USOO9063543B2
`
`(12) United States Patent
`An et al.
`
`(10) Patent No.:
`(45) Date of Patent:
`
`US 9,063,543 B2
`Jun. 23, 2015
`
`(54) APPARATUS AND METHOD FOR
`COOPERATIVE AUTONOMOUS DRIVING
`BETWEEN VEHICLE AND DRIVER
`
`(56)
`
`References Cited
`U.S. PATENT DOCUMENTS
`
`(71) Applicant: Electronics and Telecommunications
`Research Institute, Daejeon (KR)
`(72) Inventors: Kyoung-Hwan An, Daejeon (KR):
`Woo-Yong Han, Daejeon (KR)
`
`(73) Assignee: ELECTRONICS AND
`TELECOMMUNICATIONS
`RESEARCH INSTITUTE, Daejeon
`(KR)
`Subject to any disclaimer, the term of this
`patent is extended or adjusted under 35
`U.S.C. 154(b) by 0 days.
`
`(*) Notice:
`
`(21) Appl. No.: 14/023,873
`
`(22) Filed:
`
`Sep. 11, 2013
`
`(65)
`
`(30)
`
`Prior Publication Data
`US 2014/0244096 A1
`Aug. 28, 2014
`
`Foreign Application Priority Data
`
`Feb. 27, 2013 (KR) ........................ 10-2013-0021258
`
`(2006.01)
`(2006.01)
`
`(51) Int. Cl.
`GOIC 22/00
`G05D I/00
`(52) U.S. Cl.
`CPC .................................... G05D I/0055 (2013.01)
`(58) Field of Classification Search
`CPC ..... G05D 1/005; G05D 1/0274; G05D 1/024;
`G05D 1/0278; G05D 1/0027
`USPC ............. 701/24, 45,93, 97,98; 340/.575, 576
`See application file for complete search history.
`
`2/2005 Victor ........................... 340,576
`2005/0030.184 A1
`2007/0052530 A1* 3, 2007 Diebold et al. ........
`... 340/467
`2007/0225882 A1* 9/2007 Yamaguchi et al. ............ TO1/36
`2009/0326796 A1* 12/2009 Prokhorov .....
`TO1,200
`2011/0241862 A1* 10, 2011 Debouk et all
`... 340,439
`2012 0083960 A1* 4/2012 Zhu et al. ........................ TO1/23
`2012/0323479 A1* 12/2012 Nagata .......
`... 701,301
`2013/013 1907 A1* 5, 2013 Green et al. .................... TO1/23
`
`FOREIGN PATENT DOCUMENTS
`
`3, 1997
`10, 2006
`8, 2010
`
`1997-0011789
`KR
`10-2006-01-10299
`KR
`10-2010-0088943
`KR
`* cited by examiner
`Primary Examiner — Helal A Algahaim
`Assistant Examiner — Shardul Patel
`(74) Attorney, Agent, or Firm — Kile Park Reed &
`Houtteman PLLC
`
`ABSTRACT
`(57)
`The present invention relates to an apparatus and method for
`performing cooperative autonomous driving between a
`vehicle and a driver. For this, a cooperative autonomous driv
`ing apparatus according to the present invention includes a
`driver State determination unit for determining a state of a
`driver and calculating the state of the driver as a riskindex. An
`autonomous driving control unit classifies section character
`istics of respective sections included in a path to a destination
`corresponding to the driver based on section data stored in a
`database (DB), and controls autonomous driving of a vehicle
`in which the driver is riding, based on a driving environment
`recognized for the path to the destination corresponding to the
`driver. A driving control determination unit determines driv
`ing modes of the respective sections included in the path
`based on the state of the driver and the section characteristics.
`
`21 Claims, 6 Drawing Sheets
`
`100
`
`
`
`130
`
`DRIVING CONTROL
`DETERMINATION
`UNIT
`
`110
`
`DRIVER STATE
`DETERMINATION
`UNIT
`
`AUTONOMOUS
`DRIVING CONTROL -120
`UNIT
`
`AUTONOMOUS
`DRIVING DATA
`PROCESSING UNIT
`
`140
`
`IPR2025-00944
`Tesla EX1008 Page 1
`
`

