Method for determining the presence of an emergency stopping lane

Abstract

The invention relates to a method for determining the presence of an emergency stopping lane (3), the method comprising the following steps: - determining a first number of lanes (2) by analyzing the environment; - extracting a second number of lanes (2) from a map database; - comparing the first number and the second number in order to determine the presence or absence of an emergency stopping lane (3).

Description

DESCRIPTION TITLE: Method for determining the presence of an emergency stopping lane [Technical field]

[0001] The invention relates, in a context of assisted or autonomous driving, to a method for determining the presence of an emergency stopping lane. [Prior art]

[0002] For an assisted or autonomous driving system, it is useful to know whether or not an emergency stopping lane is present. This knowledge is helpful for the proper execution of certain driving functions, such as overtaking.

[0003] For driving systems with a certain degree of autonomy, typically level 4 or higher, the existence of an emergency stopping lane provides a safe refuge for stopping the vehicle in case of a problem.

[0004] Currently, determining the presence of an emergency stopping lane is typically done by analyzing images, usually from a camera mounted on the vehicle. This camera is pointed roughly in front of the vehicle and observes the horizontal markings. Image analysis allows for the detection of lane separation lines. Depending on the type of separation line, an emergency stopping lane can be identified.

[0005] An emergency stopping lane includes, on the carriageway side, to separate it from the other normal lanes, a specific separation line periodically comprising a 39 m segment separated by a 13 m gap without markings. On the outer side, the emergency stopping lane has no separation line. Other separation lines, between the normal lanes or on the inner side, are solid lines, single or double, broken lines periodically comprising a 3 m segment separated by a 10 m gap without markings outside built-up areas or periodically comprising a 1.5 m segment separated by a 5 m gap without markings within built-up areas, warning lines periodically comprising a 1.33 m segment separated by a 5 m gap without markings, or mixed lines combining a solid line and a broken line of one of the types described above.It is necessary to add to these markings additional features such as arrows or zebras, which can overlap and further complicate the analysis.

[0006] It follows from the above that determining the presence of an emergency stopping lane requires recognizing with certainty a specific separation line of an emergency stopping lane, i.e. a particular periodic pattern.

[0007] In this scenario, it appears that image processing can be quite easily fooled. Over time, a horizontal marking can be partially or completely erased. Weather conditions can worsen visibility. Therefore, it is possible to mistake a specific line for a broken line or a warning line.

[0008] Therefore, an alternative method for determining the presence of an emergency stopping lane is being sought, offering better robustness to environmental conditions. [Description of the invention]

[0009] To this end, the invention relates to a method for determining the presence of an emergency stopping lane, comprising the following steps: - Determining an initial number of pathways through environmental analysis, - Extraction of a second set of roads from a map database, - comparison of the first number and the second number in order to determine the presence or absence of an emergency stopping lane.

[0010] Specific characteristics or embodiments, usable alone or in combination, are: - If the first number is equal to the second number: there is no emergency stopping lane, and if the first number is equal to the second number plus one: there is an emergency stopping lane. - Environmental analysis is carried out through image processing, preferably using images from at least one video camera, - Image processing extracts lane separation lines from an image and counts them, the first number being equal to the number of lane separation lines, - the second number, extracted from the mapping database, only counts normal roads, - the mapping database does not contain information on the presence / absence of an emergency stopping lane.

[0011] The invention further relates to a device for determining the presence of an emergency stopping lane, implementing such a method.

[0012] The invention also relates to a vehicle comprising such a device. [Description of the drawings]

[0013] The invention will be better understood upon reading the following description, given solely by way of example, and with reference to the figures in the appendix in which:

[0014] [Fig. 1] Figure 1 shows, in a schematic top view, a vehicle on a multi-lane roadway,

[0015] [Fig. 2] Figure 2 shows, in schematic top view, an organic decomposition of the vehicle. [Description of the embodiments]

[0016] The invention relates to a method for determining the presence of an emergency stopping lane 3. According to one feature, this method comprises the following steps. In a first step, a first number N1 of lanes 2 is determined by analyzing the environment. In a second step, which may be prior to, subsequent to, or simultaneous with the first step, a second number N2 of lanes 2 is extracted from a mapping database 5. Once these two numbers N1 and N2 are available, a third step compares the first number N1 and the second number N2 to determine the presence or absence of an emergency stopping lane 3.

[0017] The environmental analysis detects and counts all lanes in the first number, N1: the two normal lanes dedicated to regular traffic and the possible emergency stopping lane (3). Conversely, the mapping database (5), which provides the second number, N2, contains only the number of two normal lanes dedicated to regular traffic. This is used, according to another characteristic, to perform the comparison. The comparison is carried out as follows: If the first number, N1, is equal to the second number, N2, then N1 = N2: there is no emergency stopping lane (3). Conversely, if the first number, N1, is equal to the second number, N2, plus one, then N1 = N2 + 1: there is an emergency stopping lane (3).

[0018] It should be noted that all other cases, N1 < N2 or N1 > N2+1, imply an error. In these other cases, it is not possible to determine whether an emergency stopping lane 3 is present or absent.

