KR102553697B1 - Method and device for searching routes in units of lanes in consideration of the operational design domain of an autonomous vehicle - Google Patents

Abstract

A method for searching a route by lane in consideration of an Operational Design Domain (ODD) of an autonomous vehicle, in a state in which global route information is being transmitted from a control server to the autonomous vehicle, (a) a specific When the global route information is transmitted, the route search device refers to the driving design area and the stored precise map information, and (i) includes information that does not satisfy the driving condition of the autonomous vehicle in the specific global route information. A first process for determining whether or not a lane change distance is included, (ii) determining whether a specific route distance value included in the specific global route information is greater than or equal to a lane change distance value, which is a distance required for the autonomous vehicle to change lanes. performing at least a part of a second process of performing a second process, and (iii) a third process of determining whether or not landmarks necessary for precise positioning are included in the specific global route information on a driving route along which the autonomous vehicle should travel. doing; (b) As a result of at least some of the processes performed, it is determined that the result of the first process is that there is no information that does not satisfy the driving condition of the autonomous vehicle in the specific global path information, and the second process The result of is determined that the specific route distance value included in the specific global route information is greater than the lane change distance value, which is a distance required for the autonomous vehicle to change lanes, and the result of the third process is the When it is determined that the landmark required for precise positioning is included in the specific global route information on the driving route along which the self-driving vehicle should travel, the route search device refers to the specific global route information searching for a plurality of regional routes located within a predetermined driving distance from the autonomous vehicle, and calculating a reliability score for each of the plurality of regional routes; and (c) the path search device compares each reliability score with a preset reliability threshold, determines whether each reliability score is equal to or greater than the preset reliability threshold, and determines whether or not each reliability score is greater than or equal to the preset reliability threshold. Selecting a specific regional route having a specific reliability score, which is the highest reliability score among reliability scores, as an optimal driving route for the autonomous vehicle, thereby driving or supporting the autonomous vehicle to operate along the specific regional route; including Disclosed are a route search method and a route search device characterized in that

Description

Method and route search device for route search in units of lanes in consideration of driving design area of autonomous vehicles

It is about a method and a route search device for searching a route in lane units considering the Operational Design Domain (ODD) of an autonomous vehicle.

Existing optimal path search technology uses a cost function such as Dijkstra or A* to calculate the cost for each path and then selects the path with the lowest cost.

However, the cost function required for route search is generally designed so that the route that arrives at the destination in the least time/shortest distance is selected by using information such as distance information, arrival time information, traffic information, and road information. The optimal route thus derived is not a problem if the driver directly drives or if the autonomous vehicle has a level 4 or higher technology level that allows autonomous driving without restrictions on the route. Areas where autonomous driving is not possible, such as senior protection areas, are included in the route, and routes that deviate from ODD (Operational Design Domain), such as unprotected left turns, are included in the driving route of autonomous vehicles, and structures such as buildings/guardrails When it is difficult to secure the field of view due to the traffic, when the lane change fails due to a short lane change or high traffic volume, or in a road environment where it is difficult to recognize the location of an autonomous vehicle, the self-driving mode of the autonomous vehicle is activated. There is a problem with the possibility of cancellation.

Therefore, there is a need for a technique for searching for and determining the safest route with a low possibility of the self-driving mode being terminated while the currently following route of the autonomous vehicle reaches the destination.

The present invention aims to solve all of the above problems.

Another object of the present invention is to make it possible to determine whether or not the global path information includes information that may cause the autonomous driving mode to be terminated.

In addition, according to the present invention, if the global route information does not include information that may cause the autonomous driving mode to be terminated, local routes included in the global route information are searched for, and reliability scores for each local route are calculated. that serves another purpose.

In addition, the present invention compares each reliability score with a preset reliability threshold and selects a specific local path having the highest reliability score among the reliability scores that are equal to or greater than the preset reliability threshold as an optimal driving route for an autonomous vehicle. Another object of the present invention is to enable a vehicle to be driven along an optimal driving path.

In addition, the present invention is to re-receive other global route information when the global route information includes information that may cause the autonomous driving mode to be terminated or when all reliability scores are less than a preset reliability threshold value. for a different purpose.

The characteristic configuration of the present invention for achieving the object of the present invention as described above and realizing the characteristic effects of the present invention described later is as follows.

According to one aspect of the present invention, in a method for searching a route in lane units in consideration of an Operational Design Domain (ODD) of an autonomous vehicle, global route information is transmitted from a control server to the autonomous vehicle In this state, (a) when specific global route information is transmitted, the route search device refers to the driving design area and stored precision map information, and (i) adds the specific global route information to the autonomous vehicle. A first process for determining whether information that does not satisfy driving conditions is included, (ii) a lane change distance where a specific route distance value included in the specific global route information is a distance required for the autonomous vehicle to change lanes. A second process for determining whether the value is greater than or equal to a value, and (iii) determining whether the specific global route information includes landmarks necessary for precise positioning on the driving route along which the autonomous vehicle should travel. performing at least part of the third process; (b) As a result of at least some of the processes performed, it is determined that the result of the first process is that there is no information that does not satisfy the driving condition of the autonomous vehicle in the specific global path information, and the second process The result of is determined that the specific route distance value included in the specific global route information is greater than the lane change distance value, which is a distance required for the autonomous vehicle to change lanes, and the result of the third process is the When it is determined that the landmark required for precise positioning is included in the specific global route information on the driving route along which the self-driving vehicle should travel, the route search device refers to the specific global route information searching for a plurality of regional routes located within a predetermined driving distance from the autonomous vehicle, and calculating a reliability score for each of the plurality of regional routes; and (c) the path search device compares each reliability score with a preset reliability threshold, determines whether each reliability score is equal to or greater than the preset reliability threshold, and determines whether or not each reliability score is greater than or equal to the preset reliability threshold. Selecting a specific regional route having a specific reliability score, which is the highest reliability score among reliability scores, as an optimal driving route for the autonomous vehicle, thereby driving or supporting the autonomous vehicle to operate along the specific regional route; including A path search method characterized by doing is disclosed.

As an example, in the first process, the route search device determines that the specific global route information includes (i) an area where autonomous driving of the autonomous vehicle is impossible, (ii) autonomous driving of the autonomous vehicle. If at least one of the cases in which a route outside the driving operating conditions is included and (iii) a route traveling in an area not included in the precise map information is included, the specific global route information is included in the driving condition Disclosed is a path search method characterized in that it is determined that information not satisfying is included.

As an example, in the second process, the specific path distance value is determined by a first navigation link, which is a navigation link of an area in which the autonomous vehicle is currently driving, among navigation links included in the specific global route information, and the first navigation link. The lane change distance value is determined according to a geometric relationship, which is a connected form between a link and a second navigation link that is a navigation link in an area in which the autonomous vehicle is scheduled to travel, and the lane change distance value is calculated for the autonomous vehicle to change lanes. A route search method is disclosed, which is determined by the sum of a required distance value and a safety distance value for lane change of the autonomous vehicle.

은 하기 수식에 따라 계산되고,As an example, the required distance value

is calculated according to the following formula,

는 상수로서 차선변경에 소요되는 최대시간을 의미하고,

는 차선변경 속도를 의미하며,

은 변경해야 하는 차선의 수를 의미하는 것을 특징으로 하고,in the above formula

is a constant and means the maximum time required for lane change,

is the lane change speed,

is characterized in that it means the number of lanes to be changed,

는 하기 수식에 따라 계산되고,The lane change speed

is calculated according to the following formula,

는 상기 자율주행차량이 위치하는 도로구간의 제한 속도를 의미하고,

는 상수로서 차선변경을 고려한 안전마진 속도를 의미하는 것을 특징으로 하는 경로 탐색 방법이 개시된다.in the above formula

Means the speed limit of the road section where the autonomous vehicle is located,

is a constant, and denotes a safety margin speed considering a lane change.