`

`U.S. Patent
`
`Jun. 23, 2015
`
`Sheet 1 of 6
`
`US 9,063,543 B2
`
`100
`O
`
`
`
`110
`
`130
`
`
`
`DRIVING CONTROL
`DETERMINATION
`UNIT
`
`DRIVER STATE
`DETERMINATION
`UNIT
`
`AUTONOMOUS
`DRIVING CONTROL
`UNIT
`
`AUTONOMOUS
`DRIVING DATA
`PROCESSING UNIT
`
`120
`
`140
`
`FG. 1
`
`IPR2025-00944
`Tesla EX1008 Page 2
`
`

`

`U.S. Patent
`
`Jun. 23, 2015
`
`Sheet 2 of 6
`
`US 9,063,543 B2
`
`110
`
`DRIVER STATE SENSOR
`MODULE
`
`111
`
`DRIVER STATE
`RECOGNITION MODULE
`
`112
`
`RISK INDEX
`DETERMINATION MODULE
`
`113
`
`FIG 2
`
`120
`
`
`
`DRIVING ENVIRONMENT
`RECOGNITION MODULE
`
`121
`
`AUTONOMOUS DRIVING
`MODULE
`
`122
`
`FG. 3
`
`IPR2025-00944
`Tesla EX1008 Page 3
`
`

`

`U.S. Patent
`
`Jun. 23, 2015
`
`Sheet 3 of 6
`
`US 9,063,543 B2
`
`17
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`MANUAL
`DRIVING
`
`DRIVER
`LEADING
`COOPERATIVE
`DRIVING
`
`
`
`
`
`
`
`
`
`EMERGENCY
`DRIVING FOR
`FAILURE IN
`AUTO
`NOMOUS
`DRIVING
`
`NOMOUS
`DRIVING
`LEADING
`
`AUTONOMOU
`DRIVING
`
`
`
`EMERGENCY
`16 DRIVING FOR
`NON
`REACTION OF
`DRIVER
`
`18
`
`FG. 4
`
`IPR2025-00944
`Tesla EX1008 Page 4
`
`

`

`U.S. Patent
`
`Jun. 23, 2015
`
`Sheet 4 of 6
`
`US 9,063,543 B2
`
`
`
`F.G. 5
`
`IPR2025-00944
`Tesla EX1008 Page 5
`
`

`

`U.S. Patent
`
`Jun. 23, 2015
`
`Sheet 5 of 6
`
`US 9,063,543 B2
`
`INPUT DESTINATION FROM USER
`
`S110
`
`
`
`
`
`
`
`S120
`
`DETERMINE DRIVER STATE
`
`
`
`
`
`
`
`ACQUIRE DRIVING
`SECTION CLASSIFICATION
`INFORMATION
`
`S130
`
`
`
`
`
`
`
`
`
`S140
`
`S150
`
`S16O
`
`S18O
`
`DETERMINE DRIVING MODES
`OF CURRENT SECTION AND
`SUBSEQUENT SECTION
`
`CONTROL VEHICLE ACCORDING
`TO DRIVING MODE
`
`CALCULATE RISK INDEX OF CURRENT
`SECTION AND UPDATE INFORMATION
`OF PREVIOUS SECTION
`
`
`
`ARRIVEDAT
`DESTINATION?
`
`F.G. 6
`
`IPR2025-00944
`Tesla EX1008 Page 6
`
`

