System and method for predicting trajectory of vehicle

Abstract

A method (50) for predicting a trajectory of a host vehicle traveling on a main lane, wherein a lane change of the host vehicle from the main lane to an adjacent lane is determined based on an estimation of dynamic behavior of a group of vehicles traveling on the adjacent lane, the group of vehicles including at least a primary vehicle located proximate to the host vehicle and a secondary vehicle located rearward of the host vehicle.

Description

TECHNICAL FIELD

[0001] The present invention relates to the field of devices and methods for predicting the path and / or trajectory of a motor vehicle, as well as computer programs intended to implement such methods.

[0002] More particularly, the present invention relates to the field of driver assistance of a motor vehicle, and in particular for assisting the driver in activating or deactivating a driver assistance system. BACKGROUND

[0003] Currently, motor vehicles are equipped with increasingly efficient advanced driver assistance systems (ADAS). The purpose of advanced driver assistance systems is to allow autonomous driving of a motor vehicle (i.e. without driver intervention or sharing the driving with the vehicle driver) in order to keep the vehicle in its lane and / or to reduce its speed. In particular, advanced driver assistance systems can be used to predict the path and / or trajectory of a motor vehicle. In the present application, the path of a vehicle will be considered as the geometric shape corresponding to the progression of the vehicle between a departure point and an arrival point. The trajectory of a vehicle will be considered as the temporal evolution of the position of the vehicle between a departure point and an arrival point.

[0004] Vehicles called "autonomous" vehicles or partially assisted driving vehicles require a sufficient model of the vehicle environment to allow algorithms to make decisions. This can be achieved by means of various proprioceptive sensors such as accelerometers, gyroscopes, etc. and various exteroceptive sensors such as cameras, radars, LIDARs, ultrasonic sensors, etc. and data fusion methods configured to process the information received and calculate the state (position, speed, acceleration, yaw, etc.) of the vehicle and the surrounding objects.

[0005] Thus, in such vehicles, it is necessary to predict the movements of the host vehicle as well as other moving objects which are present in the immediate environment of the vehicle and which are likely to become obstacles when there is a conflict between the trajectory of the host vehicle and the trajectory of the objects.

[0006] For example, lane keeping assistance (LKA) systems are known which allow the vehicle to automatically reposition itself in its lane, or even lane change assistance (LCA) systems which allow the vehicle to change lanes.

[0007] Other examples of driver assistance are also known, such as automatic emergency steering (AES) assistance systems which are able to detect any obstacles and to perform an emergency turn, predict the trajectory of moving objects allowing the analysis of potential collision risks, or even adaptive cruise control (ACC) assistance systems which are able to control the speed and to automatically maintain a safe distance from the vehicle in front.

[0008] A method for controlling a lane change of a host vehicle as a function of a first distance between two vehicles adjacent to the host vehicle as a target for the lane change of the host vehicle is known from document EP 3056405-A1.

[0009] The main drawback of these driver assistance systems is the lack of fluidity of their response reaction. Indeed, it is particularly difficult to predict the behavior of a vehicle located in a lane adjacent to the host vehicle, so the decision time of these assistance systems is particularly long.

[0010] There is therefore a need to optimize the lane change of a host vehicle from a main lane to an adjacent lane. SUMMARY

[0011] In view of the above, the aim of the present invention is to allow the trajectory of a motor vehicle to be predicted while overcoming the aforementioned drawbacks.

[0012] The aim of the present invention is a method for predicting the trajectory of a host vehicle traveling on a main lane, in which a lane change of the host vehicle from the main lane to an adjacent lane is determined as a function of an estimation of the dynamic behavior of a group of vehicles traveling on the adjacent lane. The group of vehicles comprises at least one main vehicle located in the vicinity of the host vehicle and a secondary vehicle located behind the host vehicle.

[0013] The dynamic behavior of the vehicles in the lane adjacent to the lane in which the host vehicle is located is therefore predicted in order to make a lane change decision.

[0014] The prediction of the trajectory of the host vehicle is determined as a function of the behavior of the group of vehicles present on the lane adjacent to the lane in which the host vehicle is traveling.

[0015] Advantageously, the position, orientation and speed information of the host vehicle and of the vehicles in the group of vehicles are collected and a dynamic model of each pair of successive vehicles traveling on the adjacent lane is established as a function of the collected information. This allows an accurate dynamic model of each vehicle in the adjacent vehicles to be obtained as a function of the data of the vehicles in the vicinity in the same lane.

