Method for estimating the trajectory of a vehicle, driver assistance method and system, and vehicle

Abstract

The present invention relates to a method (100) for estimating the trajectory (14) of a vehicle (1) in a road environment (2), the method comprising: a step (110) of collecting data (10) relating to the environment (2) of the vehicle (1), including at least the following data (10): a number (11) of objects (3) in the environment (2), and speed of movement (12) of the objects (3); a step (120) of estimating the trajectory (13) of the objects (3) from the collected data (10); a step (130) of creating at least one virtual road lane (5) from the estimated trajectories (13) of the objects (3); and a step (140) of estimating the trajectory (14) of the vehicle (1), the step being carried out on the basis of the following data: data (10) relating to the vehicle (1), including at least data relating to the steering wheel angle (17) and / or data relating to the turn indicators (18), and the one or more virtual road lanes (5) created in the preceding step (130).

Description

Description Title of the invention: Method for estimating the trajectory of a vehicle, method and system for driver assistance, and vehicle. Technical field

[0001] The present invention relates to a method for estimating the trajectory of a vehicle in a road environment. The invention also relates to a driver assistance method incorporating such a trajectory estimation method. The invention further relates to an advanced driver-assistance system (ADAS). The invention also relates to a vehicle equipped with such a system.

[0002] The field of the invention is that of advanced driver assistance systems, in particular for autonomous motor vehicles. Previous art

[0003] In this area, there are already different methods that take into account the road lanes in which vehicles travel.

[0004] US9261601 describes a method for estimating highway lane boundaries, as well as the position and orientation of a vehicle. A GPS system transmits the position of one or more vehicles to a trip history module, which generates a history of vehicle coordinates. A highway lane boundary estimation module generates statistical distribution curves representing the vehicle coordinate history. A processing module generates sets based on the statistical distribution curves. A lane determination module defines the width and position of each lane on the highway based on the sets. A processor combines the results for the width and position of each highway lane.

[0005] EP2899669 describes a method for estimating the position of a vehicle in a road lane. The vehicle's environment is monitored by at least one sensor, allowing lateral position information to be obtained for the vehicle within its lane of travel.

[0006] EP3835824 describes various methods for implementing an advanced driver assistance system (ADAS). The system performs trajectory estimations, taking into account the distance between an object and a vehicle.

[0007] However, the known methods are not entirely satisfactory. Description of the invention

[0008] The aim of the present invention is to propose an improved method.

[0009] To this end, the invention relates to a method for estimating the trajectory of a vehicle in a road environment, comprising: - a step of collecting data relating to the vehicle's environment, including at least the following data: • number of objects in the environment, and • speed of movement of objects; - a step of estimating the trajectory of objects from the collected data; - a step involving the creation of at least one virtual road, based on the estimated trajectories of objects; and - a step to estimate the vehicle's trajectory, carried out using the following data: • Vehicle data, including at least steering angle data and / or turn signal data, and • the virtual road(s) created in the previous step.

[0010] Thus, the invention makes it possible to improve the determination of the vehicle's trajectory, particularly when traditional lane markings are absent or take some time to be detected by the sensors, such as in roundabouts, when reaching the top of a hill, etc.

[0011] The invention provides improved navigation, enabling vehicles to navigate challenging environments with greater confidence and precision. The invention reduces reliance on physical road markings, which is beneficial in areas where road infrastructure may be incomplete or degraded.

[0012] According to other advantageous features of the method according to the invention, taken individually or in combination:

[0013] The method includes an autonomous driving stage, consisting of driving the vehicle without driver intervention along the estimated vehicle trajectory.

[0014] The autonomous driving stage includes adaptive speed regulation and / or steering angle control.

[0015] The data collection step includes a first sub-step of determining the number of objects in the environment and then, if the number of objects in the environment is greater than a threshold number of objects, a second sub-step of determining the speed of movement of the objects.

[0016] The step of creating virtual roadways is carried out when the number of objects in the environment is greater than a threshold number of objects.

[0017] The step of creating virtual roadways is carried out when objects in the environment have a movement speed greater than a speed threshold.

[0018] The vehicle trajectory estimation step is carried out taking into account the following data: - vehicle acceleration and / or - the speed of the vehicle and / or - the vehicle's trajectory at the time of the estimation stage.

[0019] The invention also relates to a driving assistance method, incorporating the trajectory estimation method described above, characterized in that the driving assistance method includes a step of updating measured data, including the detection of road markings, then a step of determining a sufficiency of data available for calculating the vehicle's trajectory; and in that the trajectory estimation method is implemented when the determination step concludes that there is insufficient data available for calculating the vehicle's trajectory.