`

`U.S. Patent
`
`Jun. 23, 2015
`
`Sheet 6 of 6
`
`US 9,063,543 B2
`
`START
`
`S161
`
`INITIALIZE
`
`S162
`
`ACQUIRE CURRENT
`LOCATION
`
`
`
`
`
`S163
`
`ACQUIRABLE
`AND FALL WITHINERROR
`RANGE2
`
`NO
`
`YES
`sau ACQUIRECURRENTSECTION
`INFORMATION
`S167
`
`NO
`
`YES
`CLASSIFY PREVIOUS SECTION
`BASED ON RISK INDEX OF
`PREVIOUS SECTION
`
`S169
`
`CALCULATE RISK INDEX
`
`S170
`
`NO
`
`
`
`
`
`CALCULATED
`RISK INDEX > RISK INDE
`OF CURRENT
`SECTION2
`YES
`
`
`
`
`
`
`
`
`
`UPDATE RISK INDEX
`
`S18O
`
`FG. 7
`
`S165
`
`PREDICT CURRENT
`LOCATION AND ACQUIRE
`PREDICTED SECTION
`
`SET RISK INDEX OF
`CURRENT SECTION TO
`MAXIMUM VALUE
`
`S166
`
`IPR2025-00944
`Tesla EX1008 Page 7
`
`