[0016] The dynamic model of each pair of successive vehicles is obtained by using an auto-regressive exogenous (ARX) calculation model to obtain the dynamic data of a pair of successive vehicles in order to determine a second-order transfer function corresponding to the behavior of the host vehicle with respect to the pair of considered successive vehicles. The behavior of the host vehicle depends on its longitudinal model and longitudinal controller.

[0017] The position, orientation and speed of the vehicles are obtained in particular by various proprioceptive sensors and exteroceptive sensors of the perception system of the host vehicle.

[0018] For example, the dynamic model established is validated by comparing the error of this auto-regressive exogenous calculation model with a threshold depending on the actual speed of each vehicle at a certain time instant and the speed of said vehicle at the preceding time instant.

[0019] According to a subsequent step, the movements of these neighbouring vehicles are predicted from the validated dynamic model and the initial position of said vehicles, the movement of the host vehicle is predicted from the movements of these neighbouring vehicles and the information originating from the proprioceptive sensors of the host vehicle, and the lane change of the host vehicle is determined from said predictions of the movements of this vehicle and of these neighbouring vehicles, the overall trajectory of the host vehicle, and information originating for example from a map of the road on which the host vehicle is travelling.

[0020] The step of determining the lane change of the host vehicle allows to evaluate a suitable time window that allows the host vehicle to safely change lane.

[0021] Then, a lane change instruction for the host vehicle is sent to a module for executing the lane change of the host vehicle.

[0022] In the case where the host vehicle cannot change lane, the host vehicle is informed that a modification of a parameter such as in particular its speed is required.

[0023] Advantageously, the steps of the method are repeated until a lane change is found possible.

[0024] According to a second aspect, the application relates to a system for predicting the trajectory of a host vehicle travelling on a main lane, the system being configured to determine a lane change of the host vehicle from the main lane to a neighbouring lane from an estimation of the dynamic behaviour of a group of vehicles travelling on the neighbouring lane, said group of vehicles comprising at least a main vehicle located in the vicinity of the host vehicle and a secondary vehicle located behind said host vehicle.

[0025] Advantageously, the system comprises:

[0026] - a module for collecting or retrieving position, orientation and speed information of the host vehicle and of the vehicles of the group of vehicles;

[0027] - a module for estimating a dynamic model of each pair of successive vehicles travelling on the neighbouring lane from the information collected. This allows to obtain an accurate dynamic model of each vehicle of the neighbouring vehicles from the data of the vehicles in the same lane that are in the vicinity.

[0028] The dynamic model of each pair of successive vehicles is obtained for example by using an auto-regressive exogenous (ARX) calculation model to obtain the dynamic data of a pair of successive vehicles to determine a second order transfer function corresponding to the behaviour of the pair of considered successive vehicles compared to the host vehicle. The behaviour of the host vehicle depends on its longitudinal model and longitudinal controller.

[0029] The position, orientation and speed of the vehicles are acquired in particular by various proprioceptive and exteroceptive sensors of the perception system of the vehicle.

[0030] Advantageously, the system comprises:

[0031] - a module for validating the dynamic model established by comparing the error of the autoregressive exogenous calculation model with a threshold depending on the actual speed of each vehicle at a certain time and on the speed of the said vehicle at the previous time;

[0032] - a module for predicting the movements of the neighbouring vehicles as a function of the validated dynamic model and the initial position of the said vehicles;

[0033] - a module for predicting the movement of the vehicle as a function of the prediction of the movements of the neighbouring vehicles and the information originating from the proprioceptive sensors of the vehicle;

[0034] - a module for determining the lane change of the vehicle as a function of the said predictions of the movements of the vehicle and of the neighbouring vehicles, the global trajectory of the vehicle and information originating for example from a map of the road on which the vehicle is travelling.

[0035] According to another aspect, the application relates to an automotive vehicle comprising a system for perception and a system for predicting the trajectory of the vehicle as previously described. BRIEF DESCRIPTION OF DRAWINGS

[0036] Further objects, features and advantages of the application will become apparent upon reading of the following description, made only by way of non-restrictive example, and with reference to the appended drawings, in which:

[0037] [ Figure 1 ] is a schematic view of two adjacent lanes in which an automotive vehicle and a plurality of neighbouring vehicles are travelling, the vehicle comprising a trajectory prediction system according to an embodiment of the application;

[0038] [ Figure 2 ] schematically illustrates a system for predicting the trajectory of the vehicle according to an embodiment of the application; and Figure 1

[0039] [ Figure 3 ] shows a flowchart of a method for predicting the trajectory of the vehicle according to an embodiment of the application implemented by the system of Figure 1 DETAILED DESCRIPTION

[0040] Figure 1 Two adjacent lanes 1, 2 in which automotive vehicles are travelling in the same direction of travel are shown highly schematically. ​​

[0041] As shown, the ego vehicle 10 is travelling on a first lane 1 and four vehicles 3, 4, 5, 6 are travelling on an adjacent lane 2 adjacent to the first lane.

