# US 11,993,254 B2 · Route search system and method for autonomous parking based on cognitive sensor > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,993,254 B2 - Document ID: US11993254B2 - Status: Granted - Kind: B2 - Application number: 17/528,350 - Filed: 2021-11-17 - Published: 2022-06-30 - Granted: 2024-05-28 - Earliest claimed: 2020-12-30 - Assignees: Hyundai Motor Company; Kia Corporation - Applicants: Hyundai Motor Company; Kia Corporation - Inventors: Kyujin Park; Junho Yang - Web page: https://patentgrep.com/patent/US11993254B2 ## CPC - B60W 30/06: Automatic manoeuvring for parking - B60W 60/0011: {involving control alternatives for a single driving scenario, e.g. planning several paths to avoid obstacles} - B60W 60/0015: {specially adapted for safety} - G01C 21/3407: {specially adapted for specific applications} - G06V 20/56: exterior to a vehicle by using sensors mounted on the vehicle - G06V 20/58: Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads - G06V 20/586: {of parking space} - 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 - G01C: MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY (measuring liquid level G01F; radio navigation, determining distance or velocity by use of propagation effects, e.g. Doppler effects, propagation time, of radio waves, analogous arrangements using other waves G01S) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A cognitive sensor-based autonomous parking route search system includes a cognitive sensor configured to obtain cognitive information for determining an existence and a location of an obstacle by sensing a space around a vehicle being driven in an inside of a parking lot, a global path search module configured to generate a global path for the inside of the parking lot as a node map by searching for a free space, in which the vehicle is capable of being driven, based on the cognitive information and by setting a node in the free space, and an optimal local path generation module configured to generate an optimal local path from a current location of the vehicle to a destination by connecting the node set in the free space while the vehicle moves inside the parking lot. ## Claims ### Claim 1 A cognitive sensor-based autonomous parking route search system, the system comprising: a cognitive sensor configured to obtain cognitive information for determining an existence and a location of an obstacle by sensing a space around a vehicle being driven in an inside of a parking lot; a global path search module configured to generate a global path for the inside of the parking lot as a node map by searching for a free space, in which the vehicle is capable of being driven, based on the cognitive information and by setting a node in the free space; and an optimal local path generation module configured to generate an optimal local path from a current location of the vehicle to a destination by connecting the node set in the free space while the vehicle moves inside the parking lot; wherein the global path search module comprises: a cell setting device configured to determine whether the obstacle is recognized for each cell by dividing the space around the vehicle detected by the cognitive sensor into a plurality of cells; a free space search device configured to distinguish between a parking space, in which the obstacle is detected, and a free space, in which the obstacle is not detected, based on the cognitive information, to search for a driving free space, which has a width at which the vehicle is capable of being driven, in the free space, and to select the driving free space as a part of the global path; and a high-priority (HP) node setting device configured to generate a HP node at an end of the driving free space, and to set the HP node for the driving free space on a parking lot map indicating the inside of the parking lot so as to be stored; wherein the HP node setting device is configured to generate an HP connection line connecting a center of the vehicle to an HP node set in the driving free space so as to be stored; and wherein the optimal local path generation module comprises: a priority update device configured to lower a priority of the HP node, through which the vehicle passes while being driven inside the parking lot, in inverse proportion to elapsed time; and a local path generation device configured to generate a path from the current location of the vehicle to the destination based on a potential field and to generate a path with a high priority of the HP node through which the vehicle passes to reach the destination as a local path. ### Claim 2 (depends on 1) The system of claim 1, wherein the global path search module comprises a principal component analysis line generation device configured to: generate a principal component indicating a distribution status of a covariance calculation region set depending on a detection location of the obstacle; and generate a principal component analysis line indicating directivity of the covariance calculation region having a probability of detecting the obstacle having a specific level or more. ### Claim 3 (depends on 2) The system of claim 2, wherein the free space search device is configured to enhance a repulsive force potential field recognized in a space with the obstacle by generating a principal component analysis force potential field having a specific radius at a principal component analysis start point location when two or more free spaces having a travelable width are recognized. ### Claim 4 (depends on 2) The system of claim 2, wherein the global path search module further comprises a low-priority (LP) connection line generation device configured to generate an LP connection line parallel to the principal component analysis line with respect to a space after the HP node. ### Claim 5 (depends on 4) The system of claim 4, wherein the LP connection line generation device is configured to generate the LP connection line so as to be parallel to an average of angles formed between two principal component analysis lines when the two principal component analysis lines are generated on both sides of a front of the vehicle. ### Claim 6 (depends on 4) The system of claim 4, wherein the global path search module further comprises an LP node setting device configured to set a point, at which the LP connection line contacts another connection line or another node, to an LP node. ### Claim 7 (depends on 1) The system of claim 1, wherein the optimal local path generation module further comprises a path generation condition determination device configured to cause a new local path to be generated when it is determined that all HP nodes are connected to two or more HP connection lines by monitoring a connection line connected to the HP node. ### Claim 8 (depends on 1) The system of claim 1, wherein the optimal local path generation module is configured to cause a new path passing through one or more LP nodes to be generated as an optimal path when a start node is capable of being reached by using a new path via the LP node other than a path on which the vehicle is moved while being driven. ### Claim 9 (depends on 1) The system of claim 1, wherein the priority update device is configured to perform autonomous parking by setting the parking space to a highest priority node when the parking space that is empty is recognized by the cognitive sensor in driving inside the parking lot. ### Claim 10 A method of searching for an autonomous parking route based on a cognitive sensor, the method comprising: obtaining cognitive information for determining an existence and a location of an obstacle by sensing a space around a vehicle being driven in an inside of a parking lot; generating a global path for the inside of the parking lot as a node map by searching for a free