`

`US 9,063,543 B2
`
`1.
`APPARATUS AND METHOD FOR
`COOPERATIVE AUTONOMOUS DRIVING
`BETWEEN VEHICLE AND DRIVER
`
`CROSS REFERENCE TO RELATED
`APPLICATION
`
`This application claims the benefit of Korean Patent Appli
`cation No. 10-2013-0021258 filed on Feb. 27, 2013, which is
`hereby incorporated by reference in its entirety into this appli
`cation.
`
`10
`
`BACKGROUND OF THE INVENTION
`
`2
`ing the state of the driver as a risk index; an autonomous
`driving control unit for classifying section characteristics of
`respective sections included in a path to a destination corre
`sponding to the driver based on section data stored in a data
`base (DB), and controlling autonomous driving of a vehicle in
`which the driver is riding, based on a driving environment
`recognized for the path to the destination corresponding to the
`driver; and a driving control determination unit for determin
`ing driving modes of the respective sections included in the
`path based on the state of the driver and the section charac
`teristics.
`Preferably, each of the driving modes of the sections may
`include one of a manual mode, a cooperative driving mode, an
`emergency driving mode for a failure in autonomous driving,
`and an emergency driving mode for non-reaction of the
`driver, and the cooperative driving mode may include a
`driver-leading cooperative driving mode and an autonomous
`driving-leading cooperative driving mode.
`Preferably, the driving control determination unit may be
`capable of changing a current driving mode to another driving
`mode corresponding to the state of the driver and the section
`characteristics in a section in which the vehicle is traveling.
`Preferably, the cooperative autonomous driving apparatus
`may further include an autonomous driving data processing
`unit for, after the vehicle has moved to a Subsequent section,
`reclassifying the section characteristics using risk indices of
`the respective sections calculated based on the driving envi
`ronment, and updating the section data stored in the DB based
`on the reclassified section characteristics.
`Preferably, the autonomous driving data processing unit
`may be configured to, after the vehicle has moved to the
`subsequent section, set a previous section to a cooperative
`driving possible section if a risk index of the previous section
`is greater than a preset driving possible risk index and is less
`than a preset driving impossible risk index.
`Preferably, the autonomous driving data processing unit
`may be configured to, after the vehicle has moved to the
`Subsequent section, set a previous section to an autonomous
`driving impossible section if a risk index of the previous
`section is greater than a preset driving impossible risk index.
`Preferably, the autonomous driving data processing unit
`may be configured to, after the vehicle has moved to the
`Subsequent section, set a previous section to an autonomous
`driving possible section if a risk index of the previous section
`is less than a preset driving possible risk index.
`Preferably, the autonomous driving data processing unit
`may be configured to, when the vehicle is traveling in a
`current section, calculate a risk index of the current section,
`and update a previously stored risk index of the current sec
`tion to the calculated risk index if the calculated risk index is
`greater than the previously stored risk index of the current
`section.
`Preferably, the driver state determination unit may deter
`mine a dozing state or an inattentive state of the driver using
`an eyetracker for tracking eye blinking or line of sight of the
`driver.
`Preferably, the driver state determination unit may deter
`mine an inattentive state of the driver by checking whether a
`nomadic device or a device in the vehicle has been used.
`Preferably, the driver state determination unit may include
`a risk index determination module for converting the State of
`the driver into a numerical risk index.
`In accordance with another aspect of the present invention
`to accomplish the above objects, there is provided a coopera
`tive autonomous driving method including determining, by a
`driver state determination unit, a state of a driver and calcu
`lating the state of the driver as a risk index; classifying, by an
`
`15
`
`25
`
`30
`
`1. Technical Field
`The present invention relates generally to an apparatus and
`method for performing cooperative autonomous driving (co
`piloting) between a vehicle and a driver and, more particu
`larly, to a cooperative autonomous driving apparatus and
`method, which determine driving performance depending on
`the states of the vehicle and the driver and dynamically
`decides on an agent of driving when the driver is driving the
`vehicle on a road.
`2. Description of the Related Art
`Currently, driver assistance systems for assisting a driver
`who is driving a vehicle are available. For example, driver
`assistance systems include an Adaptive Cruise Control
`(ACC) system, a Lane Departure Warning System (LDWS), a
`Lane Keeping System (LKS), etc. Such a driver assistance
`system is advantageous in that it partially assists a driver with
`longitudinal or lateral control of the vehicle, thus making
`driving more convenient. In contrast, such a driver assistance
`system has a limitation in that it must prepare for the driver's
`intervention. Therefore, there is a disadvantage in that when
`the driver dozes off while driving, or cannot drive due to his or
`her health condition, a conventional driver assistance system
`cannot assist the driver.
`Further, research into an autonomous driving vehicle
`capable of driving from an origin to a destination without
`intervention of the driver has recently been conducted. How
`40
`ever, there is a problem in that error in the recognition and
`determination of sensors may occur depending on a driving
`environment including road or weather conditions, thus mak
`ing it impossible to consistently guarantee the safety of the
`driver.
`In relation to this, Korean Patent Application Publication
`No. 10-2006-01 10299 entitled “Method and apparatus for
`reducing damage caused by accidents' is disclosed.
`
`35
`
`45
`
`SUMMARY OF THE INVENTION
`
`50
`
`Accordingly, the present invention has been made keeping
`in mind the above problems occurring in the prior art, and an
`object of the present invention is to provide a cooperative
`autonomous driving apparatus and method, which can allo
`cate driving control to an object having high driving perfor
`mance by taking into consideration a driving situation, the
`state of a driver, and the performance of an autonomous
`driving apparatus.
`Another object of the present invention is to provide a
`cooperative autonomous driving apparatus and method,
`which enable a vehicle to be cooperatively driven between the
`vehicle and a driver.
`In accordance with an aspect of the present invention to
`accomplish the above objects, there is provided a cooperative
`autonomous driving apparatus including a driver state deter
`mination unit for determining a state of a driver and calculat
`
`55
`
`60
`
`65
`
`IPR2025-00944
`Tesla EX1008 Page 8
`
`