[0020] The invention also relates to an advanced driver assistance system for a vehicle in a road environment, the system comprising: - a set of sensors providing data relating to the vehicle and its environment, including at least the following data: • number of objects in the environment, • speed of object movement, - an electronic control unit configured for: • estimate the trajectory of the movement of objects, based on the data relating to the objects collected by the sensors; • create virtual roadways, based on the estimated movement trajectories of objects; and • estimate the vehicle's movement trajectory from, on the one hand, data relating to the vehicle, including at least data relating to the steering angle and / or data relating to the turn signals, and, on the other hand, virtual road lanes.

[0021] The invention also relates to a vehicle, equipped with a system as described above. Description of the figures

[0022] The invention will be better understood upon reading the following description, given solely by way of non-limiting example and made with reference to the accompanying drawings in which:

[0023] [Fig.1] is a schematic representation of a vehicle according to the invention and its road environment.

[0024] [Fig.2] is a flowchart illustrating the trajectory estimation method according to the invention and its different steps.

[0025] [Fig.3] is a flowchart illustrating a driving assistance system method and its different stages, integrating the trajectory estimation method according to the invention.

[0026] [Fig.4] is a schematic representation, comprising five vignettes, of a first example of implementation of the trajectory estimation method, when the vehicle arrives at a roundabout.

[0027] [Fig.5] is a schematic representation, comprising four vignettes, of a second example of implementation of the trajectory estimation method, when the vehicle arrives at an intersection. Detailed description of the invention

[0028] Figure 1 describes a vehicle (1) according to the invention, equipped with a driver assistance system (20). The environment (2) may include moving objects (3), in particular other vehicles.

[0029] The system (20) is configured to implement a method (100) for estimating the trajectory (14) of the vehicle (1) according to the invention. More broadly, the system (20) is configured to implement a method (200) for assisting the driving of the vehicle (1), incorporating the method (100) for estimating the trajectory (14) of the vehicle (1).

[0030] The system (20) is configured to process different data (10), such as: - the number (11) of objects (3) in the environment (2), - the speed of movement (12) of objects (3), - the trajectory (13) of objects (3), - the trajectory (14) of the vehicle (1), - the speed of movement (15) of the vehicle (1), - 1' acceleration ( 16) of the vehicle ( 1 ) , - data relating to the steering angle (17) of the vehicle (1), - data relating to the turn signals (18) of the vehicle (1).

[0031] The system (20) includes a set of sensors (30) for data (10) relating to the vehicle (1) and its environment (2). The system (20) also includes an electronic control unit (40) configured to process the data (10) and execute various steps of the method (100), as detailed below. The system (20) also includes means for autonomous driving (50) the vehicle (1), including cruise control and steering angle control.

[0032] Figure [Fig.2] describes the trajectory estimation method (100) (14) and its different steps (110, 120, 130, 140, 150).

[0033] The method (100) includes a step of collecting (110) data (10) relating to the environment (2) of the vehicle (1), including at least the following data (10): number (11) of objects (3) in the environment (2), and speed of movement (12) of the objects (3).

[0034] The method (100) includes a step of estimating (120) the trajectory (13) of the objects (3) from the data (11, 12) collected in step (110) using a technique known from the prior art.

[0035] The method (100) includes a step of creating (130) virtual roadways (5), from the trajectories (13) of the objects (3) estimated in the step (120).

[0036] For a given object (3), the corresponding virtual road lane (5) is, for example, centered on the object (3), with a predefined width and parallel to the trajectory (13) estimated for the object (3) during the estimation step (120). The predefined width is, for example, the classic width of a traffic lane.

[0037] The method (100) includes a step (140) for estimating the trajectory (14) of the vehicle (1). This step (140) is performed, on the one hand, using data (10) relating to the vehicle (1), including at least data relating to the steering angle (17) and / or data relating to the turn signals (18), and on the other hand, using the virtual road lanes (5) created in the previous step (130). Furthermore, the step for estimating the trajectory (140) of the vehicle (1) can be performed taking into account the acceleration (16) of the vehicle (1) and / or the speed (15) of the vehicle (1) and / or the trajectory (14) of the vehicle (1) at the time of the estimation step (140).

[0038] The estimated trajectory (14) for vehicle (1) follows, for example, the virtual road lane (5) created during the creation step (130), which vehicle (1) is most likely to follow given the data relating to vehicle (1).

[0039] For example, if vehicle (1) has its right turn signal activated, there is a good chance that vehicle (1) will turn right at the next intersection and therefore follow a virtual road lane (5) allowing it to turn right.

[0040] The method (100) also includes an autonomous driving step (150), consisting of driving the vehicle (1) without driver intervention along the trajectory (14) of the vehicle (1) estimated in step (140). The autonomous driving step (150) includes adaptive cruise control and / or steering angle control.

