# US 12,214,775 B2 · Road surface recognition apparatus, vehicle having the same, and method of controlling vehicle > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,214,775 B2 - Document ID: US12214775B2 - Status: Granted - Kind: B2 - Application number: 17/386,031 - Filed: 2021-07-27 - Published: 2022-03-17 - Granted: 2025-02-04 - Earliest claimed: 2020-09-11 - Assignees: Hyundai Motor Company; Kia Corporation; Industry-University Cooperation Foundation Hanyang University (IUCF-HYU) - Applicants: Hyundai Motor Company; Kia Corporation; Industry-University Cooperation Foundation Hanyang University (IUCF-HYU) - Inventors: Ik Jin Um; Jun Han Kang; Jungho Park; Man Dong Kim; Chung Choo Chung; Seung-Hi Lee; Dae Jung Kim; Jin Sung Kim - Web page: https://patentgrep.com/patent/US12214775B2 ## CPC - B60W 30/02: Control of vehicle driving stability - B60W 40/06: Road conditions - B60W 10/06: including control of combustion engines - B60W 10/18: including control of braking systems - B60W 30/18172: {Preventing, or responsive to skidding of wheels} - G01B 17/08: for measuring roughness or irregularity of surfaces - B60W 2050/0057: {Frequency analysis, spectral techniques or transforms} - B60W 2420/54: Audio sensitive means, e.g. ultrasound - B60W 2520/06: Direction of travel - B60W 2520/10: Longitudinal speed - B60W 2520/105: Longitudinal acceleration - B60W 2520/125: Lateral acceleration - B60W 2520/14: Yaw - B60W 2520/28: Wheel speed - B60W 2540/18: Steering angle - B60W 2552/35: Road bumpiness, e.g. potholes - B60W 2552/40: Coefficient of friction - B60W 2555/20: Ambient conditions, e.g. wind or rain - B60W 2710/06: Combustion engines, Gas turbines - B60W 2710/18: Braking system - B60W: CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT - G01B: MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS ## Abstract A vehicle having a road surface recognition apparatus, may perform obtaining, by a controller, sound data for a sound detected by a sound detector while driving; obtaining, by the controller, driving data for driving information detected by a driving information detector; recognizing, by the controller, a type of road surface based on the obtained sound data and the obtained driving data; and controlling, by the controller, a traction control system based on information related to the recognized type of the road surface. ## Claims ### Claim 1 A road surface recognition apparatus comprising: a memory configured to store one or more instructions; and a processor configured to execute the one or more instructions stored in the memory, wherein the instructions, when executed by the processor, cause the processor to: obtain sound data for a sound detected by a sound detecting sensor; obtain driving data of a vehicle for driving information detected by a driving information detecting sensor; and recognize a type of road surface according to the sound data and the driving data, and wherein the processor is further configured to: transform the obtained sound data into sound data in a frequency domain; obtain a first feature vector using longitudinal acceleration data and driving speed data among the driving data, and the sound data in the frequency domain; classify a first type of the road surface in a first order using the first feature vector and a first classifier model; obtain a second feature vector according to a plurality of wheel speed data, the longitudinal acceleration data, steering angle data, lateral acceleration data, and yaw rate data among the driving data; classify a second type of the road surface in a second order using the second feature vector and a second classifier model; and determine the type of the road surface by selecting at least one of the first type of the road surface and the second type of the road surface based on the longitudinal acceleration data and reference longitudinal acceleration data. ### Claim 2 (depends on 1) The road surface recognition apparatus according to claim 1, wherein the processor is configured to: according to the longitudinal acceleration data and the reference longitudinal acceleration data, in a response to the longitudinal acceleration being less than the reference longitudinal acceleration, select the first type of the road surface, and in a response to the longitudinal acceleration being greater than or equal to the reference longitudinal acceleration, select the second type of the road surface. ### Claim 3 (depends on 1) The road surface recognition apparatus according to claim 1, the processor is configured to: recognize the type of the road surface according to the selected type of the road surface and state transition information for a remaining type of the road surface. ### Claim 4 (depends on 1) The road surface recognition apparatus according to claim 1, wherein the first classifier model includes a support vector machine; and wherein the second classifier model includes a deep neural network. ### Claim 