[0042] The vehicles 3, 4, 5, 6 form a group 7 of vehicles travelling on the adjacent lane 2 adjacent to the lane 1 of the ego vehicle 10.

[0043] The ego vehicle 10 comprises a system 11 for perceiving the environment of said vehicle, configured to detect the group 7 of vehicles travelling on the adjacent lane 2.

[0044] Typically, the group 7 of vehicles travelling on the adjacent lane 2 comprises at least a primary vehicle located in the immediate vicinity of the ego vehicle 10 and at least a secondary vehicle located behind the ego vehicle 10.

[0045] The perception system 11 allows to detect the primary vehicle 6 located in the immediate vicinity of the ego vehicle 10 in the adjacent lane 2.

[0046] The perception system 11 comprises various proprioceptive sensors such as accelerometers, gyroscopes, etc. and various exteroceptive sensors such as cameras, radars, LIDARs, ultrasonic sensors, etc. and a data fusion method configured to process the received information and calculate the state (position, speed, acceleration, yaw, etc.) of the ego vehicle 10 and of the surrounding objects 7.

[0047] The behaviour of the vehicle detecting the primary vehicle 6 allows to determine a prediction of the position of the secondary vehicles 3, 4, 5 located behind the primary vehicle 6 in the direction of travel of the vehicles.

[0048] The speed fluctuations of the primary vehicle propagate to the secondary vehicles, so that the more the number of secondary vehicles, the more the ability to predict a time window for a lane change for the ego vehicle 10.

[0049] The prediction allows to create a time window for a lane change for the ego vehicle 10 from all the vehicles travelling on the lane adjacent to the ego vehicle 10.

[0050] The ego vehicle comprises a system 12 for predicting the trajectory of said ego vehicle, configured to send a lane change instruction for the ego vehicle as a function of the estimation of the behaviour of the vehicles travelling in the adjacent lane 2.

[0051] As shown in detail in Figure 2 The system 12 for predicting the trajectory of the ego vehicle 10 comprises a module 13 for determining the state of the vehicles 3, 4, 5, 6 travelling on the adjacent lane 2 adjacent to the ego vehicle 10, as shown in detail in

[0052] To this end, the module 13 comprises a module 13a for determining the position P and the orientation O of the vehicles 3, 4, 5, 6 travelling on the adjacent lane 2 adjacent to the host vehicle 10, and a module 13b for determining the speed V of the vehicles 3, 4, 5, 6 travelling on the adjacent lane 2 adjacent to the host vehicle 10. The position P, the orientation O and the speed V of the adjacent vehicles are in particular obtained by means of the various proprioceptive and exteroceptive sensors of the perception system 11 of the host vehicle 10.

[0053] The system 12 for predicting the trajectory of the host vehicle 10 further comprises a module 14 for estimating a dynamic model of the vehicles travelling on the adjacent lane 2. The module 14 is configured to establish a dynamic model of each pair of consecutive vehicles travelling on the adjacent lane 2. This allows to obtain an accurate dynamic model of each of the adjacent vehicles from the data of the vehicles adjacent in the same lane.

[0054] The module 14 for estimating a dynamic model of a vehicle is configured to determine a second order transfer function corresponding to the behaviour of the vehicle with respect to the adjacent vehicle. The behaviour of the host vehicle depends on its longitudinal model and longitudinal controller.

[0055] The module 14 uses an auto-regressive exogenous (ARX) computational model to obtain the dynamic data of a pair of consecutive vehicles.

[0056] The auto-regressive exogenous computational model is expressed according to the following equation:

[0057] A(z).y(t) = B(z).u(t - nk) + e(t).

[0058] wherein:

[0059] z is the time offset;

[0060] nk is the delay;

[0061] u(t) is the input data, in this case the speed of the aforementioned secondary vehicle;

[0062] y(t) is the output data, in this case the speed of the primary vehicle;

[0063] e(t) is the error value; and

[0064] A(z) and B(z) are second order polynomials.