[0041] Step (110) is performed by the sensors (30) and the electronic control unit (40). Steps (120, 130, 140) are performed by the electronic control unit (40). Step (150) is performed by the autonomous driving means (50).

[0042] Figure 3 describes the driving assistance method (200) integrating the trajectory estimation method (100) (14), with their different stages (210, 220, 230, 240, 250, 260, 270; 110, 120, 130, 140, 150).

[0043] The method (200) includes a step (210) of updating the data (10) relating to the vehicle (1). This data (10) may include the current trajectory (14) of the vehicle (1), its speed of travel (15), its acceleration (16), its steering angle (17), as well as data relating to its turn signals (18).

[0044] The method (200) includes a step (220) of updating the data (10) measured by the sensors (30) concerning the environment (2) of the vehicle (1), including the detection of road markings. The sensors (30) may include a front camera and a front-facing central radar.

[0045] The method (200) includes a step (230) for determining whether sufficient data (10) is available for calculating the trajectory (14) of the vehicle (1). When the determination step (230) concludes that sufficient data (10) is available for calculating the trajectory (14) of the vehicle (1), then step (240) is implemented. When the determination step (230) concludes that insufficient data (10) is available for calculating the trajectory (14) of the vehicle (1), then the trajectory estimation method (100) is implemented.

[0046] The method (200) includes a step (240) of calculating the trajectory (14) of the vehicle (I) from the available data (10).

[0047] The method (200) includes a step (250) of taking into account the ground markings to calculate the trajectory (14).

[0048] The method (200) includes a step (260) of taking into account the past trajectory (14) to calculate the future trajectory (14).

[0049] The method (200) includes a step (270) of predicting a trajectory (14) for the vehicle (1).

[0050] The stages (140, 270) are followed by the autonomous driving stage (150).

[0051] As shown in [Fig.3], the data collection step (110) (10) includes a first substep (111) and a second substep (112).

[0052] Substep (111) consists of determining the number (11) of objects (3) in the environment (2), at a detection distance from the vehicle (1). The detection distance corresponds to the area of ​​interest where the system (20) detects the objects (3). This detection distance depends on the speed (15) of the vehicle (1): the higher the speed (15), the greater the detection distance. Only objects (3) detected within the detection distance are taken into account. It is therefore necessary to have a minimum number (II) of objects (3), greater than a threshold number of objects (3), to construct the scene according to the properties of the objects (3). The minimum number (11) of objects (3) varies according to the detection distance and the capabilities of the sensors (30).

[0053] If the number (13) of objects (3) in the environment (2) is greater than the threshold number of objects (3), substep (111) is followed by substep (112), which consists of determining the speed of movement (12) of the objects (3).

[0054] Steps (120, 130) are performed following step (110) only when the number (13) of objects (3) in the environment (2) exceeds the object number threshold, and when the objects (3) in the environment (2) have a movement speed (14) exceeding a speed threshold. Otherwise, step (110) is followed by a kinetic trajectory prediction step (114) for the vehicle (1).

[0055] The threshold for the number of objects (3) can be defined for example at 2, preferably 3. The speed threshold can be defined for example at 2 km / h.

[0056] Between steps (130, 140) there is a step (132) for verifying data (10) relating to the vehicle (1). This data (10) may include the current trajectory (14) of the vehicle (1), its speed (15), its acceleration (16), its steering angle (17), as well as data relating to its turn signals (18).

[0057] Figure 4 shows a first example of the implementation of system (20) and of method (100), when the environment (2) includes a roundabout. [Fig.4] contains five vignettes 4a, 4b, 4c, 4d and 4e, showing the vehicle (1) and its environment (2).

[0058] Thumbnail 4a shows step (111), during which four objects (3) (11) are detected in the environment (2) of the vehicle (1).

[0059] Vignette 4b shows step (112), during which the speeds (12) of the four objects (3) are detected, concluding that one of the objects (3) is at rest, while the other three objects (3) are moving at a speed (12) above the speed threshold.

[0060] Vignette 4c shows step (120), during which the trajectories (13) of the three moving objects (3) are estimated.

[0061] Thumbnail 4d shows step (130), during which a virtual roadway (5) is created from the trajectories (13).

[0062] The vignette 4e shows the step (140), during which the trajectory (14) of the vehicle (1) on the lane (6) is estimated taking into account the virtual road lane (5), as well as data (10) relating to the vehicle (1), including at least data relating to the steering angle (17) and / or data relating to the turn signals (18).

