Method and system for learning free space available for a vehicle to travel

Abstract

The invention relates to a method for ascertaining free space (F) in which a self-propelled vehicle (2) can travel by means of a plurality of free space segments (S0-S3), in which the free space in which the self-propelled vehicle (2) can travel without a collision is checked on the basis of the free space segments.

Description

Technical Field

[0001] This invention relates to the field of vehicle assistance systems. In particular, this invention relates to a method and system for determining the free space in which a vehicle can move. Background Technology

[0002] Systems for planning the free space in which vehicles can travel without collisions are generally known. In particular, systems are known that iteratively determine the free space for collision-free travel over the entire prediction time period, based on the estimated trajectories of the vehicle itself and a third vehicle (hereinafter referred to as the third vehicle trajectory).

[0003] The problem here is that the calculation of free space that allows for collision-free driving is very computationally expensive because it is based on a large number of objects that may be moving in the same way, making the iterative calculation of free space very complex. Summary of the Invention

[0004] Based on this, the objective of the present invention is to provide a method for obtaining the free space in which a vehicle can drive, which enables reliable and computationally efficient calculation of the free space for collision-free driving.

[0005] This task is solved by a method having the features of independent patent claim 1. Preferred embodiments are the subject of the dependent claims. A system for determining the free space in which the vehicle can travel is the subject of co-claim 10.

[0006] According to the first aspect, a method for determining the free space in which a vehicle can travel without collisions is disclosed. The method includes the following steps:

[0007] First, estimate the vehicle's trajectory and the trajectory of at least one third vehicle / other vehicle within the prediction time period. The vehicle's trajectory predicts the vehicle's movement within the prediction time period, and the third vehicle's trajectory predicts the third vehicle's movement within the prediction time period.

[0008] Subsequently, a first free space is established. This first free space forms the basic framework for the free space to be known, hereinafter also referred to as the second free space or final free space, which is passed to the trajectory planner for planning a collision-free trajectory for the vehicle. The establishment of the first free space is preferably performed without considering third-party objects, particularly without considering the predicted positions of moving objects, especially third vehicles.

[0009] Subsequently, based on the estimated vehicle trajectory, the future position of the vehicle at different time points within the prediction period is determined. In other words, the prediction period is discretized in time, thereby defining multiple time points within the prediction period. Based on the estimated vehicle trajectory, the position of the vehicle at these time points is determined. These time points can be selected equidistantly within the prediction period.

[0010] Subsequently, based on the known future position of the vehicle, the first free space is divided into free space segments. Continuous free space segments are preferably directly connected to each other. Each free space segment preferably has a length extending in the direction of travel, which is greater than the length of the vehicle.

[0011] The future positions of the third vehicle are estimated at different points in time within the prediction period based on the trajectory of the third vehicle. Here, these points in time correspond to those used to determine the position of the primary vehicle.

[0012] Subsequently, for each corresponding time point within the predicted time period, the future position of the third vehicle at that time point is compared with the free space segment where the primary vehicle is located at that time point. In particular, it is determined whether the third vehicle approaches the corresponding free space segment in a manner below a safe threshold, or whether it is at least partially located within the corresponding free space segment.

[0013] Subsequently, if the estimated future position of the third vehicle at the corresponding time point indicates that the third vehicle approaches the free space segment where the primary vehicle is located at that time point in a manner below a safety threshold, or that the third vehicle at least partially intrudes into the free space segment, then at least one free space segment is restricted. Here, "restriction" is specifically understood to reduce the area of ​​the free space segment by modifying its lateral / lateral boundaries, thereby reducing the width of the free space segment extending laterally in the direction of travel. This results in a free space segment that is at least partially modified.

[0014] Finally, a second final free space is formed by merging the free space segments, wherein at least one of the free space segments is modified.