5 (depends on 1) The road surface recognition apparatus according to claim 1, wherein the processor is configured to receive the sound data through an Electronic Stability Control (ESC) system. ### Claim 6 A vehicle comprising: a sound detecting sensor; a driving information detecting sensor; and a controller electrically connected to the sound detecting sensor and the driving information detecting sensor and configured to: obtain sound data for a sound detected by the sound detecting sensor, obtain driving data for driving information detected by the driving information detecting sensor, recognize a type of road surface according to the obtained sound data and the obtained driving data, and control a driving force of a vehicle according to information related to the recognized type of the road surface, wherein the controller is configured to: transform the obtained sound data into sound data in a frequency domain; obtain a first feature vector using longitudinal acceleration data and driving speed data among the driving data, and the sound data in the frequency domain; classify a first type of the road surface in a first order using the first feature vector and a first classifier model; obtain a second feature vector according to a plurality of wheel speed data, the longitudinal acceleration data, steering angle data, lateral acceleration data, and yaw rate data among the driving data; classify a second type of the road surface in a second order using the second feature vector and a second classifier model; and determine the type of the road surface by selecting at least one of the first type of the road surface and the second type of the road surface based on the longitudinal acceleration data and reference longitudinal acceleration data. ### Claim 7 (depends on 6) The vehicle according to claim 6, wherein the controller is configured to: according to the longitudinal acceleration data and the reference longitudinal acceleration data, in a response to the longitudinal acceleration being less than the reference longitudinal acceleration, select the first type of the road surface, and in a response to the longitudinal acceleration being greater than or equal to the reference longitudinal acceleration, select the second type of the road surface. ### Claim 8 (depends on 6) The vehicle according to claim 6, wherein the controller is configured to: recognize the type of the road surface according to the selected type of the road surface and state transition information for a remaining type of the road surface. ### Claim 9 (depends on 6) The vehicle according to claim 6, wherein the first classifier model includes a support vector machine; and wherein the second classifier model includes a deep neural network. ### Claim 10 (depends on 6) The vehicle according to claim 6, wherein the sound detecting sensor is mounted on a right rear side of a vehicle body. ### Claim 11 (depends on 6) The vehicle according to claim 6, further including: a traction control system configured to control the driving force in a response to a control command of the controller. ### Claim 12 (depends on 6) The vehicle according to claim 6, wherein the controller is configured to: obtain a friction coefficient corresponding to the information related to the recognized type of the road surface, and control an operation of at least one of a brake and an engine based on the obtained friction coefficient. ### Claim 13 A method of controlling a vehicle, the method comprising: obtaining, by a controller, sound data for a sound detected by a sound detecting sensor while driving; obtaining, by the controller, driving data of the vehicle for driving information detected by a driving information detecting sensor; recognizing, by the controller, a type of road surface according to the obtained sound data and the obtained driving data; and controlling, by the controller, a traction control system according to information related to the recognized type of the road surface, wherein the controller is further configured to: transform the obtained sound data into sound data in a frequency domain; obtain a first feature vector using longitudinal acceleration data and driving speed data among the driving data, and the sound data in the frequency domain; classify a first type of the road surface in a first order using the first feature vector and a first classifier model; obtain a second feature vector according to a plurality of wheel speed data, the longitudinal acceleration data, steering angle data, lateral acceleration data, and yaw rate data among the driving data; classify a second type of the road surface in a second order using the second feature vector and a second classifier model; and determine the type of the road surface by selecting at least one of the first type of the road surface and the second type of the road surface based on the longitudinal acceleration data and reference longitudinal acceleration data.