`

`3
`autonomous driving control unit, section characteristics of
`respective sections included in a path to a destination corre
`sponding to the driver based on section data stored in a data
`base (DB); determining, by a driving control determination
`unit, driving modes of the respective sections included in the
`path based on the state of the driver and the section charac
`teristics; and controlling, by the autonomous driving control
`unit, autonomous driving of a vehicle in which the driver is
`riding, based on a driving environment recognized for the
`path to the destination corresponding to the driver.
`Preferably, each of the driving modes of the sections may
`include one of a manual mode, a cooperative driving mode, an
`emergency driving mode for a failure in autonomous driving,
`and an emergency driving mode for non-reaction of the
`driver, and the cooperative driving mode includes a driver
`leading cooperative driving mode and an autonomous driv
`ing-leading cooperative driving mode.
`Preferably, determining the driving modes of the sections
`may be configured to be capable of changing a current driving
`mode to another driving mode corresponding to the state of
`the driver and the section characteristics in a section in which
`the vehicle is traveling.
`Preferably, the cooperative autonomous driving method
`may further include after controlling the autonomous driving
`of the vehicle in which the driver is riding, after the vehicle
`has moved to a Subsequent section, reclassifying, by the
`autonomous driving data processing unit, the section charac
`teristics using risk indices of the respective sections calcu
`lated based on the driving environment, and updating the
`section data stored in the DB based on the reclassified section
`characteristics.
`Preferably, updating the section data may be configured to,
`after the vehicle has moved to the Subsequent section, set a
`previous section to a cooperative driving possible section if a
`risk index of the previous section is greater than a preset
`driving possible risk index and is less than a preset driving
`impossible risk index.
`Preferably, updating the section data may be configured to,
`after the vehicle has moved to the Subsequent section, set a
`previous section to an autonomous driving impossible section
`if a risk index of the previous section is greater than a preset
`driving impossible risk index.
`Preferably, updating the section data may be configured to,
`after the vehicle has moved to the Subsequent section, set a
`previous section to an autonomous driving possible section if
`a risk index of the previous section is less than a preset driving
`possible risk index.
`Preferably, updating the section data may be configured to,
`when the vehicle is traveling in a current section, calculate a
`risk index of the current section, and update a previously
`stored risk index of the current section to the calculated risk
`index if the calculated risk index is greater than the previously
`stored risk index of the current section.
`
`10
`
`15
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`US 9,063,543 B2
`
`4
`FIG. 3 is a block diagram showing the autonomous driving
`control unit of the cooperative autonomous driving apparatus
`according to an embodiment of the present invention;
`FIG. 4 is a diagram illustrating examples of the change of
`a driving mode which can be performed by the driving control
`determination unit of the cooperative autonomous driving
`apparatus according to an embodiment of the present inven
`tion;
`FIG. 5 is a diagram showing an embodiment to which the
`cooperative autonomous driving apparatus according to the
`present invention is applied;
`FIG. 6 is a flowchart showing a cooperative autonomous
`driving method according to an embodiment of the present
`invention; and
`FIG. 7 is a flowchart showing a driving section analysis and
`update method in the cooperative autonomous driving
`method according to an embodiment of the present invention.
`
`DESCRIPTION OF THE PREFERRED
`EMBODIMENTS
`
`The present invention will be described in detail below
`with reference to the accompanying drawings. In the follow
`ing description, redundant descriptions and detailed descrip
`tions of known functions and elements that may unnecessar
`ily make the gist of the present invention obscure will be
`omitted. Embodiments of the present invention are provided
`to fully describe the present invention to those having ordi
`nary knowledge in the art to which the present invention
`pertains. Accordingly, in the drawings, the shapes and sizes of
`elements may be exaggerated for the sake of clearer descrip
`tion.
`FIG. 1 is a block diagram showing a cooperative autono
`mous driving apparatus 100 according to an embodiment of
`the present invention. Below, the cooperative autonomous
`driving apparatus 100 according to the embodiment of the
`present invention will be described in detail with reference to
`FIG. 1. The cooperative autonomous driving apparatus 100
`according to the embodiment of the present invention
`includes a driver state determination unit 110, an autonomous
`driving control unit 120, a driving control determination unit
`130, and an autonomous driving data processing unit 140.
`The components of the cooperative autonomous driving
`apparatus 100 are described below.
`The driver state determination unit 110 functions to deter
`mine an abnormal state of a driver, such as dozing off inat
`tention, or non-response of the driver. The driver state deter
`mination unit 110 tracks the blinking of the driver's eyes or
`the line of sight using, for example, an eyetracker, and then
`determines whether the driver dozes off or is inattentive.
`Further, the driver state determination unit 110 may check
`whether a nomadic terminal or each device in the vehicle is
`used and then determine whether the driver is inattentive.
`Furthermore, the driver state determination unit 110 may
`determine the state of the driver based on driving information,
`Such as the speed, steering angle, and variable speed of the
`vehicle calculated by the autonomous driving control unit
`120. Furthermore, the driver state determination unit 110 may
`convert the state of the driver into a risk index. The risk index
`of the driver may be determined using the following Equation
`(1):
`
`driver risk index=(driver state risk index)*(driving
`(1)
`environment risk index)
`As shown in Equation (1), the driver risk index is calculated
`by multiplying a driver state risk index by a driving environ
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`55
`
`The above and other objects, features and advantages of the
`present invention will be more clearly understood from the
`following detailed description taken in conjunction with the
`accompanying drawings, in which:
`FIG. 1 is a block diagram showing a cooperative autono
`mous driving apparatus according to an embodiment of the
`present invention;
`FIG. 2 is a block diagram showing a driver state determi
`nation unit included in the cooperative autonomous driving
`apparatus according to an embodiment of the present inven
`tion;
`
`60
`
`65
`
`IPR2025-00944
`Tesla EX1008 Page 9
`
`

