Path planning for autonomous and semi-autonomous vehicles

Abstract

The present invention relates to path planning for autonomous and semi-autonomous vehicles. A method (400, 500) for path planning for an autonomous or semi-autonomous vehicle (1) comprises obtaining a drivable area of ​​the vehicle's surroundings and generating a path (2) within the drivable area for a time step t based on a predetermined set of constraints and a predetermined set of characteristics of the path. The predetermined set of constraints includes at least one constraint (7, 8) based on the vehicle's current posture.

Description

Technical Field

[0001] The present disclosure relates to autonomous driving (AD) and advanced driver assistance systems (ADAS). More specifically, the present disclosure relates to path planning for a vehicle within a drivable area. Background Art

[0002] Many vehicles today have a variety of driver support features in the form of advanced driver assistance systems (ADAS). Furthermore, many of these support features form the basis for current and future autonomous driving (AD) features. Examples of ADAS features or functions include lane departure warning systems, lane centering, lane keeping assist, driver assistance, lane change assist, parking sensors, pedestrian protection systems, blind spot monitors, adaptive cruise control (ACC), anti-lock braking systems, and the like. These features supplement the vehicle's traditional driver controls with one or more warnings or automated actions in response to certain scenarios.

[0003] Over the past few years, the development of autonomous vehicles has exploded, with many different approaches being developed. Both autonomous driving (AD) and advanced driver assistance systems (ADAS) (i.e., semi-autonomous driving) are currently being developed across many different technology areas within these fields. One such area is path planning, i.e., how to plan a path so that the vehicle can be safely maneuvered accurately and in a manner consistent with the comfort of the vehicle's occupants, i.e., path or trajectory planning.

[0004] Typically, conventional path planning systems generate a target path for an autonomous vehicle from a given drivable area, which is typically provided by a perception system or module. The target path hypothesis is sent to the vehicle controller, which calculates the steering angle and causes the vehicle to travel along the path. A schematic diagram of a conventional path planning scheme is illustrated in Figure 1. The vehicle's path planning module generates a target path 2' for vehicle 1'. However, because the path tracking controller commands the steering angle to converge with the target path 2', the initial vehicle position is far from the generated target path 2', which results in an uneven vehicle motion (as indicated by the dotted arrow 3'). The unevenness of motion 3' can be traced back to two stages. First, a large steering angle is commanded immediately after automation is activated, and second, after reaching path 2', due to the large travel angle that had been increased in the previous stage, vehicle 1' deviates from path 2' instead of converging with it. Therefore, vehicle motion 3' undergoes several such deviations until it completely converges with path 2'. In some cases where the initial vehicle offset is very large, the magnitude of the overshoot may increase over time, and vehicle 1' may therefore fail to rejoin path 2'. For example, this may be a problem in automated valet parking applications, where the user parks in an arbitrary posture before initiating automated parking, but the target path is optimized around the center of the drivable area.

[0005] Although currently known solutions may produce adequate results in terms of safe path planning, there is always a need for improvements in the art, particularly in terms of comfort and overall user experience. Summary of the Invention

[0006] It is therefore an object of the present disclosure to provide a method for path planning of an autonomous or semi-autonomous vehicle, a computer-readable storage medium, a control device and a vehicle comprising such a control device that alleviate all or at least some of the drawbacks of currently known solutions.

[0007] This object is achieved by means of a method for path planning for an autonomous or semi-autonomous vehicle, a computer-readable storage medium, a control device and a vehicle comprising such a control device as defined in the appended claims. The term exemplary should in this context be understood as an example, instance or illustration.

[0008] According to a first aspect of the present disclosure, a method for path planning for an autonomous or semi-autonomous vehicle is provided. The method comprises obtaining a drivable area of ​​the vehicle's surroundings, and generating a path within the drivable area for a time step t based on a predetermined set of constraints and a predetermined set of characteristics of the path. The predetermined set of constraints includes at least one constraint based on the vehicle's posture.

[0009] The proposed method enables smoother and more stable path planning modules for autonomous and semi-autonomous vehicles compared to traditional schemes. More specifically, the proposed method mitigates the risk of failed path convergence trajectories in situations where the vehicle's initial pose deviates from the target path (such as, for example, in automated valet parking applications).

[0010] Furthermore, the present inventors have realized that by forming constraints that depend on the vehicle's current posture when generating a target path for the vehicle, the risk of failing to meet the generated path due to a non-optimal initial vehicle position can be reduced. Specifically, the present inventors have realized that by using constraints in the form of two side boundaries (the two side boundaries extend a predetermined length in front of the vehicle at a predetermined distance from the vehicle's longitudinal center axis), simple and computationally efficient path optimization can be achieved.