As an example, in the third process, the route search device includes (i) a distance from a predetermined line connecting the driving route among a plurality of landmarks included in the precision map information included in the autonomous vehicle. Search for at least one specific landmark that exists within the maximum recognition distance of each sensor and is included in the vertical angle range and horizontal angle range of the field of view (FOV) for each sensor (ii) calculates the use score of each of the specific landmarks by applying each of the importance weights preset for each landmark to each of the specific landmarks, and obtains a total score obtained by summing all the use scores Compared with a preset precise positioning threshold, (iii) if the total score is equal to or greater than the preset precise positioning threshold, it is determined that a landmark required for the precise positioning is included in the specific global route information, and the total score is Disclosed is a path search method characterized in that it is determined that a landmark required for the precise positioning is not included in the specific global path information if the predetermined precise positioning threshold is smaller than the preset precise positioning threshold.

As an example, in the step (b), the route search device includes a plurality of precision map links included in the specific global route information, wherein the precision map link means the center line of a road, located within the predetermined travel distance. The plurality of local routes, which are predetermined paths in which at least some of the specific precision map links are connected to each other, by searching for specific precision map links of and assigning a unique ID to each of the specific precision map links using a predetermined algorithm. A path search method characterized by searching is disclosed.

As an example, the route search device calculates the reliability score of each of the local routes according to the following formula,

는 전역경로 추종 점수를 의미하고,

는 전역경로 추종에 필요한 차선변경 점수를 의미하며,

는

의 중요도(Weight)를 의미하고,

는

의 중요도(Weight)를 의미하며,

는 상기 지역경로 각각의 인덱스를 의미하는 것을 특징으로 하는 경로 탐색 방법이 개시된다.In the above formula,

denotes the global path following score,

denotes the lane change score required to follow the global route,

Is

means the weight of

Is

means the weight of

Disclosed is a route search method characterized in that denotes an index of each of the local routes.

As an example, the route search device, for each regional route, sets the number of specific first precision map links that are specific precision map links included in the corresponding regional route to the specific global route among the specific first precision map links. The global path following score is obtained by multiplying a value divided by the number of second specific precision map links that are corresponding specific precision map links by a predetermined K value, which can be set within the range of 0 to 1. Disclosed is a path search method characterized by calculating.

As an example, the route search device determines (i) when the specific first precision map link is the same as the specific second precision map link, and (ii) a global path at a prior time point that is a time point before a predetermined time from the current time point. If at least one of the cases in which the dictionary precision map link, which is a precision map link corresponding to the dictionary global path, and the specific first precision map link are connected, the predetermined K value is set to 1, and the specific first precision map link is connected. Even if the link is the same as the second specific precision map link or if the prior precision map link and the specific first precision map link are connected, if an event occurs in the corresponding local path, the predetermined K value A route search method characterized by setting to a value smaller than 1 is disclosed.

를 계산하고,As an example, the route search device calculates the lane change score according to the following formula.

calculate,

는 상수로서 정밀지도 링크간 횡방향 거리를 의미하고,

는 변경하려는 타겟 정밀지도 링크까지의 차선변경의 수를 의미하며,

는 교통상황을 고려한 복잡도 계수를 의미하는 것을 특징으로 하되, 상기 복수의 지역경로 중 차선변경이 필요하지 않은 지역경로의 경우에는 상기 차선변경 점수가 1로 설정되는 것을 특징으로 하는 경로 탐색 방법이 개시된다.In the above formula,

is a constant, which means the lateral distance between precision map links,

Means the number of lane changes to the target precision map link to be changed,

denotes a complexity coefficient considering traffic conditions, but in the case of a regional route that does not require a lane change among the plurality of regional routes, the lane change score is set to 1. do.

As an example, in the step (b), the route search device, as a result of at least some of the processes performed, (i) the result of the first process is assigned to the specific global route information for the operation of the autonomous vehicle. If it is determined that information that does not satisfy the condition is included, (ii) the result of the second process is that the specific route distance value included in the specific global route information is necessary for the autonomous vehicle to change lanes. If it is determined to be a value smaller than the lane change distance value, which is the distance, and (iii) the result of the third process indicates that the landmark required for precise positioning in the specific global route information is the distance the autonomous vehicle must drive. If it falls under one of the cases determined not to be included in the driving route, a request is made to the control server to transmit global route information different from the specific global route information, and if the other global route information is transmitted, the global route information is transmitted. A route search method is disclosed wherein the route search apparatus performs at least a part of the first process, the second process, and the third process with respect to the other global route information.

As an example, in the step (c), the path search apparatus, as a result of comparing each of the reliability scores with the preset reliability threshold, when all of the plurality of reliability scores are smaller than the preset reliability threshold, the A route search method characterized by requesting transmission of global route information different from the specific global route information to a control server and re-receiving the other global route information is disclosed.

According to another aspect of the present invention, in a route search device for searching a route in lane units in consideration of an Operational Design Domain (ODD) of an autonomous vehicle, global route information is transmitted from a control server to the autonomous vehicle. at least one memory to store instructions in the state being transmitted; and at least one processor configured to execute the instructions, wherein the processor: (1) when specific global path information is transmitted, with reference to the driving design area and stored precision map information, (i ) a first process of determining whether the specific global route information includes information that does not satisfy the driving condition of the autonomous vehicle, (ii) a specific route distance value included in the specific global route information A second process for determining whether the value is greater than or equal to the lane change distance value, which is the distance required for the vehicle to change lanes, and (iii) the specific global path information must include a landmark required for precise positioning when the autonomous vehicle is traveling. A process of performing at least a part of a third process for determining whether a driving path is included in a traveling route, and (2) as a result of the at least part of the process, the result of the first process is associated with the specific global path information. It is determined that there is no information that does not satisfy the driving condition of the self-driving vehicle, and the result of the second process is that the value of the specific route distance included in the specific global route information causes the autonomous vehicle to change lanes. It is judged to be a value greater than the lane change distance value, which is a required distance, and the result of the third process is that the landmark required for precise positioning in the specific global route information is on the driving route along which the autonomous vehicle should travel. If it is determined to be included, searching for a plurality of local routes located within a predetermined driving distance from the autonomous vehicle with reference to the specific global route information, and calculating a reliability score for each of the plurality of local routes and (3) comparing each of the reliability scores with a preset reliability threshold to determine whether each reliability score is equal to or greater than the preset reliability threshold, and determining the highest reliability score among predetermined reliability scores equal to or greater than the preset reliability threshold. By selecting a specific regional route having a specific reliability score, which is a high reliability score, as an optimal driving route for the autonomous vehicle, a process for driving or supporting the autonomous vehicle to operate along the specific regional route Characterized in that A route search device is initiated.

As an example, the processor may, in the first process, determine, in the specific global path information, (i) a region where autonomous driving of the autonomous vehicle is impossible is included, (ii) an autonomous driving operation of the autonomous vehicle If at least one of the cases in which a route out of the condition is included and (iii) a route traveling in an area not included in the detailed map information is included, the specific global route information satisfies the driving condition. Disclosed is a path search device characterized in that it is determined that information that is not included is included.

As an example, in the second process, the specific path distance value is determined by a first navigation link, which is a navigation link of an area in which the autonomous vehicle is currently driving, among navigation links included in the specific global route information, and the first navigation link. The lane change distance value is determined according to a geometric relationship, which is a connected form between a link and a second navigation link that is a navigation link in an area in which the autonomous vehicle is scheduled to travel, and the lane change distance value is calculated for the autonomous vehicle to change lanes. A route search device is disclosed, which is determined by the sum of a required distance value and a safety distance value for lane change of the self-driving vehicle.