[0063] Figure 5 shows a second example of the implementation of system (20) and method (100), where environment (2) has an intersection. Figure 5 contains four thumbnails: 5a, 5b, 5c, and 5d.

[0064] Vignette 5a shows step (112), during which the velocities (12) of two objects (3) present in the environment (2) are detected, concluding that one of the objects (3) is moving in one direction, while the other object (3) is moving in the opposite direction.

[0065] Vignette 5b shows step (120), during which the trajectories (13) of the two moving objects (3) are estimated.

[0066] Thumbnail 5c shows step (130), during which two parallel virtual roadways (5) are created from the trajectories (13) of the objects (3).

[0067] The vignette 5d shows the step (140), during which the possible trajectories (14) of the vehicle (1) are estimated taking into account the virtual road lanes (5), as well as the data (10) relating to the vehicle (1), including at least data relating to the steering angle (17) and / or data relating to the turn signals (18).

[0068] Furthermore, the vehicle (1), the system (20), and the method (100) can be configured differently from Figures 1 to 5 without departing from the scope of the invention, which is defined by the claims. In addition, the technical characteristics of the various embodiments and variants mentioned above can be combined, in whole or in part. Thus, the vehicle (1), the system (20), and the method (100) can be adapted in terms of cost, functionality, and performance.

Claims

Demands

1. Method (100) for estimating the trajectory (14) of a vehicle (1) in a road (2) environment, comprising: - a data collection step (110) of data (10) relating to the environment (2) of the vehicle (1), including at least the following data (10): • number (11) of objects (3) in the environment (2), and • speed of movement (12) of objects (3); - an estimation step (120) of the trajectory (13) of the objects (3) from the data (10) collected; - a step of creating (130) at least one virtual road (5), from the trajectories (13) of the estimated objects (3); and - an estimation step (140) of the trajectory (14) of the vehicle (1), carried out using the following data: • data (10) relating to the vehicle (1), including at least data relating to the steering angle (17) and / or data relating to the turn signals (18), and • the virtual roadway(s) (5) created in the previous step (130).

2. Method (100) according to claim 1, characterized in that it comprises an autonomous driving step (150), consisting of driving the vehicle (1) without driver intervention along the estimated trajectory (14) of the vehicle (1).

3. Method (100) according to claim 2, characterized in that the autonomous driving step (150) includes adaptive speed regulation and / or steering angle control.

4. Method (100) according to any one of the preceding claims, characterized in that the data (10) collection step (110) comprises a first substep consisting of determining the number (11) of objects (3) in the environment (2) and then, if the number (13) of objects (3) in the environment (2) is greater than a threshold number of objects, a second substep consisting of determining the speed of movement (12) of the objects (3).

5. Method (100) according to any one of the preceding claims, characterized in that the virtual road (5) creation step (130) is carried out when the number (13) of objects (3) in the environment (2) is greater than a threshold number of objects.

6. Method (100) according to any one of the preceding claims, characterized in that the step of creating (130) virtual roadways (5) is carried out when the objects (3) in the environment (2) have a movement speed (14) greater than a speed threshold.

7. Method (100) according to any one of the preceding claims, characterized in that the estimation step (140) of the trajectory (14) of the vehicle (1) is carried out taking into account the following data: - the acceleration (16) of the vehicle (1) and / or - the speed of movement (15) of the vehicle (1) and / or - the trajectory (14) of the vehicle (1) at the time of the estimation step (140).

8. A driving assistance method (200), incorporating the trajectory estimation method (100) (14) according to any one of the preceding claims, characterized in that the driving assistance method (200) comprises a step (220) of updating measured data (10), including the detection of road markings, and then a step of determining whether sufficient data (10) is available for calculating the trajectory (14) of the vehicle (1); and in that the trajectory estimation method (100) (14) is implemented when the determination step (230) concludes that there is insufficient data (10) available for calculating the trajectory (14) of the vehicle (1).

9. Advanced system (20) for assisting the driving of a vehicle (1) in a road environment (2), the system (20) comprising: - a set of sensors (30) for data (10) relating to the vehicle (1) and its environment (2), including at least the following data (10): • number (11) of objects (3) in the environment (2), • speed of movement (12) of objects (3), - an electronic control unit (40) configured for: • estimate the trajectory of movement (13) of the objects (3), from the data (10) relating to the objects (3) collected by the sensors (30); • create virtual roadways (5), based on the estimated movement trajectories (13) of the objects (3); and • estimate the trajectory of movement (14) of the vehicle (1) from, on the one hand, the data (10) relating to the vehicle (1), including at least data relating to the steering angle (15) and / or data relating to the turn signals (16), and, on the other hand, the virtual road lanes (5).

10. Vehicle (1), equipped with a system (20) according to claim 9.

Images