[0011] According to an exemplary embodiment of the present disclosure, generating a path for the time step t within the drivable area includes the following steps:

[0012] Calculating a cost function, wherein said cost function is a mathematical representation of said predetermined set of characteristics of said path,

[0013] Calculating the constraint set, wherein the constraint set further includes: at least one constraint based on a boundary of the drivable area and at least one constraint based on vehicle characteristics,

[0014] Solve the optimization problem based on the calculated cost function and the calculated constraint set,

[0015] The path is formed based on the solved optimization problem.

[0016] In other words, path planning is formulated as a quadratic programming problem.

[0017] Accordingly, by using this formula, the global optimal solution can be derived in real time using an efficient but traditional optimization algorithm.

[0018] Moreover, according to another exemplary embodiment of the present disclosure, the vehicle has a longitudinal axis that intersects a predetermined center point of the vehicle, and the at least one constraint based on the current pose of the vehicle includes two lateral boundaries that extend parallel to the longitudinal axis of the vehicle at a first predetermined lateral distance from the longitudinal axis. The predetermined center point of the vehicle can be interpreted as a reference point based on which path tracking control is performed (e.g., the center of the vehicle's true center of mass axis). The lateral boundaries provide a simple and effective means to impose constraints related to the vehicle pose in path planning applications.

[0019] More specifically, according to yet another exemplary embodiment, the area between the two side boundaries defines an allowed region within which at least a portion of the path of the time step t is allowed to be formed.

[0020] According to a second aspect of the present disclosure, there is provided a (non-transitory) computer-readable storage medium storing one or more programs configured to be executed by one or more processors of a vehicle control system, the one or more programs comprising instructions for performing a method according to any one of the embodiments disclosed herein. Together with this aspect of the present disclosure, similar advantages and preferred features as previously discussed in the first aspect of the present disclosure are provided.

[0021] As used herein, the term "non-transitory" is intended to describe computer-readable storage media (or "memory") that excludes propagating electromagnetic signals, but is not intended to otherwise limit the type of physical computer-readable storage device encompassed by the phrase computer-readable medium or memory. For example, the terms "non-transitory computer-readable medium" or "tangible memory" are intended to encompass various types of storage devices that do not necessarily store information permanently, including, for example, random access memory (RAM). Program instructions and data stored in a non-transitory form on a tangible computer-accessible storage medium may be further transmitted by a transmission medium or signal, such as an electrical, electromagnetic, or digital signal, which may be carried via a communication medium such as a network and / or wireless connection. Thus, as used herein, the term "non-transitory" is a limitation on the medium itself (e.g., tangible rather than a signal), rather than a limitation on the permanence of the data storage (e.g., RAM vs. ROM).

[0022] Further, according to a third aspect of the present disclosure, a control device for path planning of an autonomous or semi-autonomous vehicle is provided. The control device comprises a control circuit (which may also be referred to as one or more processors) and an associated storage device. The control circuit is configured to: obtain a drivable area of ​​the vehicle's surroundings, and generate a path for a time step t within the drivable area based on a predetermined set of characteristics of the path according to a predetermined set of constraints. Moreover, the predetermined set of constraints includes at least one constraint based on the current posture of the vehicle. Together with this aspect of the disclosure, similar advantages and preferred features as previously discussed in the first aspect of the disclosure are proposed.

[0023] According to a fourth aspect of the present disclosure, a vehicle is provided that includes a perception system comprising at least one sensor for monitoring the vehicle's surroundings. The vehicle further includes a control device according to any of the embodiments disclosed herein. Similar advantages and preferred features as previously discussed in connection with the first aspect of the present disclosure are provided in conjunction with this aspect of the present disclosure.

[0024] Further embodiments of the present disclosure are defined in the dependent claims. It should be emphasized that when used in this specification, the term "comprises" is used to specify the presence of a stated feature, integer, step, or component. It does not exclude the presence or addition of one or more other features, integers, steps, components, or groups thereof.

[0025] These and other features and advantages of the present disclosure will be further elucidated with reference to the embodiments described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Further objects, features, and advantages of the embodiments of the present disclosure will become apparent from the following detailed description made with reference to the accompanying drawings, in which:

[0027] FIG1 is a schematic top view of a vehicle traveling on a road section according to an example of the prior art.

[0028] Figure 2 is a schematic top view of a vehicle having a control device for path planning according to an exemplary embodiment of the present disclosure.