[0015] This method offers the following technical advantages: by dividing the free space spanning the entire prediction time period into multiple free space segments based on the vehicle's position selection, and checking for each free space segment whether a third vehicle intrudes into that free space segment at the corresponding time point when the vehicle is already within that segment, computational complexity is significantly reduced. Furthermore, by considering the vehicle's motion when creating the free space segments, it is possible to distinguish whether the predicted motion of the third vehicle is relevant to the free space planning.

[0016] According to one embodiment, the first free space is constructed in a tubular shape and extends along the estimated trajectory of the vehicle. Thus, the first free space forms a corridor in which the vehicle can move substantially during the predicted time period, and this corridor can serve as a basis for determining whether the first free space needs to be restricted due to a third vehicle or other obstacle.

[0017] According to one embodiment, each free space segment includes the entire width of the first free space and also includes a portion of the length of the first free space. Here, the portion of length extends along the vehicle's direction of travel, and the width of the first free space is measured transversely to this direction of travel. In other words, the first free space is segmented only along the vehicle's direction of travel, and not transversely to this direction. Thus, the first free space can be divided into free space segments, which can be allocated to different time points within a predicted time period and to different positions of the vehicle within that predicted time period.

[0018] According to one embodiment, the free space segments are formed centered on the position of the vehicle at discrete time points within the predicted time period. Preferably, the free space segment has a length greater than the length of the vehicle. Thus, the free space segment in which the vehicle is located at a given time point can be determined. Similarly, the position of a third vehicle can be assigned to the free space segment in which the vehicle is located at that time point, based on the given time point. Therefore, the free space for collision-free driving can be determined based on the relevant area of ​​the first free space.

[0019] According to one embodiment, limiting at least one free space segment includes at least partially reducing the width of the free space segment, i.e., the free space segment is defined laterally in such a way that it does not overlap with a third vehicle or a safety area allocated to the third vehicle.

[0020] According to one embodiment, a safety zone is formed around a third vehicle. If the estimated future position of the third vehicle at a certain point in time indicates that its safety zone at least partially encroaches on the free space segment in which the third vehicle is located at that point in time, then that free space segment is restricted or clipped. The safety zone provides a safety buffer that allows the second free space created by merging free space segments to have higher collision-free safety.

[0021] According to one embodiment, the size of the safe zone is selected based on the quality of the third vehicle trajectory. Specifically, when the quality of the third vehicle trajectory is low, i.e., when the third vehicle trajectory has relatively high uncertainty, the safe zone can be selected to be larger than the safe zone when the quality of the third vehicle trajectory is high. This can significantly improve the safety of assisted or autonomous driving operations.

[0022] According to one embodiment, the length of the free space segment is selected based on the quality of the vehicle trajectory and / or the quality of the third vehicle trajectory. Specifically, when the quality of the vehicle trajectory and / or the third vehicle trajectory is low, the length of the free space segment is selected to be greater than the length of the free space segment when the vehicle trajectory and / or the third vehicle trajectory have higher quality. This prevents situations where the vehicle and / or third vehicle position estimates are inaccurate, resulting in insufficient or no free space constraint due to overly fine discretization of the free space.

[0023] According to one embodiment, the length of the free space segment is selected based on the vehicle's speed. Specifically, the length of the free space segment can increase as the vehicle's speed increases. This improves the safety of the planned free space.

[0024] According to another aspect, a system for determining the free space in which a vehicle can drive is disclosed. The system has a computing unit configured to perform the following steps:

[0025] - Estimate the trajectory of the vehicle itself and the trajectory of at least one third vehicle for the predicted time period.

[0026] - Establish the first free space;

[0027] - Based on the estimated vehicle trajectory, the future position of the vehicle at different points in time within the predicted time period can be determined;

[0028] - Based on the known future location of the vehicle, the first free space is divided into free space segments;

[0029] - Estimate the future location of the third vehicle at different points in time within the prediction period based on the trajectory of the third vehicle;

[0030] - For each time point within the predicted time period, compare the future position of the third vehicle at the corresponding time point with the free space segment where the own vehicle is located at that time point;

[0031] - If the estimated future position of the third vehicle at the corresponding time point indicates that the third vehicle approaches the free space segment where the vehicle is located at that time point in a manner below a safety threshold, or the third vehicle at the corresponding time point at least partially intrudes into the free space segment where the vehicle is located at that time point, then at least one free space segment is restricted.