[0065] The polynomials A(z) and B(z) are expressed according to the following equations:

[0066] A(z) = 1 + a1.z -1 + a2.z -2

[0067] B(z) = b1+ b2.z -1+ b3.z -2 .

[0068] The estimated dynamic model is then validated in a module 16 for validating the dynamic model, which is configured to validate the dynamic model as a function of the actual speed V(t) at the instant t and the preceding speed V(t-1) at the preceding instant t-1.

[0069] The system 12 for predicting the trajectory of the host vehicle 10 further comprises a module 18 for predicting the movements of the neighboring vehicles as a function of the validated dynamic model and the initial positions of the neighboring vehicles, and a module 20 for predicting the movement of the host vehicle 10 as a function of the movement predictions provided by the module 18 and the information originating from the proprioceptive sensors C of the host vehicle 10.

[0070] The system 12 for predicting the trajectory of the host vehicle 10 further comprises a module 22 for determining a lane change of the host vehicle 10 as a function of the predictions of the movements of the host vehicle and of the neighboring vehicles, the overall trajectory T of the host vehicle 10, and the information originating non-limitatively from a map Cart of the road on which the host vehicle is driving.

[0071] The module 22 for determining a lane change by the host vehicle 10 is configured to evaluate a suitable time window that allows the host vehicle to safely change lane.

[0072] A lane change instruction for the host vehicle 10 is sent to a module 24 for executing the lane change.

[0073] In the event that the host vehicle cannot change lane, the module 22 for determining a lane change can be configured to inform the module 20 for predicting the trajectory of the host vehicle 10, in particular with a view to modifying a parameter such as in particular its speed.

[0074] As Figure 3 illustrated, the method 50 for predicting the trajectory of the host vehicle 10 comprises a step 51 for determining the state of the vehicles 3, 4, 5, 6 driving on the neighboring lanes 2 adjacent to the host vehicle 10.

[0075] The determining step 51 allows the position P, the orientation O and the speed information of the host vehicle 10 and of the vehicles 3, 4, 5, 6 driving on the neighboring lanes 2 adjacent to the host vehicle 10 to be collected or retrieved. The position P, the orientation O and the speed V of the neighboring vehicles are in particular acquired by the various proprioceptive and exteroceptive sensors of the perception system 11 of the host vehicle 10.

[0076] The method 50 for predicting the trajectory of the host vehicle 10 further comprises a step 52 of estimating a dynamic model of the vehicles driving on the adjacent lane 2. During this step 52, a dynamic model is established for each pair of successive vehicles driving on the adjacent lane 2. This allows to obtain an accurate dynamic model for each of the vehicles of the adjacent vehicles from the data of the vehicles adjacent in the same lane.

[0077] The dynamic model of each pair of successive vehicles is obtained by using an auto regressive exogenous (ARX) computation model to obtain the dynamic data of a pair of successive vehicles to determine a second order transfer function corresponding to the behavior of the vehicle with respect to the adjacent vehicles. The behavior of the host vehicle depends on its longitudinal model and longitudinal controller. The auto regressive exogenous computation model is explained with reference to the above equations, i.e. mathematical formula 1 to mathematical formula 3.

[0078] The estimated dynamic model is then validated in a step 53 of validating the dynamic model, which is configured to validate the dynamic model from the actual speed V(t) of the vehicle at time t and the previous speed V(t-1) of the vehicle at the previous time t-1. For example, to validate the dynamic model, the error of the ARX computation model is compared to a threshold. If the error is below said threshold, the dynamic model is validated.

[0079] The method 50 for predicting the trajectory of the host vehicle 10 further comprises a step 54 of predicting the movement of the adjacent vehicles from the dynamic model validated in step 53 and the initial position of the adjacent vehicles.

[0080] The method 50 for predicting the trajectory of the host vehicle 10 further comprises a step 55 of predicting the movement of the host vehicle 10 from the movement prediction provided in step 53 and the information originating from the proprioceptive sensors C of the host vehicle 10.

[0081] In a step 56, the lane change of the host vehicle 10 is then determined from the prediction of the movement of the host vehicle and the adjacent vehicles, the overall trajectory T of the host vehicle 10 and the information originating from the road map Cart on which the host vehicle is driving.

[0082] The step 56 of determining the lane change of the host vehicle allows to evaluate a suitable time window that allows the host vehicle 10 to safely change lane.