은 하기 수식에 따라 계산되고,As an example, the required distance value

is calculated according to the following formula,

는 상수로서 차선변경에 소요되는 최대시간을 의미하고,

는 차선변경 속도를 의미하며,

은 변경해야 하는 차선의 수를 의미하는 것을 특징으로 하고,in the above formula

is a constant and means the maximum time required for lane change,

is the lane change speed,

is characterized in that it means the number of lanes to be changed,

는 하기 수식에 따라 계산되고,The lane change speed

is calculated according to the following formula,

는 상기 자율주행차량이 위치하는 도로구간의 제한 속도를 의미하고,

는 상수로서 차선변경을 고려한 안전마진 속도를 의미하는 것을 특징으로 하는 경로탐색장치가 개시된다.in the above formula

Means the speed limit of the road section where the autonomous vehicle is located,

is a constant, and denotes a safety margin speed considering a lane change.

As an example, the processor, in the third process, (i) a distance from a predetermined line connecting the driving route among a plurality of landmarks included in the precision map information is included in the autonomous vehicle Searching for and storing at least one specific landmark that exists within the maximum recognition distance of each sensor and is included in the vertical angle range and horizontal angle range of the field of view (FOV) for each sensor, and , (ii) calculating the use score of each of the specific landmarks by applying each of the predetermined importance weights for each corresponding landmark to each of the specific landmarks, and calculating the total score by summing all the use scores of each of the Compared with the precision positioning threshold, (iii) if the total score is equal to or greater than the preset precision positioning threshold, it is determined that a landmark required for the precise positioning is included in the specific global route information, and the total score is equal to or greater than the preset precision positioning threshold. A route search apparatus is disclosed which determines that a landmark necessary for the precise positioning is not included in the specific global route information when the precision positioning threshold is smaller than the precise positioning threshold.

As an example, in the process (2), the processor may, in the process (2), select a plurality of specific links included in the specific global route information, wherein the precise map link means a center line of a road, located within the predetermined mileage. Searching for precision map links, and assigning a unique ID to each of the specific precision map links using a predetermined algorithm to search for the plurality of local routes, which are predetermined paths in which at least some of the specific precision map links are connected to each other Disclosed is a path search device characterized in that

As an example, the processor calculates the reliability score of each of the local paths according to the following formula,

는 전역경로 추종 점수를 의미하고,

는 전역경로 추종에 필요한 차선변경 점수를 의미하며,

는

의 중요도(Weight)를 의미하고,

는

의 중요도(Weight)를 의미하며,

는 상기 지역경로 각각의 인덱스를 의미하는 것을 특징으로 하는 경로탐색장치가 개시된다.In the above formula,

denotes the global path following score,

denotes the lane change score required to follow the global route,

Is

means the weight of

Is

means the weight of

denotes an index of each of the local routes.

As an example, the processor determines the number of specific first precision map links, which are specific precision map links included in the corresponding local path, for each regional path, corresponding to the specific global path among the specific first precision map links. Calculating the global path following score by multiplying a value divided by the number of specific second precision map links, which are specific precision map links, by a predetermined K value, which may be set within the range of 0 to 1 Disclosed is a path search device characterized in that

As an example, the processor determines (i) when the specific first precision map link is the same as the specific second precision map link, and (ii) a dictionary that is a global path at a prior time point that is a time point before a predetermined time from the current time point. If at least one of the cases in which the pre-precision map link, which is a precision map link corresponding to the global path, and the first specific precision map link are connected, the predetermined K value is set to 1, and the specific first precision map link is The predetermined K value is set to 1 even when an event occurs in the corresponding local path even if it is the same as the specific second precision map link or if the prior precision map link and the specific first precision map link are connected. A path search device characterized in that the value is set to a smaller value is disclosed.

를 계산하고,As an example, the processor calculates the lane change score according to the following formula.

calculate,

는 상수로서 정밀지도 링크간 횡방향 거리를 의미하고,

는 변경하려는 타겟 정밀지도 링크까지의 차선변경의 수를 의미하며,

는 교통상황을 고려한 복잡도 계수를 의미하는 것을 특징으로 하되, 상기 복수의 지역경로 중 차선변경이 필요하지 않은 지역경로의 경우에는 상기 차선변경 점수가 1로 설정되는 것을 특징으로 하는 경로탐색장치가 개시된다.In the above formula,

is a constant, which means the lateral distance between precision map links,

Means the number of lane changes to the target precision map link to be changed,

denotes a complexity coefficient considering traffic conditions, but in the case of a regional route that does not require a lane change among the plurality of regional routes, the lane change score is set to 1. do.

As an example, the processor, in the (2) process, as a result of at least some of the processes performed, (i) the result of the first process sets the driving condition of the autonomous vehicle to the specific global path information. If it is determined that unsatisfactory information is included, (ii) the result of the second process is that the specific route distance value included in the specific global route information is a distance required for the autonomous vehicle to change lanes. When it is determined to be a value smaller than the lane change distance value, and (iii) the result of the third process is that the landmark required for precise positioning in the specific global route information is the driving route along which the autonomous vehicle should travel. If it falls under one of the cases determined not to be included in the above, a request is made to the control server to transmit global path information different from the specific global path information, and if the other global path information is transmitted, the route search The device performs at least a part of the first process, the second process, and the third process with respect to the other global path information.

As an example, the processor, in the process (3), when all of the plurality of reliability scores are smaller than the preset reliability threshold as a result of comparing each of the reliability scores with the preset reliability threshold, the control server A route search device characterized in that the specific global route information and other global route information is requested to be transmitted and the other global route information is re-received.

According to the present invention, there are the following effects.

According to the present invention, it is possible to determine whether or not the global path information includes information that may cause the autonomous driving mode to be terminated.

In addition, in the present invention, if the global route information does not include information that may cause the autonomous driving mode to be terminated, local routes included in the global route information may be searched and a reliability score for each local route may be calculated. has the effect of allowing

In addition, the present invention compares each reliability score with a preset reliability threshold and selects a specific local path having the highest reliability score among the reliability scores that are equal to or greater than the preset reliability threshold as an optimal driving route for an autonomous vehicle. There is an effect of allowing the driving vehicle to be operated along an optimal driving route.

In addition, the present invention has an effect of allowing other global route information to be re-received when the global route information includes information that may cause the autonomous driving mode to be terminated, or when all reliability scores are smaller than a preset reliability threshold. there is.

1 is a diagram showing a schematic configuration of a path search device for searching a path in units of lanes in consideration of a driving design area of an autonomous vehicle according to an embodiment of the present invention.
2 is a diagram for explaining a schematic sequence of a method of searching for a path in units of lanes in consideration of a driving design area of an autonomous vehicle according to an embodiment of the present invention.
3A and 3B are diagrams for determining whether a specific route distance value included in specific global route information is greater than a lane change distance value, which is a distance required for an autonomous vehicle to change lanes, according to an embodiment of the present invention. It is a figure for explaining an example of the 2nd process.
4 illustrates an example of a third process for determining whether landmarks required for precise positioning are included in specific global route information on a driving route along which an autonomous vehicle should travel according to an embodiment of the present invention. It is a drawing for
5 is a diagram for explaining an example of searching for a plurality of local routes located within a predetermined driving distance from an autonomous vehicle with reference to specific global route information according to an embodiment of the present invention.
6 is a diagram for explaining an example of calculating a global path following score according to an embodiment of the present invention.
7 is a diagram for explaining an example of calculating a lane change score according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The detailed description of the present invention which follows refers to the accompanying drawings which illustrate, by way of illustration, specific embodiments in which the present invention may be practiced. These embodiments are described in sufficient detail to enable one skilled in the art to practice the present invention. It should be understood that the various embodiments of the present invention are different from each other but are not necessarily mutually exclusive. For example, specific shapes, structures, and characteristics described herein may be implemented in one embodiment in another embodiment without departing from the spirit and scope of the invention. Additionally, it should be understood that the location or arrangement of individual components within each disclosed embodiment may be changed without departing from the spirit and scope of the invention. Accordingly, the detailed description set forth below is not to be taken in a limiting sense, and the scope of the present invention, if properly described, is limited only by the appended claims, along with all equivalents as claimed by those claims. Like reference numbers in the drawings indicate the same or similar function throughout the various aspects.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily practice the present invention.