[0032] - A second free space is formed by merging free space segments.

[0033] In the context of this invention, the expressions “approximately,” “substantially,” or “about” indicate a deviation of + / -10%, preferably + / -5%, from the corresponding precise value, and / or a deviation in a form that is not significant to the function.

[0034] The improvements, advantages, and applicability of the present invention are also derived from the following description and drawings of the embodiments. Here, all described and / or illustrated features, either individually or in any combination thereof, are the subject matter of the invention, regardless of their combination in the claims or their reference relationships. The content of the claims is also an integral part of the specification. Attached Figure Description

[0035] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. Wherein shown:

[0036] Figure 1 An exemplary schematic diagram of a system for determining the free space in which collision-free driving is possible is shown;

[0037] Figure 2 An exemplary and schematic top view of a vehicle's driving situation is shown, in which free space planning is used based on multiple free space segments; and

[0038] Figure 3 An exemplary block diagram illustrates the method steps for determining a free space where collision-free driving is possible. Detailed Implementation

[0039] Figure 1 A system 1 for determining the free space F in which a vehicle 2 can drive is illustrated, by way of example and schematic. System 1 is preferably disposed within the vehicle 2. System 1 includes a computing unit 4. The computing unit 4 receives information from sensing devices 5 of the vehicle 2. The sensing devices 5 are configured to detect the surrounding environment of the vehicle 2, such that the information provided by the sensing devices 5 includes environmental information. The sensing devices 5 may, for example, include one or more radar sensors, ultrasonic sensors, one or more lidar sensors, and / or one or more cameras.

[0040] The computing unit 4 is configured to determine the free space F in which the vehicle 2 can drive without collisions based on environmental information. The free space F can be defined, for example, by left and right boundary lines. The left and right boundary lines can be formed, for example, by broken lines / connecting lines that link multiple support points.

[0041] The free space F obtained by the computing unit 4 can be passed to the trajectory planner 6, which uses this free space F to determine the trajectory of the vehicle 2.

[0042] Figure 2 This schematically illustrates traffic conditions in free space where the system can be applied to determine where vehicle 2 can travel without collisions.

[0043] In the illustrated embodiment, vehicle 2 travels along the direction of travel FR on a road having multiple lanes for that direction of travel. Vehicle 2 is exemplarily traveling in the right-hand lane. An obstacle H is present in this lane; in the illustrated example, the obstacle is a stationary vehicle. Vehicle 2 must go around the obstacle H, and for this purpose, vehicle 2 performs a lane change. Therefore, the obstacle H restricts the free space F for collision-free travel.

[0044] Furthermore, the free space F, which allows for collision-free driving, is also affected by a third vehicle 3 located in the left lane, which, for example, moves at a higher speed and performs a lane change to the middle lane.

[0045] The knowledge of the free space F in which vehicle 2 can travel without collision is obtained by: setting a first free space based on the vehicle's expected movement during the predicted time period; dividing the first free space into multiple consecutive free space segments S0–S3 in the driving direction FR; and locally restricting the first free space within the corresponding free space segment S0–S3 when obstacles H, third vehicles 3, etc., are present. In other words, instead of performing free space planning for the entire predicted time period—for example, the predicted time period can be 0s to 5s—the first free space is segmented, giving the spatial area in which vehicle 2 will move during the predicted time period. Subsequently, based on these free space segments S0–S3, it is checked whether obstacles, third vehicles, etc., exist in the corresponding free space segments S0–S3 at the time point when vehicle 2 is in the corresponding free space segment S0–S3, so as to focus the free space planning on the relevant area and thereby limit the required computational resources.

