliest claimed: 2019-01-24
- Assignees: Acconeer AB
- Applicants: Acconeer AB
- Inventors: Peter Almers; Bo Lincoln; Rikard Nelander; David Montgomery; Gaston Holmén
- Web page: https://patentgrep.com/patent/US20220091257A1

## CPC

- G01S 13/881: {for robotics}
- G05D 1/0257: {using a radar (radar systems designed for anti-collision purposes between land vehicles or between land vehicle and fixed obstacles G01S13/931)}
- A01D 34/008: {for automated or remotely controlled operation}
- G01S 7/412: {based on a comparison between measured values and known or stored values}
- G01S 13/30: using more than one pulse per radar period
- G01S 13/88: Radar or analogous systems specially adapted for specific applications (electromagnetic prospecting or detecting of objects, e.g. near-field detection, G01V3/00)
- G05D 1/0088: {characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours (using knowledge based models G06N5/00)}
- G05D 1/02: Control of position or course in two dimensions
- G05D 1/227: Handing over between remote control and on-board control; Handing over between remote control arrangements
- G05D 1/43: Control of position or course in two dimensions [2D]
- G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G05D: SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- A01D: HARVESTING; MOWING

## Abstract

An autonomous moving object comprising a radar sensor is provided. The radar sensor is configured to, during movement, acquire data sets representing reflections from surface portions located within a distance range, and, at least at a sequence of occasions, illuminate a surface region and acquire a data set representing, for each of a set of distances within said distance range, an amplitude and a phase of reflected radar signals received from surface portions located at said distance. Said surface regions comprise common sub-region illuminated at each of said occasions. A radar signal processor is configured to receive the data sets acquired at each of said sequence of occasions. The received data sets form a collection of data sets, wherein each data set of said collection comprises a data subset pertaining to said common sub-region. A surface classifier processor is configured to output a classification of a surface type of the surface based on said collection of data subsets.

## Claims

### Claim 1

An autonomous moving object comprising:
a radar sensor configured to, during movement of the moving object along a movement path along a surface, acquire radar reflection data sets representing radar reflections originating from reflecting surface portions located within a measurement distance range,
wherein the radar sensor is configured to, at each measurement occasion of a temporal sequence of measurement occasions during the movement of the moving object along the movement path, illuminate a respective surface region and acquire a radar reflection data set pertaining to the illuminated surface region, the radar reflection data set representing, for each measurement distance of a set of measurement distances within said measurement distance range, an amplitude and a phase of reflected radar signals received from reflecting surface portions located at said measurement distance, and wherein said respective surface regions comprise a common sub-region which is illuminated at each one of said measurement occasions;
a radar signal processor configured to receive the acquired radar reflection data sets acquired at each one of said sequence of measurement occasions, wherein the received acquired radar reflection data sets form a collection of radar reflection data sets, wherein each radar reflection data set of said collection comprises a radar reflection data subset pertaining to said common sub-region; and
a surface classifier processor configured to output a classification of a surface type of the surface based on said collection comprising said reflection data subsets pertaining to said common sub-region.

### Claim 2 (depends on 1)

An autonomous moving object according to claim 1, wherein the radar sensor is configured to, at each measurement occasion of each one of a plurality of temporal sequences of measurement occasions, acquire a radar reflection data set representing, for each measurement distance of a set of measurement distances within said measurement distance range, an amplitude and a phase of reflected radar signals received from reflecting surface portions located at said measurement distance;
wherein the radar signal processor is configured to receive the acquired radar reflection data sets acquired at each one of said respective sequences of measurement occasions, wherein the received acquired radar reflection data sets acquired at each one of said respective sequences of measurement occasions forms a respective collection of radar reflection data sets, wherein each radar reflection data set of each respective collection of radar reflection data sets comprises a radar reflection data subset pertaining to a respective common sub-region; and
wherein the surface classifier processor is configured to output a respective classification of a surface type of the surface based on each one of said respective collections.

### Claim 3 (depends on 2)

An autonomous moving object according to claim 2, further comprising a surface classifier post-processor configured to output a time filtered classification based on each one of said respective classifications.

### Claim 4 (depends on 1)

An autonomous moving object according to claim 1, wherein said classification is a binary prediction of surface type or a multiclass prediction of surface type.

### Claim 5 (depends on 1)

An autonomous moving object according to claim 1, wherein the autonomous moving object is operable in a training mode in which the autonomous moving object is configured to move along a training surface of a known surface type, wherein the radar sensor is configured to, during movement along the training surface, acquire radar reflection training data sets representing radar reflections originating from said training surface;
wherein the radar signal processor is configured to receive the acquired radar reflection training data sets acquired during said movement along the training surface, wherein the received acquired radar reflection training data sets form a training collection of radar reflection training data sets; and
wherein the surface classifier processor is configured to receive a said training collection and a parameter representing the surface type of said training surface as a training input.

### Claim 6 (depends on 1)

An autonomous moving object according to claim 1, wherein each acquired radar reflection data set comprises I/Q demodulated samples for each measurement distance of said set of measurement distances within said measurement distance range.

### Claim 7 (depends on 1)

An autonomous moving object according to claim 1, wherein the radar signal processor is configured to pre-process said collection of radar reflection data sets, comprising the radar reflection data subsets pertaining to the common sub-region of the surface, to calculate a radar reflection metric as a function of amplitude, phase, measurement distance and time of measurement occasion, and wherein the surface classifier processor is configured to output the classification of the surface type based on the radar reflection metric.

### Claim 8 (depends on 7)

An autonomous moving object according to claim 7, wherein the radar signal processor is configured to pre-process said collection of radar reflection data sets by calculating one of a frequency domain transform or an auto-covariance based on the radar reflection data sets of the collection.

### Claim 9 (depends on 1)

An autonomous moving object according to claim 1, wherein said radar reflection data sets are acquired at positions along the movement path separated by ¼ of a radar signal wavelength, or less.

### Claim 10 (depends on 1)

An autonomous moving object according to claim 1, wherein the radar sensor is configured to, at each one of said measurement occasions, illuminate said respective surface region with a respective burst of radar pulses and acquire said radar reflection data set by performing time-diluted sampling of reflected radar pulses of said respective burst of radar pulses.

### Claim 11 (depends on 10)

An autonomous moving object according to claim 10, wherein a carrier frequency of each radar pulse lies in the millimeter wave band.

### Claim 12 (depends on 1)

An autonomous moving object according to claim 1, wherein said common sub-region is at least 1 cm2.

### Claim 13 (depends on 1)

An autonomous moving object according to any preceding claim 1, further comprising a controller circuit configured to receive an output from the surface classifier processor and, in response, output a control command for controlling a movement and/or an operation of the moving object.

### Claim 14 (depends on 1)

An autonomous moving object according to claim 1, wherein the autonomous moving object is a robotic vacuum cleaner, a robotic lawn mower, a warehouse robot, an autonomous moving robotic toy or an autonomous moving companion robot.