1 is a diagram showing a schematic configuration of a path search device 100 that searches a path in units of lanes in consideration of an Operational Design Domain (ODD) of an autonomous vehicle according to an embodiment of the present invention.

As shown in FIG. 1, the route search device 100 of the present invention may include a memory 110 and a processor 120.

The memory 110 of the path search device 100 may store instructions of the processor 120. Specifically, the instructions are codes generated for the purpose of causing the path search device 100 to function in a specific way. , may be stored in a computer usable or computer readable memory that may be directed to a computer or other programmable data processing equipment. Instructions may perform processes for executing functions described in the specification of the present invention.

In addition, the processor 120 of the path search device 100 may include a hardware configuration such as a micro processing unit (MPU) or a central processing unit (CPU), a cache memory, and a data bus. . In addition, it may further include a software configuration of an operating system and an application that performs a specific purpose.

In addition, the route search device 100 may be interlocked with a database 900 including information used to search a route in lane units in consideration of a driving design area of an autonomous vehicle. Here, the database 900 is a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg SD or XD memory), At least one of RAM (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (ReadOnly Memory, ROM), EEPROM (Electrically Erasable Programmable ReadOnly Memory), PROM (Programmable ReadOnly Memory), magnetic memory, magnetic disk, and optical disk It may include one type of storage medium, but is not limited thereto and may include any medium capable of storing data. In addition, the database 900 may be installed separately from the path search device 100, or may be installed inside the path search device 100 to transmit data or record received data. The above may be implemented separately, which may vary depending on the implementation conditions of the invention. In addition, the database 900 may store precision map information, and the precision map information may be updated at regular intervals or according to the needs of users/managers.

In addition, the route search device 100 may transmit and receive necessary information with a control server (not shown) through a communication unit (not shown). Here, the route search device 100 may receive global route information considering traffic information, shortest distance, minimum time, etc. from the control server. At this time, the route search device 100 directly receives global route information from the control server. It may be transmitted or received, or the global route information may be transmitted and received indirectly through a global route receiving unit that receives global route information from the control server. Therefore, in the following description, the description that the route search device 100 receives the global route information from the control server may mean that the global route information can be directly transmitted and received from the control server, and the global route receiver It may mean that global path information can be transmitted indirectly through . For reference, the same will be true even when the route search device 100 transmits information to the control server.

In addition, the route search device 100 may transmit and receive necessary information with a location receiver (not shown) for receiving location information such as GPS information and UTM information for querying precision map information through a communication unit (not shown).

In addition, the path search device 100 can transmit and receive necessary information with a sensor data receiver (not shown) that receives information on surrounding objects acquired from sensors such as lidar, camera, and radar through a communication unit (not shown). .

In addition, the route search device 100 may transmit/receive necessary information with a traffic information receiving unit (not shown) that receives traffic information and traffic information of a current driving route through a communication unit (not shown).

In addition, the route search device 100 may transmit/receive necessary information with a route display unit (not shown) displaying global routes and local routes on a screen through a communication unit (not shown).

In addition, the route search device 100 may establish a route plan for an autonomous vehicle and transmit/receive necessary information with a control unit (not shown) that controls the vehicle.

For reference, the path search device 100 may be located and installed inside the self-driving vehicle, or may be installed outside the self-driving vehicle to transmit and receive necessary information to and from the self-driving vehicle.

A method of searching for a route by lane in consideration of a driving design area of an autonomous vehicle using the route search device 100 according to an embodiment of the present invention configured as described above will be described with reference to FIG. 2 as follows.

2 is a diagram for explaining a schematic sequence of a method of searching for a path in units of lanes in consideration of a driving design area of an autonomous vehicle according to an embodiment of the present invention.

First, the path search apparatus 100 may transmit and receive specific global path information from the control server (S210).

Here, the specific global route information is at least a part of the global route information transmitted from the control server to the route search device 100, and may be information on a global route corresponding to a driving route on which the self-driving vehicle is currently running. It will not be limited. At this time, the control server may transmit the global route information to the route search device 100 at regular intervals, or may transmit the route search device 100 only in specific situations, but is not limited thereto.

In addition, the route search device 100 determines whether information that does not satisfy the driving condition of the autonomous vehicle is included in the specific global route information with reference to the driving design area of the autonomous vehicle and the stored precision map information. A first process for determining, a second process for determining whether a specific route distance value included in the specific global route information is greater than a lane change distance value, which is a distance required for an autonomous vehicle to change lanes, and a specific global route At least a part of the third process of determining whether or not landmarks necessary for precise positioning are included in the information on a driving route along which the autonomous vehicle should travel may be performed (S220).

At this time, in the first process, the route search apparatus 100 determines, in the first process, when the specific global route information includes an area where autonomous driving of the autonomous vehicle is impossible, the specific global route information deviates from the autonomous driving operating conditions of the autonomous vehicle. If at least one of the case where a route is included and the specific global route information includes a route traveling in an area not included in the precise map information, the specific global route information does not satisfy the driving condition. It can be judged that information is included.

Here, the area in which autonomous driving is impossible may mean an area in which autonomous driving is not permitted, such as a children's protection area, an elderly protection area, or a school zone, but is not limited thereto. In addition, routes that deviate from autonomous driving operating conditions may refer to routes that are not technically supported depending on the technological level of autonomous vehicles, whether or not a driver is on board, and the demonstration environment of autonomous vehicles, such as unprotected left turns and roundabouts. It will not be limited to this.

In addition, in the second process, the route search apparatus 100 connects a first navigation link, which is a navigation link of an area in which an autonomous vehicle is currently driving, among navigation links included in specific global route information, and the first navigation link. and determine a specific path distance value according to the geometric relationship, which is a connected form between the second navigation link, which is the navigation link of the area where the autonomous vehicle is scheduled to drive, and the required distance value calculated for the autonomous vehicle to change lanes and autonomous driving The lane change distance value may be determined by the sum of the safety distance values for the lane change of the vehicle.

을 계산할 수 있다.At this time, the path search device 100 is the required distance value according to the following formula

can be calculated.

는 상수로서 차선변경에 소요되는 최대시간을 의미하고,

는 차선변경 속도를 의미하며,

은 변경해야 하는 차선의 수를 의미할 수 있다. 여기서, 차선변경 속도인

는 하기 수식에 따라 계산될 수 있다.In the above formula,

is a constant and means the maximum time required for lane change,

is the lane change speed,

may mean the number of lanes to be changed. Here, the lane change speed is

Can be calculated according to the formula below.

는 자율주행차량이 위치하는 도로구간의 제한 속도를 의미하고,

는 상수로서 차선변경을 고려한 안전마진 속도를 의미할 수 있다.In the above formula,

denotes the speed limit of the road section where the autonomous vehicle is located,

Is a constant and may mean a safety margin speed considering a lane change.