[0046] The following is based on Figure 2 The method will be described in more detail.

[0047] First, the trajectory ET of vehicle 2 within the predicted time period is estimated. This trajectory ET is preferably a coarse estimate, derived based on information from previous predicted time periods. Similarly, the trajectory DT of a third vehicle 3 is obtained from vehicle 2, wherein the third vehicle 3 is detected, for example, by the sensor 5 of vehicle 2. When the third vehicle trajectory DT is obtained, quality information, which is a measure of the estimated quality of the third vehicle trajectory DT, can also be provided. Therefore, for example, if the third vehicle 3 has been detected by the sensor 5 of vehicle 2 for a longer period, the quality of the third vehicle trajectory DT can be higher than if the third vehicle 3 was detected only recently.

[0048] Subsequently, a first free space is established. This is preferably based on the estimated trajectory ET of vehicle 2. The first free space can be constructed in a tubular shape and has a length and a width, wherein the length extends along the direction of travel and at least substantially corresponds to the distance traveled by vehicle 2 within the predicted time period. The width of the first free space extends laterally, and in particular perpendicularly to, the direction of travel FR of vehicle 2. Thus, the first free space provides a corridor in which vehicle 2 can move substantially within the predicted time period without considering objects restricting the free space.

[0049] Subsequently, based on the vehicle trajectory ET of vehicle 2 and the discrete time points t0–t3 within the prediction time period, the future positions of vehicle 2 are determined. This position determination only involves the position of vehicle 2 in the driving direction FR, that is, the lateral position of vehicle 2 does not need to be considered.

[0050] Figure 2 The diagram shows the position of vehicle 2 at time points t0–t3. Based on these positions, the first free space can be segmented, forming free space segments S0–S3. Free space segments S0–S3 extend along the driving direction FR, each passing through a portion of the first free space. Laterally to the driving direction, these free space segments are defined by the boundaries of the first free space, such as… Figure 2 As shown by the thick lines, free space segments S0–S3 are assigned to time points t0–t3 within the predicted time period, for example, free space segment S0 is assigned to time point t0, free space segment S1 is assigned to time point t1, and so on. Viewed from the direction of travel FR, each free space segment S0–S3 is arranged centered around the position of vehicle 2 at the corresponding time point. Free space segments S0–S3 have a length extending along the direction of travel FR, which is greater than the length of vehicle 2. The widths of free space segments S0–S3 correspond to the widths of the first free space.

[0051] Subsequently, based on the object or obstacle H that falls into the corresponding free space segment S0–S3 at time points t0–t3, one or more free space segments S0–S3 are restricted. In other words, the object or obstacle H is used to restrict the free space segment S0–S3 only if the vehicle 2 also falls at least partially into the corresponding free space segment S0–S3 at time points t0–t3. Here, a safety zone S can be formed around the object or obstacle H. This safety zone S can, for example, be larger than the base area of ​​the object or obstacle H, thereby forming a safety buffer to account for, for example, position estimation of the object or obstacle H with uncertainty. Figure 2 In the diagram, the safe area S surrounding the obstacle H and the third vehicle 3 is indicated by a rectangular outline surrounding the obstacle H and the third vehicle 3.

[0052] As mentioned earlier, the future motion of the third vehicle 3 within the predicted time period is described by the third vehicle trajectory DT. Based on this third vehicle trajectory DT, the position of the third vehicle 3 at time points t0–t3 can be determined.

[0053] If a third vehicle 3, an obstacle H, or another object also intrudes into or is completely located within the corresponding free space segment S0–S3 where vehicle 2 is located at time t0–t3 (e.g., obstacle H and third vehicle 3 intrude into free space segment S2 at time t2), then the free space segment S0–S3 is correspondingly restricted or clipped so that the object or obstacle H, along with its necessary safety area S, is located outside the corresponding free space segment S0–S3.