`

`5
`ment risk index. Here, the driver state risk index is calculated
`using the following Equation (2):
`(2)
`driver state risk index fidriver state)
`In Equation (2), function f() denotes a mapping function.
`Therefore, the driver state risk index is calculated by mapping
`the driver state to a lookup table configured using experi
`ments between risk indices. Here, the lookup table is a table in
`which the degrees of error in recognition or determination are
`converted into risk indices depending on the degrees of the
`driver state, for example, dozing off while driving, inatten
`tion, and non-reaction, and the risk indices are stored. Further,
`the driving environment risk index shown in Equation (1) is
`calculated by the following Equation (3):
`driving environment risk index=1/(W1*Min(TTC)+
`
`10
`
`15
`
`As shown in Equation (3), the driving environment risk
`index is an index for measuring the dynamic environment of
`a road, that is, a driving environment. That is, the driving
`environment risk index is an index obtained by converting a
`risk of collision between a vehicle in which the driver is riding
`and neighboring vehicles into a numerical value. For this,
`weights W1 and W2 are respectively multiplied by a mini
`mum value of Time To Collision (TTC) values with neigh
`boring vehicles and a minimum value of Inter Vehicular Time
`(TIV) values with the neighboring vehicles, the sum of these
`multiplication results is obtained, and then a resulting value is
`obtained in the form of an inverse function. Here, equations
`required to obtain TTC and TIV are given in the following
`Equations (4) and (5):
`
`25
`
`30
`
`(relative distance between host vehicle and target vehicle)
`T (relative speed between host vehicle and target vehicle)
`(relative distance between host vehicle and target vehicle)
`TV =
`(host vehicle speed)
`
`(4)
`
`(5)
`
`35
`
`The autonomous driving control unit 120 functions to clas
`Sify the section characteristics of a path based on section data
`stored in a separate database (DB), and control the autono
`mous driving of the vehicle based on a driving environment
`recognized for a path to the destination corresponding to the
`driver. Here, the section characteristics are classified into an
`autonomous driving possible section, an autonomous driving
`impossible section, a cooperative driving possible section,
`and an undecided section. In this case, the autonomous driv
`ing possible section denotes a section in which an autono
`mous driving risk index is less than a preset driving possible
`risk index. The cooperative driving possible section denotes a
`section in which the autonomous driving risk index is greater
`than the preset driving possible risk index and is less than a
`preset driving impossible risk index. The autonomous driving
`impossible section denotes a section in which the autono
`mous driving risk index is greater than the preset driving
`impossible risk index. Finally, the undecided section denotes
`a section in which a vehicle has not yet been driven. In the
`above description, the autonomous driving risk index may be
`calculated by the following Equation (6):
`autonomous driving risk index=(autonomous driving
`state risk index)*(driving environment risk
`(6)
`index)
`As shown in Equation (6), the autonomous driving risk
`index is calculated by multiplying an autonomous driving
`state risk index by a driving environment risk index. Here, the
`autonomous driving State risk index is an index indicating
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`US 9,063,543 B2
`
`6
`whether basic autonomous driving is possible in the driving
`environment. Such an autonomous driving state risk index
`may be calculated by the following Equation (7):