에는 경우에 따라 속도 또는 속력의 단위를 변환하기 위한 소정의 변환상수가 적용될 수 있을 것이다. 예를 들어,

의 단위가 kph(kilometer per hour)라면 이를 mps(meter per second)로 변환하기 위해

값을 변환상수인 3.6으로 나눔으로써

의 단위를 kph에서 mps로 변환할 수 있을 것이다.here,

In some cases, a predetermined conversion constant for converting a unit of speed or speed may be applied. for example,

If the unit of is kph (kilometer per hour), to convert it to mps (meter per second)

By dividing the value by the conversion constant of 3.6

You will be able to convert the units of kph to mps.

For example, referring to FIG. 3A, the geometric relationship in which the first navigation link 310 and the second navigation link 320 are connected form an angle to the left with respect to the driving direction of the autonomous vehicle. and, assuming that the autonomous vehicle is currently located at the starting point 330, the autonomous vehicle needs to make a left turn according to the geometric relationship, which is a connection form between the navigation links included in the specific global route information. must move from the starting point 330 to the first target point 340 in order to perform a left turn. Accordingly, the specific path distance value in this situation may be d ₁ (350). At this time, d ₁ (350) is 200m, the safety distance value for lane change is 20m, the maximum time required for lane change is 7 seconds, the speed limit of the road section where the autonomous vehicle is located is 50kph, and the safety distance is 50kph. Assuming that the marginal speed is 20kph and the number of lanes to be changed is 4, the required distance value will be calculated as approximately 233m by (7*((50-20)/3.6)*4), where Since the unit of the lane change speed, which is the difference between the speed limit of 50 kph and the safety margin speed of 20 kph, is kph, conversion constant 3.6 can be applied to convert it to mps unit. Also, since the safety distance value is 20 m, the lane change distance value may be calculated as approximately 253 m. Therefore, the route search device 100 may determine that the specific route distance value of 200 m is smaller than the calculated lane change distance value of 253 m, and determine the route as a route on which the autonomous vehicle cannot change lanes, thereby determining the control server You can request global path information other than specific global path information with .

As another example, referring to FIG. 3B, the geometric relationship in which the first navigation link 310 and the second navigation link 370 are connected is in the form of a straight line in the direction in which the autonomous vehicle travels. , Assuming that the self-driving vehicle is currently located at the starting point 330 and the self-driving vehicle is scheduled to drive to the second target point 380, the self-driving vehicle is While going straight according to the geometric relationship, which is a connected form, it is necessary to change lanes in order to move to the second target point 380. Therefore, the specific path distance value in this situation may be the sum of d ₁ (350) and d ₂ (360). At this time, d ₁ (350) is 200 m, d ₂ (360) is 200 m, the safety distance value for lane change is 20 m, the maximum time required for lane change is 7 seconds, and the road where the autonomous vehicle is located. Assuming that the speed limit of the section is 50kph, the safety margin speed is 20kph, and the number of lanes to be changed is 3, the required distance value is calculated as 175m by (7*((50-20)/3.6)*3) It will be. Here, since the unit of the lane change speed, which is the difference between the speed limit of 50 kph and the safety margin speed of 20 kph in the road section, is kph, conversion constant 3.6 can be applied to convert it to mps. Also, since the safety distance value is 20 m, the lane change distance value may be calculated as 195 m. Accordingly, the route search apparatus 100 may determine that the specific route distance value of 400m is greater than the calculated lane change distance value of 175m.

In addition, in the third process, the path search apparatus 100 determines that the maximum distance of each sensor included in the autonomous vehicle from a predetermined line connecting the driving path among a plurality of landmarks included in the precision map information. At least one specific landmark that exists within the recognition distance and is included in the vertical angle range and horizontal angle range of each sensor's FOV (Field of View) is searched for and stored, and each specific landmark is searched for and stored. The use score of each specific landmark is calculated by applying each predetermined importance weight for each corresponding landmark, and the total score obtained by summing all the use scores of each landmark can be compared with a preset precision positioning threshold. Here, as a result of comparing the total score with a preset precision positioning threshold, the path search apparatus 100 determines that the landmark required for precise positioning is included in specific global path information if the total score is equal to or greater than the preset precision positioning threshold. and if the total score is smaller than the predetermined precise positioning threshold, it may be determined that the landmark required for precise positioning is not included in the specific global path information.

Here, the landmark may be one of a road guide sign, a building, a guardrail, a sign, a lane, and a curb, but is not limited thereto.

For example, referring to FIG. 4, a curb (410: 410a, 410b, 410c, 410d), a building (420: 420a, 420b, 420c, 420d), a first lane (430: 430a, 430b, 430c, 430d, 430e, 430f, 430g, 430h) and the second lane (440: 440a, 440b, 440c, 440d, 440e, 440f, 440g, 440h) are included, and the driving path of the autonomous vehicle Assuming that a predetermined line connecting is the purple line 450 indicated in purple, the route search device 100 is included in the autonomous vehicle from the purple line 450 among the landmarks included in the precision map information One of the sensors located within the indigo triangle range 460, which is the field of view (FOV) range of the camera, which is one of the sensors available, or the lidar, which is another sensor included in the autonomous vehicle, can be sensed from the purple line 450. The curb (410b, 410c, 410d), the first lane (430c, 430d, 430e, 430f, 430g, 430h) and the second lane (440c, 440d, 440e, 440f, 440g, 440h) may be searched for and stored as specific landmarks. In addition, when the score for each specific landmark is set to 1, the use score is calculated for each specific landmark by multiplying the importance weights preset for each landmark as shown in <Table 1> below, and each calculated use score The total score can be obtained by summing all of them.

land mark importance weight curb One lane One sign 0.8 building, guardrail 0.6 Road guidance sign 0.4

Next, as a result of at least some of the processes performed above, the route search apparatus 100 determines that the result of the first process is that there is no information that does not satisfy the driving condition of the autonomous vehicle in the specific global route information, As a result of the second process, it is determined that the specific route distance value included in the specific global route information is greater than the lane change distance value, which is a distance required for the autonomous vehicle to change lanes, and the result of the third process is determined as a specific global route distance value. When it is determined that the route information includes landmarks necessary for precise positioning on the driving route on which the autonomous vehicle must travel, a plurality of points located within a predetermined driving distance from the autonomous vehicle with reference to specific global route information It is possible to search for local paths of , and calculate reliability scores for each of a plurality of local paths (S230).

At this time, the route search apparatus 100 searches for a plurality of specific precision map links located within the predetermined driving distance among the precision map links included in the specific global route information, and uses a predetermined algorithm to search for each specific precision map link. By assigning a unique ID to , it is possible to search for a plurality of local routes, which are predetermined routes in which at least some of the specific precision map links are connected to each other. Here, the precision map link may mean the center line of a road.

Here, the route search device 100 may search all local routes included in specific global route information by utilizing a graph algorithm such as DFS (Depth-First Search), BFS (Breadth-First Search), etc. with a predetermined algorithm. There will be, but it will not be limited to this.

For example, referring to FIG. 5, as a result of the route search apparatus 100 searching for a plurality of local routes, a unique ID of 1 to 21 may be assigned to each of a plurality of searched precision map links. In addition, the route search device 100 is a plurality of local routes, which are predetermined routes in which at least some of the specific precision map links are connected to each other, such as "1-6-10", "2-6-11-17", " 2-6-10", "3-7-12-18", "4-8-13-19", "4-8-13-21", "5-9-14-20" and "5- You will be able to explore 8 local routes like 9-14-15".

In addition, the route search apparatus 100 may calculate the reliability score of each local route according to the following formula.