[0054] like Figure 2 As shown, neither obstacle H nor the third vehicle 3 is within free space segments S0 and S1. Therefore, these free space segments remain unrestricted.

[0055] At time point t2, vehicle 2 is in free space segment S2, and at this time, obstacle H and third vehicle 3 are also located in free space segment S2. Therefore, the boundary restricting the width of free space segment S2 is changed so that third vehicle 3 and obstacle H, along with their necessary safety zone S, are located outside free space segment S2. This is... Figure 2 The angled shape of the lateral boundary of the free space segment S2 (thick line) is schematically represented in the diagram. A similar approach is used for the free space segment S3 and time point t3.

[0056] After modifying one or more free space segments, a second free space is formed by merging free space segments S0–S3. This second free space constitutes the free space in which collision-free driving is possible within the predicted time period, and this free space can be used by the trajectory planner 6 to calculate a collision-free trajectory.

[0057] The length of the free space segment S0–S3, measured along the direction of travel FR of vehicle 2, can be selected based on the estimated mass of the vehicle trajectory ET and / or the mass of the third vehicle trajectory DT. In particular, when the masses of the vehicle trajectory ET and / or the third vehicle trajectory DT are low, the length of the free space segment S0–S3 can be selected to be greater than the length when the masses of the vehicle trajectory ET and / or the third vehicle trajectory DT are high, in order to improve safety when determining the free space where collision-free driving is possible.

[0058] Alternatively or additionally, the length of the free space segment S0–S3 measured along the driving direction FR of vehicle 2 can depend on the speed of vehicle 2. In particular, the length of the free space segment S0–S3 can increase as the speed of vehicle 2 increases.

[0059] As previously mentioned, a safety zone S can be established around the third vehicle 3. This safety zone is used to restrict the free space segment S0–S3 when the third vehicle 3, along with its safety zone S, enters the free space segment S0–S3. The size of the safety zone S, particularly the distance from the edge of the safety zone to the outer contour line of the third vehicle 3, can be determined based on the mass of the third vehicle trajectory DT. In particular, when the prediction of the third vehicle trajectory DT is uncertain and thus the mass of the third vehicle trajectory DT is low, the safety zone S can be selected to be larger than the safety zone in the case of high mass, thereby improving safety when determining the free space for collision-free driving.

[0060] Figure 3 A schematic block diagram illustrating the steps of a method for determining the free space in which a vehicle can drive is shown.

[0061] First, estimate the trajectory of the vehicle and the trajectory of at least one third vehicle for the predicted time period (S10).

[0062] In addition, a first free space (S11) is established.

[0063] Subsequently, based on the estimated vehicle trajectory, the future position of the vehicle at different time points within the predicted time period is obtained (S12).

[0064] Based on the known future location of the vehicle, the first free space is divided into free space segments (S13).

[0065] Estimate the future position of the third vehicle at different time points within the predicted time period based on the trajectory of the third vehicle (S14).

[0066] Subsequently, for each time point within the predicted time period, the future position of the third vehicle at the corresponding time point is compared with the free space segment where the own vehicle is located at that time point (S15).

[0067] Subsequently, if the estimated future position of the third vehicle at the corresponding time point indicates that the third vehicle approaches the corresponding free space segment where the vehicle is located at that time point in a manner below a safety threshold (which can be pre-given by a safety area, for example), or the third vehicle at least partially intrudes into the free space segment, then at least one free space segment is restricted or clipped (S16).

[0068] Finally, a second free space (S17) is formed by merging the free space segments.

[0069] The invention has been described above based on embodiments. It should be understood that many changes and modifications can be made without departing from the scope of protection defined by the patent claims.