`
`autonomous driving state risk index=W1* recognition
`error--W2* determination error--W3*control
`error--W4*map data error
`(7)
`Referring to Equation 7, the autonomous driving state risk
`index is calculated by multiplying weights W1, W2. W3, and
`W4 by a recognition error, a determination error, a control
`error, and a map data error, respectively, and by obtaining the
`sum of multiplication results. In this case, the weights W1,
`W2. W3, and W4 are values determined according to the
`importance level used in an autonomous driving algorithm.
`Here, the recognition error includes a GPS reception error, a
`lane recognition error, an obstacle recognition error, a traffic
`light recognition error, etc. Further, the determination error
`includes errors in driving behavior, such as maintaining a
`lane, changing a lane and going through an intersection. Fur
`ther, the control error denotes a difference between a planned
`path and an actually tracked path. Finally, the map data error
`denotes a difference between map data stored in a database
`(DB), as in the case of a navigation terminal, and actually
`recognized map data.
`The autonomous driving control unit 120 may use a loca
`tion acquisition device. Such as a Global Positioning System
`(GPS) or an Inertial Navigation System (INS), in order to
`acquire the location of the vehicle in which the driver is
`riding. Further, the autonomous driving control unit 120 may
`recognize a driving environment using a camera, lidar, radar,
`or the like So as to recognize a road and an obstacle. Further
`more, the autonomous driving control unit 120 may utilize an
`intra-vehicle sensor or the like so as to detect the states of the
`vehicle, for example, the speed, acceleration, Steering angle,
`yaw angle, etc. of the vehicle.
`In addition, the autonomous driving control unit 120
`executes a path planning and behavior control algorithm
`required to conduct autonomous driving. Here, the autono
`mous driving control unit 120 may calculate an autonomous
`driving possible section based on the section information
`updated by the autonomous driving data processing unit 140.
`Further, the autonomous driving control unit 120 may
`exchange information, such as driving mode change notifica
`tion, risk information notification, and driver command rec
`ognition, with the driver, through a Human Vehicle Interface
`(HVI). Furthermore, the autonomous driving control unit 120
`may automatically control the vehicle using a vehicle actua
`tor, or exchange various types of information, Such as the
`location, speed, lane change intention, and event information
`of a neighboring vehicle, with the neighboring vehicle.
`The driving control determination unit 130 determines the
`driving modes of respective sections included in the path to a
`destination, based on the driver state determined by the driver
`state determination unit 110 and the section characteristics
`classified by the autonomous driving control unit 120. Here,
`the driving modes are configured to include a manual mode,
`a cooperative driving (co-piloting) mode, an autonomous
`driving mode, an emergency driving mode for a failure in
`autonomous driving, and an emergency driving mode for the
`non-reaction of a driver. Further, the cooperative driving
`mode includes a driver-leading cooperative driving mode in
`which the driver leads driving and the cooperative autono
`mous driving apparatus assists driving, and an autonomous
`driving-leading cooperative driving mode in which the coop
`erative autonomous driving apparatus leads driving and the
`driver assists driving. Individual driving modes are described
`in detail below.
`
`IPR2025-00944
`Tesla EX1008 Page 10
`
`