는 전역경로 추종 점수를 의미하고,

는 전역경로 추종에 필요한 차선변경 점수를 의미하며,

는

의 중요도(Weight)를 의미하고,

는

의 중요도(Weight)를 의미하며,

는 상기 지역경로 각각의 인덱스를 의미할 수 있다. In the above formula,

denotes the global path following score,

denotes the lane change score required to follow the global route,

Is

means the weight of

Is

means the weight of

may mean an index of each of the local paths.

Here, the route search device 100 may use traffic information and sensor data of a route currently driven by an autonomous vehicle to calculate a reliability score.

At this time, the route search apparatus 100, for each regional route, sets the number of specific first precision map links that are specific precision map links included in the corresponding regional route to a specific global route among specific first precision map links. A global path following score may be calculated by multiplying a value divided by the number of specific second precision map links, which are specific precision map links, by a predetermined K value. Here, a predetermined K value may be set within a range of 0 to 1.

Here, the path search apparatus 100 corresponds to a case where a specific first precision map link is the same as a specific second precision map link, and a prior global path that is a global path at a prior time point, which is a time point before a predetermined time from the current time point. The predetermined K value may be set to 1 if at least one of the cases in which a pre-precision map link, which is a precision map link, and a specific first precision map link are connected. Here, the path search apparatus 100 corresponds to the case where the specific first precision map link is the same as the specific second precision map link or the case in which the prior precision map link and the specific first precision map link are connected. Even if an event occurs in the corresponding local path, a predetermined K value may be set to a value smaller than 1.

Here, the case where an event occurs in the following area route may include, but is not limited to, the case of an accident in the following area route, the case of construction work, the existence of illegal parking and stopping.

For example, referring to FIG. 6, a first path 610 and a second path 620 in which unique IDs from 1 to 21 are assigned to each precision map link and a specific global path is indicated by a purple line. Assuming that, in the case of the "4-8-13-21" route among the plurality of regional routes, the specific first precision map links included in the "4-8-13-21" route are "4" and "8". Since ", "13" and "21", the number will be 4, and the specific precision map link corresponding to the first path 610 or the second path 620, which is a specific global path among the specific first precision map links, is also " Since 4", "8", "13" and "21", the number of specific second precision map links will also be 4. Accordingly, in the case of the “4-8-13-21” path, the K value may be 1 because the specific first precision map link and the specific second precision map link correspond to the same case. For reference, the "5-9-14-15" route will also be the same as the "4-8-13-21" route. In addition, in the case of the "3-7-12-18" route among the plurality of regional routes, the specific first precision map links included in the "3-7-12-18" route are "3", "7", Since "12" and "18", the number will be 4, and among the specific first precision map links, the specific precision map link corresponding to the first path 610 or the second path 620, which is a specific global path, is "3". , "7" and "12", the number of specific second precision map links will also be three. On the other hand, if the self-driving vehicle is located at the first point 630 at a prior time point, which is a time point before a predetermined time from the current time point, a pre-precision map link corresponding to the pre-global route, which is the global route at that time, is "4", "8", "13" and "21", and the autonomous vehicle is located at the second point 640 at the present time, so that the specific first precision map link is "8", "13" and "21". ", the prior precision map links of "4", "8", "13", and "21" and the specific first precision map links of "8", "13", and "21" are connected to each other, so the K value is could be 1.

를 하기 수식에 따라 계산할 수 있다.In addition, the route search device 100 is a lane change score

can be calculated according to the following formula.

는 상수로서 정밀지도 링크간 횡방향 거리를 의미하고,

는 변경하려는 타겟 정밀지도 링크까지의 차선변경의 수를 의미하며,

는 교통상황을 고려한 복잡도 계수를 의미할 수 있다. 여기서, 복수의 지역경로 중 차선변경이 필요하지 않은 지역경로의 경우에는 차선변경 차선변경 점수인

을 1로 설정할 수 있다.In the above formula,

is a constant, which means the lateral distance between precision map links,

Means the number of lane changes to the target precision map link to be changed,

may mean a complexity coefficient considering traffic conditions. Here, in the case of a regional route that does not require a lane change among multiple regional routes, the lane change lane change score is

can be set to 1.

That is, the route search device 100, when the autonomous vehicle moves along a following regional route corresponding to a specific global route or when an event such as an accident occurs and it is necessary to change lanes or merge into a lane to avoid this, It will be possible for the self-driving vehicle to select another local route that is closest to the current driving local route.

For example, referring to FIG. 7, as a result of the route search device 100 searching for a plurality of regional routes, a plurality of regions in which a plurality of fine map links each assigned a unique ID from 1 to 13 exist. In a state where a route is searched and the following local route corresponding to a specific global route is the "1-8-9-11" route 710, if a lane change is required due to an event such as an accident on the following local route Assuming that, the route search device 100 selects the "10-12-13" route 720 closest to the "1-8-9-11" route 710 as the target area route, and changes lanes therefor. score can be calculated. At this time, the lateral distance between the precision map links may be d ₃ (730), and the lane change for moving from the 1-8-9-11 "route 710 to the "10-12-13" route 720 The number could be 1.

는, 해당 지역경로에서 현재 측정되는 차량의 통과속도와 정밀지도에 저장된 속도정보의 비율을 고려하여 아래 <표 2>와 같이 교통상황을 원활, 보통 및 혼잡으로 구분하고, 각각의 교통상황별로 다르게 설정될 수 있을 것이다.Here, the complexity coefficient considering the traffic situation is

Considering the ratio of the current measured vehicle passing speed and the speed information stored in the precision map in the local route, traffic conditions are classified into smooth, normal, and congested as shown in <Table 2> below, and different for each traffic situation will be able to set

traffic situation speed ratio k smoothly 70 to 100 One commonly 35 to 70 0.7 confusion 0 to 35 0.3

Next, the path search apparatus 100 compares each reliability score with a preset reliability threshold, determines whether each reliability score is equal to or greater than the preset reliability threshold, and determines whether or not each reliability score is equal to or greater than the preset reliability threshold. By selecting a specific regional route having a specific reliability score, which is the highest reliability score, as an optimal driving route for the autonomous vehicle, the autonomous vehicle may be driven or supported to operate along the specific regional route (S240).

Here, the preset reliability threshold may be automatically set according to the location of the road, road conditions, etc., or may be arbitrarily set by a user or administrator, but is not limited thereto.

On the other hand, in step S240, the path search apparatus 100, as a result of comparing each reliability score with a preset reliability threshold, when all of the plurality of reliability scores are smaller than the preset reliability threshold, send specific global path information to the control server. You can re-receive other global path information by requesting transmission of different global path information.

At this time, the path search apparatus 100 may perform steps S220 to S240 again with reference to the re-received other global path information.

That is, if all of the reliability scores are less than the preset reliability threshold, it may mean that all local routes are unreliable, so a specific global route information and other global route information are requested again, and the other global route information received again is referred to. Thus, by performing steps S220 to S240, a reliable and optimal driving route is found, and the autonomous vehicle can be operated along the optimal driving route.

Meanwhile, as a result of at least some of the processes performed in step S230, the path search apparatus 100 determines that the result of the first process includes information that does not satisfy the driving condition of the autonomous vehicle in the specific global path information. When the result of the second process is determined that the specific path distance value included in the specific global route information is smaller than the lane change distance value, which is the distance required for the autonomous vehicle to change lanes, and the third If the result of the process corresponds to one of the cases in which it is determined that the specific global route information does not include the landmark required for precise positioning on the driving route on which the autonomous vehicle should travel, the specific global route information and You can request transmission of other global path information. In this case, when another global path information is transmitted, the path search apparatus 100 may perform at least a part of the first process, the second process, and the third process for the different global path information.