space, in which the vehicle is capable of being driven, based on the cognitive information and by setting a node in the free space; and generating an optimal local path from a current location of the vehicle to a destination by connecting the node set in the free space while the vehicle moves inside the parking lot; wherein generating the global path comprises: dividing the space around the vehicle; distinguishing between a parking space, in which the obstacle is detected, and a free space, in which the obstacle is not detected, based on the cognitive information, searching for a driving free space, which has a width at which the vehicle is capable of being driven, in the free space, and selecting the driving free space as a part of the global path; and generating a high-priority (HP) node at an end of the driving free space, and setting the HP node for the driving free space on a parking lot map indicating the inside of the parking lot so as to be stored; wherein generating the global path further comprises generating a low-priority (LP) connection line parallel to a principal component analysis line with respect to a space after the HP node; and wherein generating the optimal local path comprises: lowering a priority of the HP node, through which the vehicle passes while being driven inside the parking lot, in inverse proportion to elapsed time; and generating a path from the current location of the vehicle to the destination based on a potential field and generating a path with a high priority of the HP node through which the vehicle passes to reach the destination as a local path. ### Claim 11 (depends on 10) The method of claim 10, wherein searching for the driving free space comprises enhancing a repulsive force potential field recognized in a space with the obstacle by generating a principal component analysis force potential field having a specific radius at a principal component analysis start point location when two or more free spaces having a travelable width are recognized. ### Claim 12 (depends on 10) The method of claim 10, wherein generating the global path further comprises: generating a principal component indicating a distribution status of a covariance calculation region set depending on a detection location of the obstacle; and generating the principal component analysis line indicating directivity of the covariance calculation region having a probability of detecting the obstacle having a specific level or more. ### Claim 13 (depends on 10) The method of claim 10, wherein generating the optimal local path further comprises causing a new local path to be generated when it is determined that all HP nodes are connected to two or more HP connection lines by monitoring a connection line connected to the HP node. ### Claim 14 A cognitive sensor-based autonomous parking route search system, the system comprising: a cognitive sensor configured to obtain cognitive information for determining an existence and a location of an obstacle by sensing a space around a vehicle being driven in an inside of a parking lot; a global path search module configured to generate a global path for the inside of the parking lot as a node map by searching for a free space, in which the vehicle is capable of being driven, based on the cognitive information and by setting a node in the free space; and an optimal local path generation module configured to generate an optimal local path from a current location of the vehicle to a destination by connecting the node set in the free space while the vehicle moves inside the parking lot; wherein the global path search module comprises: a cell setting device configured to determine whether the obstacle is recognized for each cell by dividing the space around the vehicle detected by the cognitive sensor into a plurality of cells; a free space search device configured to distinguish between a parking space, in which the obstacle is detected, and a free space, in which the obstacle is not detected, based on the cognitive information, to search for a driving free space, which has a width at which the vehicle is capable of being driven, in the free space, and to select the driving free space as a part of the global path; and a high-priority (HP) node setting device configured to generate a HP node at an end of the driving free space, and to set the HP node for the driving free space on a parking lot map indicating the inside of the parking lot so as to be stored; wherein the global path search module further comprises: a principal component analysis line generation device configured to generate a principal component indicating a distribution status of a covariance calculation region set depending on a detection location of the obstacle and to generate a principal component analysis line indicating directivity of the covariance calculation region having a probability of detecting the obstacle having a specific level or more; a low-priority (LP) connection line generation device configured to generate an LP connection line parallel to the principal component analysis line with respect to a space after the HP node; an LP node setting device configured to set a point, at which the LP connection line contacts another connection line or another node, to an LP node; and wherein the optimal local path generation module comprises: a priority update device configured to lower a priority of the HP node, through which the vehicle passes while being driven inside the parking lot, in inverse proportion to elapsed time; and a local path generation device configured to generate a path from the current location of the vehicle to the destination based on a potential field and to generate a path with a high priority of the HP node through which the vehicle passes to reach the destination as a local path. ### Claim 15 (depends on 14) The system of claim 14, wherein the HP node setting device is configured to generate an HP connection line connecting a center of the vehicle to an HP node set in the driving free space so as to be stored. ### Claim 16 (depends on 14) The system of claim 14, wherein the free space search device is configured to enhance a repulsive force potential field recognized in a space with the obstacle by generating a principal component analysis force potential field having a specific radius at a principal component analysis start point location when two or more free spaces having a travelable width are recognized. ### Claim 17 (depends on 14) The system of claim 14, wherein the LP connection line generation device is configured to generate the LP connection line so as to be parallel to an average of angles formed between two principal component analysis lines when the two principal component analysis lines are generated on both sides of a front of the vehicle. ### Claim 18 (depends on 14) The system of claim 14, wherein the optimal local path generation module further comprises a path generation condition determination device configured to cause a new local path to be generated when it is determined that all HP nodes are connected to two or more HP connection lines by monitoring a connection line connected to the HP node. ### Claim 19 (depends on 14) The system of claim 14, wherein the optimal local path generation module is configured to cause a new path passing through one or more LP nodes to be generated as an optimal path when a start node is capable of being reached by using a new path via the LP node other than a path on which the vehicle is moved while being driven. ### Claim 20 (depends on 14) The system of claim 14, wherein the priority update device is configured to perform autonomous parking by setting the parking space to a highest priority node when the parking space that is empty is recognized by the cognitive sensor in driving inside the parking lot.