`

`US 9,063,543 B2
`
`10
`
`15
`
`7
`First, the manual mode is a mode in which the driver
`performs all of recognition, determination, and control. The
`driver-leading cooperative driving mode is a mode in which
`the driver makes a final decision on all of recognition, deter
`mination, and control and drives a vehicle, and in which the
`cooperative autonomous driving apparatus 100 provides rec
`ognition information, determination information, control
`information, etc. to the driver. That is, the driver-leading
`cooperative driving mode is a mode in which the driver leads
`driving and the cooperative autonomous driving apparatus
`assists the driver in driving. The autonomous driving-leading
`cooperative driving mode is a mode in which the cooperative
`autonomous driving apparatus performs all of recognition,
`determination, and control, and if necessary, driving is per
`formed under the confirmation of the driver. Here, the driver
`must continuously monitor the driving of the cooperative
`autonomous driving apparatus, and may intervene the control
`of driving if necessary. The autonomous driving mode is a
`mode in which the driver does not intervene the driving of the
`cooperative autonomous driving apparatus and the coopera
`tive autonomous driving apparatus performs all of recogni
`tion, determination, and control. The emergency driving
`mode for a failure in autonomous driving is a mode prepared
`against a failure in autonomous driving. That is, the emer
`gency driving mode for a failure in autonomous driving is
`configured to Suddenly stop the vehicle, turn on/off an emer
`gency lamp, or contact a call center. The emergency driving
`mode for the non-reaction of a driver is a mode performed in
`a case where, due to the impossibility of autonomous driving,
`it is intended to hand the control of driving over to the driver,
`but the driver does not make a response. In the emergency
`driving mode for the non-reaction of the driver, activity such
`as stopping the vehicle at the shoulder of a road, turning
`on/off the emergency lamp, or contacting a call center may be
`taken.
`The autonomous driving data processing unit 140 reclas
`sifies section characteristics based on the risk indices of pre
`vious sections after the vehicle in which the driver rides has
`passed through a plurality of sections on the path. In greater
`detail, the autonomous driving data processing unit 140
`reclassifies section characteristics using the risk indices of the
`previous sections calculated based on the driving environ
`ment after the vehicle has moved to a Subsequent section.
`Thereafter, pieces of section data stored in the DB are updated
`based on the reclassified section characteristics. By means of
`45
`this, the vehicle uses updated information when passing
`through the plurality of sections later, thus enabling safer
`driving to be conducted. A procedure performed by the
`autonomous driving data processing unit 140 is described
`below.
`First, the initialization of devices required for update is
`performed. Thereafter, the autonomous driving control unit
`120 acquires the current location of a recognized vehicle.
`Here, when the acquisition of the current location fails or
`when the current location falls out of a predicted error range,
`the current location is predicted based on existing data. There
`after, a current section is predicted by performing map match
`ing based on the predicted location. Then, the risk index of the
`current section stored in section data is set to a maximum
`value. The reason for this is that it is difficult to currently
`acquire the location from the corresponding section, thus
`causing various problems in autonomous driving. In contrast,
`when the acquisition of the current location Succeeds and the
`current location falls within the predicted error range, current
`section information is acquired. Based on the current section
`information, it is determined whether the vehicle in which the
`driver is riding has entered a new section. In this case, if it is
`
`55
`
`8
`determined that the vehicle has not entered a new section, an
`autonomous driving risk index is calculated, and the calcu
`lated risk index is compared with a value previously stored in
`the section data. As a result of the comparison, if the calcu
`lated risk index is greater than the value stored in the section
`data, the value Stored in the section data is updated to the
`calculated risk index. Further, if it is determined that the
`vehicle has entered the new section, a previous section is
`reclassified based on the risk index of the previous section.
`Such reclassification is performed by comparing the risk
`index of the previous section, a preset driving possible risk
`index, and a preset driving impossible risk index. That is, if
`the risk index of the previous section is less than the driving
`possible risk index, the previous section is set to an autono
`mous driving possible section. Further, if the risk index of the
`previous section is greater than the preset driving possible
`risk index and is less than the preset driving impossible risk
`index, the previous section is set to a cooperative driving
`possible section. Furthermore, if the risk index of the previous
`section is greater than the preset driving impossible risk
`index, the previous section is set to an autonomous driving
`impossible section.
`FIG. 2 is a block diagram showing the driver state deter
`mination unit 110 of FIG.