That is, if the specific global route information includes information that does not satisfy the operating conditions of the autonomous vehicle, the specific route distance value is smaller than the lane change distance value, or the landmark required for precise positioning is not included, the autonomous vehicle Since there is a possibility that the autonomous driving mode may be canceled while the driving vehicle is driving, in order to prevent this or reduce the possibility of the autonomous driving mode being canceled, request global route information other than specific global route information and refer to other global route information received again. So that the above step S220 can be performed again.

Embodiments according to the present invention described above may be implemented in the form of program instructions that can be executed through various computer components and recorded on a computer-readable recording medium. The computer readable recording medium may include program instructions, data files, data structures, etc. alone or in combination. Program instructions recorded on the computer-readable recording medium may be specially designed and configured for the present invention, or may be known and usable to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magnetooptical media such as floptical disks. , and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter or the like as well as machine language codes generated by a compiler. The hardware device may be configured to act as one or more software modules to perform processing according to the present invention and vice versa.

In the above, the present invention has been described by specific details such as specific components and limited embodiments and drawings, but these are provided to help a more general understanding of the present invention, and the present invention is not limited to the above embodiments. , Those skilled in the art to which the present invention pertains may seek various modifications and variations from these descriptions.

Therefore, the spirit of the present invention should not be limited to the above-described embodiments, and not only the claims described later, but also all modifications equivalent or equivalent to these claims belong to the scope of the spirit of the present invention. will do it

Claims (24)

In the method of searching for a route by lane in consideration of the Operational Design Domain (ODD) of an autonomous vehicle,
In a state where global route information is being transmitted from the control server to the autonomous vehicle,
(a) When specific global route information is transmitted, the route search device refers to the driving design area and the stored precise map information, and (i) satisfies the operating condition of the autonomous vehicle in the specific global route information A first process for determining whether or not information that does not exist is included, (ii) a value in which a specific route distance value included in the specific global route information is greater than the lane change distance value, which is a distance required for the autonomous vehicle to change lanes. (iii) a third process for determining whether or not landmarks required for precise positioning are included in the specific global route information on a driving route along which the autonomous vehicle should travel. performing at least some;
(b) As a result of at least some of the processes performed, it is determined that the result of the first process is that there is no information that does not satisfy the driving condition of the autonomous vehicle in the specific global path information, and the second process The result of is determined that the specific route distance value included in the specific global route information is greater than the lane change distance value, which is a distance required for the autonomous vehicle to change lanes, and the result of the third process is the When it is determined that the landmark required for precise positioning is included in the specific global route information on the driving route along which the self-driving vehicle should travel, the route search device refers to the specific global route information searching for a plurality of regional routes located within a predetermined driving distance from the autonomous vehicle, and calculating a reliability score for each of the plurality of regional routes; and
(c) The path search device compares each of the reliability scores with a predetermined reliability threshold to determine whether each reliability score is equal to or greater than the predetermined reliability threshold, and determines whether or not each reliability score is greater than or equal to the predetermined reliability threshold. selecting a specific regional route having a specific reliability score, which is the highest reliability score, as an optimal driving route for the autonomous vehicle, thereby allowing the autonomous vehicle to operate or support driving along the specific regional route;
Path search method comprising a. According to claim 1,
In the first process,
The route search device determines that the specific global route information includes (i) a region where autonomous driving of the autonomous vehicle is impossible, (ii) a route that is out of autonomous driving operating conditions of the autonomous vehicle is included, and and (iii) a case in which a route traveling in an area not included in the precise map information is included, the specific global route information includes information that does not satisfy the driving condition. A path search method characterized in that it is determined that. According to claim 1,
In the second process,
The specific path distance value is connected to a first navigation link, which is a navigation link of an area in which the self-driving vehicle is currently driving, among navigation links included in the specific global path information, and the first navigation link. It is determined according to the geometric relationship, which is a connection form between the second navigation link, which is the navigation link of the area to be driven,
The lane change distance value is determined by a sum of a required distance value calculated for the autonomous vehicle to change lanes and a safety distance value for lane change of the autonomous vehicle. According to claim 3,
The required distance value

is calculated according to the following formula,

in the above formula

is a constant and means the maximum time required for lane change,

is the lane change speed,

is characterized in that it means the number of lanes to be changed,
The lane change speed

is calculated according to the following formula,

in the above formula

Means the speed limit of the road section where the autonomous vehicle is located,

Is a constant and means a safety margin speed considering a lane change. According to claim 1,
In the third process,
The route search device (i) a distance located from a predetermined line connecting the driving route among a plurality of landmarks included in the precision map information is within the maximum recognition distance of each sensor included in the autonomous vehicle. exist, search for and store at least one specific landmark included in the vertical angle range and horizontal angle range of the field of view (FOV) for each sensor, and (ii) the specific land The use score of each of the specific landmarks is calculated by applying each of the importance weights preset for each corresponding landmark to each mark, and the total score obtained by summing all the use scores of each is compared with a preset precision positioning threshold, (iii) If the total score is greater than or equal to the preset precise positioning threshold, it is determined that a landmark necessary for the precise positioning is included in the specific global route information, and if the total score is smaller than the preset precise positioning threshold, the A path search method characterized in that it is determined that a landmark required for precise positioning is not included in the specific global path information. According to claim 1,
In step (b),
The route search device searches for a plurality of specific precision map links located within the predetermined driving distance among precision map links included in the specific global route information, wherein the precision map link means the center line of a road, and A path search method characterized in that by assigning a unique ID to each of the specific precision map links using an algorithm of searching for the plurality of local routes, which are predetermined routes in which at least some of the specific precision map links are connected to each other. According to claim 6,
The route search device calculates the reliability score of each of the local routes according to the following formula,

In the above formula,

denotes the global path following score,

denotes the lane change score required to follow the global route,

Is

means the weight of

Is

means the weight of

A route search method characterized in that denotes an index of each of the local routes. According to claim 7,
The route search device determines the number of specific first precision map links, which are specific precision map links included in the corresponding local route, for each regional route, and determines the number of specific first precision map links corresponding to the specific global route among the specific first precision map links. Calculating the global path following score by multiplying a value divided by the number of specific second precision map links, which are precision map links, by a predetermined K value - the predetermined K value may be set within the range of 0 to 1. Characterized route search method. According to claim 8,
The route search device is configured to: (i) when the specific first precision map link is the same as the specific second precision map link, and (ii) a global path at a prior time point, which is a time point before a predetermined time from the current point in time, a pre-global path. Set the predetermined K value to 1 if at least one of the cases in which a pre-precision map link, which is a precision map link corresponding to a route, and the specific first precision map link are connected,
Even if the first specific precision map link is the same as the specific second precision map link or the case in which the prior precision map link and the specific first precision map link are connected, if an event occurs in the corresponding local path A path search method characterized in that the predetermined K value is set to a value smaller than 1. According to claim 7,
The route search device calculates the lane change score according to the following formula.