[0083] In a step 57, a lane change instruction for the host vehicle 10 is sent to the module 24 for performing the lane change of the host vehicle 10.

[0084] In case the host vehicle cannot change lane, the host vehicle 10 is informed that a modification of a parameter such as in particular its speed is required.

[0085] The method 50 for predicting the trajectory of the host vehicle is repeated until a lane change can be found.

[0086] With the present invention, it is possible to reliably predict in real time the movement of the traffic flow in the lane adjacent to the lane in which the host vehicle is traveling and to allow the position of the host vehicle after it has changed lanes from the main lane to the adjacent lane to be predicted.

[0087] Furthermore, the present invention allows the constraints of the road on which the host vehicle is traveling to be taken into account.

[0088] This system and method for predicting the trajectory of the host vehicle allow the traffic flow to flow more smoothly without compromising the safety of the host vehicle and the surrounding vehicles.

Claims

1. A method (50) for predicting a trajectory of a host vehicle (10) driving on a main lane (1), wherein determining a lane change of the host vehicle from the main lane (1) to the adjacent lane (2) as a function of an estimation of the dynamic behavior of a group of vehicles driving on the adjacent lane (2), said group of vehicles (7) comprising at least one main vehicle located in the vicinity of the host vehicle and a secondary vehicle located behind said host vehicle, wherein position (P), orientation (O) and speed (V) information of the host vehicle (10) and of the vehicles (3, 4, 5, 6) of the group of vehicles (7) are collected and a dynamic model of each pair of successive vehicles driving on the adjacent lane (2) is built from the collected information, wherein the dynamic model of each pair of successive vehicles is obtained by determining a second order transfer function corresponding to the behavior of the host vehicle with respect to each pair of considered successive vehicles using an auto regressive exogenous computation model, wherein the behavior of the host vehicle depends on its longitudinal model and on a longitudinal controller.

2. The method (50) of claim 1, wherein, The established dynamic model is validated by comparing the error e(t) of the auto regressive exogenous computation model with a threshold depending on the actual speed V(t) of each vehicle at time t and on the speed V(t-1) of said vehicle at the previous time t-1.

3. The method (50) of claim 2, wherein, The movements of the adjacent vehicles are predicted from the validated dynamic model and from the initial positions of said vehicles, the movement of the host vehicle (10) is predicted from the prediction of the movements of the adjacent vehicles and from the information originating from the proprioceptive sensors (C) of the host vehicle (10), and the lane change of the host vehicle (10) is determined from said prediction of the movements of the host vehicle and of the adjacent vehicles, the global trajectory (T) of the host vehicle (10).

4. The method (50) of any of the preceding claims, wherein, The steps of the method (50) are repeated until a lane change is found.

5. A system (12) for predicting a trajectory of a host vehicle (10) driving on a main lane (1), the system being configured to determine a lane change of the host vehicle from the main lane (1) to an adjacent lane (2) as a function of an estimation of the dynamic behavior of a group of vehicles driving on the adjacent lane (2), said group of vehicles (7) comprising at least one main vehicle located in the vicinity of the host vehicle and a secondary vehicle located behind said host vehicle, wherein said system (12) comprising: - a module (13) for collecting position (P), orientation (O) and speed (V) information of the host vehicle (10) and of the vehicles (3, 4, 5, 6) of the group of vehicles (7); - a module (14) for estimating a dynamic model of each pair of successive vehicles driving on the adjacent lane (2) from the collected information; - a module (16) for validating the established dynamic model by comparing the error e(t) of the auto regressive exogenous computation model with a threshold depending on the actual speed V(t) of each vehicle at time t and on the previous speed V(t-1) of said vehicle at the previous time t-1; - a module (18) for predicting the movements of the adjacent vehicles from the validated dynamic model and from the initial positions of said vehicles; - a module (20) for predicting the movement of the host vehicle (10) from the prediction of the movements of the adjacent vehicles and from the information originating from the proprioceptive sensors (C) of the host vehicle (10); and - a module (22) for determining a lane change of the host vehicle (10) from the main lane (1) to the adjacent lane (2) as a function of the prediction of the movements of the host vehicle and of the adjacent vehicles, the global trajectory (T) of the host vehicle (10). - a module for determining a lane change of the host vehicle (10) as a function of said predictions of movements of the host vehicle and of the neighboring vehicles and of the global trajectory (T) of the host vehicle (10).

6. A host motor vehicle comprising a system (11) for perception and a system (12) for predicting a trajectory of the host vehicle as claimed in claim 5.

Images