[0019] Another characteristic is that environmental analysis is performed through image processing, preferably using images from at least one video camera. This image processing focuses primarily on horizontal marking.

[0020] According to another feature, image processing extracts the 4 lane separation lines from an image and counts them. Here, all the 4 separation lines are considered, the separation lines 4 between the normal lanes and the specific separation line 4 bordering a possible emergency stopping lane 3.

[0021] The first number, N1, is equal to the number of lane separation lines (4). This is because the outermost normal lane (2), on the inside of the carriageway, is bordered by a lane separation line (4) on the inside of the carriageway; a lane separation line (4) is present between any pair of adjacent lanes (2), including any emergency stopping lane (3); and an emergency stopping lane (3) is not bordered by a lane separation line (4) on the outside of the carriageway. Therefore, the number of lane separation lines (4) is equal to the number of lanes (2), including the normal lanes (2) and any emergency stopping lane (3).

[0022] The major advantage of the invention lies at this stage. Indeed, the prior art performs a precise analysis of the separation lines 4 in order to detect a specific separation line 4 of an emergency stopping lane 3 to confirm the presence of an emergency stopping lane 3. Such an analysis presents numerous risks of misleading results, which can lead to a false detection, positive or negative.

[0023] On the contrary, the invention is limited to line detection, regardless of its type, without performing a detailed analysis of the type. Such image processing, simply detecting the separation lines for the purpose of counting them, is much simpler and therefore much more robust to disturbances (faded paint, weather, etc.).

[0024] According to another characteristic, the second number N2, extracted from the cartographic database 5, only counts the normal roads 2. This advantageously allows the use of existing databases and therefore relatively low cost.

[0025] Another advantage of the method according to the invention, in that it uses a cartographic database 5, is that the information provided, in this case the second number N2, is very reliable. Cartographic databases are checked regularly and updated quickly when road infrastructure changes. This information N2 thus benefits from a very high confidence level, close to 100%. This information is not subject to environmental conditions, unlike data measured or perceived by at least one sensor observing the environment, which generally has a lower confidence level.

[0026] It is considered, as a hypothesis of the invention, that the cartographic database 5 does not contain the information of presence / absence of an emergency stopping lane 3.

[0027] This approach is advantageous because it allows the use of a standard-definition (SD) map database. In contrast to high-definition (HD) map databases, an SD map database is less expensive, requires less memory, and covers a larger map area.

[0028] A map database typically includes the road type: national highway, expressway, or motorway. However, it should be noted that while an emergency stopping lane (3) is generally absent from a national highway, it may occasionally be present, particularly when the national highway becomes an expressway, but not always. Conversely, a motorway usually includes an emergency stopping lane (3), but this is not always the case. An emergency stopping lane (3) may be locally or temporarily absent or closed, for example, due to roadworks.

[0029] Also, the type of road alone does not allow us to determine the presence or absence of an emergency stopping lane 3.

[0030] The process of the invention, in that it closely observes the reality of the environment, is therefore very advantageous.

[0031] The invention also relates to a device for determining the presence of an emergency stopping lane 3, implementing the previous method.

[0032] The invention also relates to a vehicle 1, equipped with such a device.

[0033] As illustrated in Figure 2, such a device includes a camera 6, aimed at the front of the vehicle 1. The device preferably incorporates a map database 5. Alternatively or in addition to an onboard map database 5, the device can be connected to a remote map database 5 via a wireless communication network. All these elements (map database 5, camera 6) are interfaced by a processing unit 7. This processing unit 7 is capable of implementing the method.

[0034] The invention has been illustrated and described in detail in the drawings and the preceding description. This description is to be considered illustrative and given by way of example, and not as limiting the invention to this single description. Numerous embodiments are possible.

Claims

Demands

1. A method for determining the presence of an emergency stopping lane (3), characterized in that it comprises the following steps: - determination of a first number (N1) of paths (2), by analysis of the environment, - extraction of a second number (N2) of roads (2), from a cartographic database (5), - comparison of the first number (N1) and the second number (N2) in order to determine the presence or absence of an emergency stopping lane (3).

2. Method according to claim 1, where: - if the first number (N1) is equal to the second number (N2): there is no emergency stopping lane (3), and - if the first number (N1) is equal to the second number (N2) plus one: there is an emergency stopping lane (3).

3. A method according to any one of claims 1 or 2, wherein the analysis of the environment is carried out by image processing, preferably from at least one video camera (6).

4. Method according to claim 3, wherein the image processing extracts from an image the lane separation lines (4) (2) and counts them, the first number (N1) being equal to the number of lane separation lines (4).

5. A method according to any one of claims 1 to 4, wherein the second number (N2), extracted from the map database (5), counts only normal lanes.

6. A method according to any one of claims 1 to 5, wherein the map database (5) does not contain the presence / absence information of an emergency stopping lane (3).

7. Device for determining the presence of an emergency stopping lane (3), characterized in that it implements a method according to any one of the preceding claims.

8. Vehicle (1), characterized in that it comprises a device according to the preceding claim.

Images