calculate,

In the above formula,

is a constant, which means the lateral distance between precision map links,

Means the number of lane changes to the target precision map link to be changed,

Is characterized in that it means a complexity coefficient considering traffic conditions,
Wherein among the plurality of local routes, the lane change score is set to 1 in the case of a local route that does not require a lane change. According to claim 1,
In step (b),
The route search device, as a result of at least some of the processes performed, (i) as a result of the first process, information that does not satisfy the driving condition of the autonomous vehicle is included in the specific global route information (ii) the result of the second process is that the specific route distance value included in the specific global route information is a value smaller than the lane change distance value, which is a distance required for the autonomous vehicle to change lanes. and (iii) a result of the third process determines that the specific global route information does not include the landmark required for precise positioning on the driving route along which the autonomous vehicle should travel. If one of the cases is true, requesting the control server to transmit global path information different from the specific global path information;
When the other global path information is transmitted, the path search apparatus performs at least a part of the first process, the second process, and the third process with respect to the different global path information. According to claim 1,
In step (c),
The path search device, as a result of comparing each of the reliability scores with the predetermined reliability threshold value, determines that all of the plurality of reliability scores are smaller than the predetermined reliability threshold value, to the control server different from the specific global path information. A route search method characterized by requesting transmission of global route information and re-receiving the other global route information. In a route search device for searching a route in lanes in consideration of an Operational Design Domain (ODD) of an autonomous vehicle,
In a state where global route information is being transmitted from the control server to the autonomous vehicle,
at least one memory for storing instructions; and
at least one processor configured to execute the instructions;
Including,
(1) When specific global route information is transmitted, the processor refers to the driving design area and stored precision map information, and (i) does not satisfy the driving condition of the autonomous vehicle in the specific global route information. A first process for determining whether or not information that does not exist is included, (ii) whether a specific route distance value included in the specific global route information is greater than or equal to a lane change distance value, which is a distance required for the autonomous vehicle to change lanes. A second process for determining whether or not, and (iii) a third process for determining whether or not landmarks necessary for precise positioning are included in the specific global route information on a driving route along which the autonomous vehicle should travel. (2) as a result of at least some of the processes performed, it is determined that the result of the first process is that there is no information that does not satisfy the driving condition of the autonomous vehicle in the specific global path information; and, as a result of the second process, it is determined that the specific route distance value included in the specific global route information is greater than or greater than the lane change distance value, which is a distance required for the autonomous vehicle to change lanes. 3 When the result of the process is determined that the landmark required for precise positioning is included in the specific global route information on the driving route along which the autonomous vehicle should travel, with reference to the specific global route information A process of searching for a plurality of local routes located within a predetermined driving distance from the autonomous vehicle and calculating a reliability score for each of the plurality of local paths, and (3) calculating each of the reliability scores with a predetermined reliability threshold By comparing, it is determined whether each reliability score is equal to or greater than the predetermined reliability threshold, and a specific local path having a specific reliability score, which is the highest reliability score among predetermined reliability scores equal to or greater than the predetermined reliability threshold, is determined by the self-driving vehicle. A route search device characterized in that performing a process of driving or supporting the autonomous vehicle to run along the specific regional route by selecting an optimal driving route for the route. According to claim 13,
the processor,
In the first process, when the specific global route information includes (i) an area where autonomous driving of the autonomous vehicle is impossible, (ii) a route that deviate from the autonomous driving operating condition of the autonomous vehicle is included, and (iii) a case in which a route traveling in an area not included in the precise map information is included, the specific global route information includes information that does not satisfy the driving condition. Path search device, characterized in that for determining that. According to claim 13,
In the second process,
The specific path distance value is connected to a first navigation link, which is a navigation link of an area in which the self-driving vehicle is currently driving, among navigation links included in the specific global path information, and the first navigation link. It is determined according to the geometric relationship, which is a connection form between the second navigation link, which is the navigation link of the area to be driven,
The lane change distance value is determined by a sum of a required distance value calculated for the autonomous vehicle to change lanes and a safety distance value for lane change of the autonomous vehicle. According to claim 15,
The required distance value

is calculated according to the following formula,

in the above formula

is a constant and means the maximum time required for lane change,

is the lane change speed,

is characterized in that it means the number of lanes to be changed,
The lane change speed

is calculated according to the following formula,

in the above formula

Means the speed limit of the road section where the autonomous vehicle is located,

Is a constant and means a safety margin speed considering a lane change. According to claim 13,
the processor,
In the third process, (i) a distance from a predetermined line connecting the driving route among a plurality of landmarks included in the precision map information is within the maximum recognition distance of each sensor included in the autonomous vehicle. exist, search for and store at least one specific landmark included in the vertical angle range and horizontal angle range of the field of view (FOV) for each sensor, and (ii) the specific land The use score of each of the specific landmarks is calculated by applying each of the importance weights preset for each corresponding landmark to each mark, and the total score obtained by summing all the use scores of each is compared with a preset precision positioning threshold, (iii) If the total score is greater than or equal to the preset precise positioning threshold, it is determined that a landmark necessary for the precise positioning is included in the specific global route information, and if the total score is smaller than the preset precise positioning threshold, the A route search device characterized in that it is determined that a landmark required for precise positioning is not included in the specific global route information. According to claim 13,
the processor,
In the process (2) above,
A plurality of specific precision map links located within the predetermined driving distance among precision map links included in the specific global route information, wherein the precision map link means the center line of a road, are searched for, and the precision map links are searched for by using a predetermined algorithm. A route search apparatus, characterized in that, by assigning a unique ID to each of the specific precision map links, the plurality of local routes, which are predetermined routes in which at least some of the specific precision map links are connected to each other, are searched. According to claim 18,
the processor,
Calculate the reliability score of each of the local paths according to the following formula,

In the above formula,

denotes the global path following score,

denotes the lane change score required to follow the global route,

Is

means the weight of

Is

means the weight of

denotes an index of each of the local routes. According to claim 19,
the processor,
For each of the regional routes, the number of specific first precision map links, which are specific precision map links included in the corresponding regional route, is calculated by calculating the number of specific first precision map links, which are specific precision map links corresponding to the specific global path, among the specific first precision map links. The global route tracking score is calculated by multiplying a value divided by the number of 2 precision map links by a predetermined K value - the predetermined K value can be set within the range of 0 to 1. . According to claim 20,
the processor,
(i) when the first specific precision map link is the same as the second specific precision map link, and (ii) a precision map corresponding to a global path in advance at a prior time point, which is a time point before a predetermined time from the current time point. If at least one of the cases in which the pre-precision map link, which is a link, and the specific first precision map link are connected, set the predetermined K value to 1,
Even if the first specific precision map link is the same as the specific second precision map link or the case in which the prior precision map link and the specific first precision map link are connected, if an event occurs in the corresponding local path A path search device characterized in that the predetermined K value is set to a value smaller than 1. According to claim 19,
the processor,
According to the following formula, the lane change score

calculate,

In the above formula,

is a constant, which means the lateral distance between precision map links,

Means the number of lane changes to the target precision map link to be changed,

Is characterized in that it means a complexity coefficient considering traffic conditions,
The route search device according to claim 1 , wherein the lane change score is set to 1 in the case of a regional route that does not require a lane change among the plurality of regional routes. According to claim 13,
the processor,
In the process (2) above,
As a result of at least some of the processes performed, (i) when it is determined that the result of the first process includes information that does not satisfy the driving condition of the autonomous vehicle in the specific global route information, ( ii) when a result of the second process determines that the specific route distance value included in the specific global route information is smaller than the lane change distance value, which is a distance required for the autonomous vehicle to change lanes; and (iii) A result of the third process corresponds to one of cases in which it is determined that the specific global route information does not include the landmark required for precise positioning on the driving route along which the autonomous vehicle should travel. If so, request the control server to transmit global path information different from the specific global path information;
When the other global path information is transmitted, the path search apparatus performs at least a part of the first process, the second process, and the third process with respect to the different global path information. According to claim 13,
the processor,
In the above (3) process,
As a result of comparing each of the reliability scores with the preset reliability threshold, when all of the plurality of reliability scores are smaller than the preset reliability threshold, the specific global path information and other global path information are transmitted to the control server. A route search device characterized in that the other global route information is re-received by requesting a transfer.

Images