[0070] List of reference numerals

[0071]

Claims

1. A method for determining the free space (F) in which a vehicle (2) can move, wherein, The method includes the following steps: - Estimate the vehicle trajectory (ET) of the vehicle (2) and the third vehicle trajectory (DT) of at least one third vehicle (3) for the predicted time period (S10). - Set the first free space (S11); - Based on the estimated vehicle trajectory (ET), the future position of the vehicle (2) at different time points (t1–t3) within the predicted time period is known (S12). - Based on the known future position of the vehicle (2), the first free space is divided into free space segments (S0–S3) (S13). - Estimate the future location of the third vehicle (3) at different time points within the prediction time period based on the third vehicle trajectory (DT) (S14). - For each time point (t0–t3) within the predicted time period, compare the future position of the third vehicle (3) at the corresponding time point (t0–t3) with the free space segment (S0–S3) where the self vehicle (2) is located at that time point (t0–t3) (S15). - If the estimated future position of the third vehicle (3) at the corresponding time point (t0–t3) indicates that the third vehicle (3) approaches the free space segment (S0–S3) where the vehicle (2) is located at that time point in a manner below a safety threshold, or the third vehicle (3) at least partially intrudes into the free space segment (S0–S3) where the vehicle (2) is located at that time point (t0–t3), then at least one free space segment (S0–S3) is restricted (S16). - A second free space (S17) is formed by merging the free space segments (S0–S3).

2. The method according to claim 1, characterized in that, The first free space is constructed in a tubular shape and extends along the estimated vehicle trajectory (ET).

3. The method according to claim 1 or 2, characterized in that, The free space segments (S0–S3) each include the entire width of the first free space and a portion of the length of the first free space.

4. The method according to any one of the preceding claims, characterized in that, The free space segments (S0–S3) are formed by centering on the position of the vehicle (2) at discrete time points (t0–t3) in the prediction time period.

5. The method according to any one of the preceding claims, characterized in that, The restriction on at least one free space segment (S0–S3) includes at least locally reducing the width of the free space segment (S0–S3).

6. The method according to any one of the preceding claims, characterized in that, A safety zone (S) is formed around the third vehicle (3), and at least one free space segment (S0–S3) is restricted if the estimated future position of the third vehicle (3) indicates that the safety zone (S) at least partially encroaches on the corresponding free space segment (S0–S3).

7. The method according to claim 6, characterized in that, The size of the safe zone (S) is selected based on the quality of the third vehicle trajectory (DT).

8. The method according to any one of the preceding claims, characterized in that, The length of the free space segment (S0–S3) is selected based on the quality of the voluntary vehicle trajectory (ET) and / or the quality of the third vehicle trajectory (DT).

9. The method according to any one of the preceding claims, characterized in that, The length of the free space segment (S0–S3) is selected based on the speed of the vehicle (2).

10. A system for knowing the free space (F) in which a vehicle (2) can move, wherein, The system (1) has a computing unit (4) configured to perform the following steps: - Estimate the trajectory of the vehicle (2) and the trajectory of at least one third vehicle (3) for the predicted time period (DT). - Establish the first free space; - Based on the estimated vehicle trajectory (ET), the future position of the vehicle (2) at different time points (t0–t3) within the predicted time period is known; - Based on the known future position of the vehicle (2), the first free space is divided into free space segments (S0–S3). - Estimate the future location of the third vehicle (3) at different time points (t0–t3) within the prediction time period based on the third vehicle trajectory (DT); - For each time point (t0–t3) within the predicted time period, compare the future position of the third vehicle (3) at the corresponding time point (t0–t3) with the free space segment (S0–S3) where the vehicle (2) is located at that time point (t0–t3); - If the estimated future position of the third vehicle (3) at the corresponding time point (t0–t3) indicates that the third vehicle (3) approaches the free space segment (S0–S3) where the vehicle (2) is located at that time point in a manner below a safety threshold, or if the third vehicle (3) at least partially intrudes into the free space segment (S0–S3) where the vehicle (2) is located at that time point (t0–t3), then at least one free space segment (S0–S3) is restricted. - A second free space is formed by merging the free space segments (S0–S3).

Images