[0029] Figure 3 (a) to Figure 3 (b) is a top schematic diagram illustrating a vehicle having a control device for path planning for a valet parking application according to an exemplary embodiment of the present disclosure.

[0030] Figure 3 (c) to Figure 3 (d) is a schematic top view of a vehicle with a valet parking application according to a prior art example.

[0031] Figure 4 is a schematic block diagram representation of a method for path planning for an autonomous or semi-autonomous vehicle according to an embodiment of the present disclosure.

[0032] Figure 5 is a schematic flow chart representation of a method for path planning for an autonomous or semi-autonomous vehicle according to an embodiment of the present disclosure.

[0033] Figure 6 is a schematic side view of a vehicle having a control device for path planning of an autonomous or semi-autonomous vehicle according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0034] It will be understood by those skilled in the art that the steps, services, and functions explained herein can be implemented using separate hardware circuits, using software in combination with programmed microprocessors or general-purpose computers, using one or more application-specific integrated circuits (ASICs), and / or using one or more digital signal processors (DSPs). It will also be understood that when the present disclosure is described in terms of a method, it can also be embedded in one or more processors and one or more memories coupled to the one or more processors, wherein the one or more memories store one or more programs that, when executed by the one or more processors, perform the steps, services, and functions disclosed herein.

[0035] In the following description of exemplary embodiments, identical reference numerals refer to identical or similar components.A vehicle is to be understood in this context as a road vehicle (such as a car, bus, truck, construction vehicle, etc.).

[0036] Generally, there are two main approaches to path planning problems. The first is a sample-based approach, in which the drivable area is discretized into multiple units or nodes. The target path is found by performing a type of graph-based search from the starting area to the ending area. Another approach is to use a cost function and a set of constraints to solve the optimization problem. The cost function represents the path characteristics desired by the user, and the constraints include restrictions on the drivable area within which the path is allowed to exist. Accordingly, the optimal path is found as the one that can minimize the cost function. The present disclosure is particularly related to the latter approach.

[0037] Figure 2 is a schematic top view of a vehicle 1 traveling on a road section 4 (here in the form of a lane on a road) according to an exemplary embodiment of the present disclosure. Figure 2 A method or process for path planning for an autonomous or semi-autonomous vehicle is at least generally described in accordance with an exemplary embodiment of the present disclosure. The illustrated vehicle 1 is therefore to be interpreted as an autonomous vehicle or a semi-autonomous vehicle (i.e., a vehicle equipped with a suitable advanced driver assistance system) having a control device for path planning in accordance with an exemplary embodiment of the present disclosure. A path may in this context be a sequence of vehicle configurations that begins and ends with a boundary configuration. These boundary configurations may also be referred to as start and end points. In other words, a planned path may be understood as a series of positions or configurations that are planned to be assumed by the vehicle 1 within a drivable area.

[0038] Assume that a path planning system generates a target path 2 for a vehicle 1 from a given drivable area, which is typically determined by a perception module (see e.g. Figure 667 in the figure). Assume that the target path 2 is sent to the vehicle controller that calculates the steering angle and acceleration values ​​so that the vehicle 1 travels along the generated path 2. In this context, the perception system is to be understood as a system responsible for acquiring raw sensor data from sensors such as cameras, laser radar (LIDAR), radar (RADAR), and ultrasonic sensors and converting this raw data into scene understanding. Of course, the drivable area can be determined by additional means such as high-definition (HD) map data, position data obtained from a positioning system (e.g., GNSS), etc.

[0039] In addition, Figure 2 In FIG, it is assumed that the vehicle 1 has a posture (ie, geographic location and travel angle) away from the target path (ie, the center of the lane). In the illustration, it is assumed that the vehicle 1 is stationary, so the target path is generated to move the vehicle 1 to the center of the lane 4.

[0040] The path planning module (i.e., the control device) of vehicle 1 obtains the drivable area of ​​the surrounding environment. The drivable area can be, for example, one or more lanes of a road. Then, within the drivable area, the Path 2 generation module generates a target path using a predetermined sampling rate. If the sampling rate is, for example, 10 Hz, the Path 2 generation module updates the target path every 0.1 seconds.

[0041] In more detail, a path 2 is generated for time step t based on a predetermined set of path characteristics and further based on a predetermined set of constraints, wherein the predetermined set of constraints includes at least one constraint 7, 8 based on (depending on) the current posture of the vehicle. The current posture of the vehicle can, for example, be retrieved from an auxiliary vehicle control unit such as a positioning system (e.g., GNSS), an inertial measurement unit (IMU), or any other suitable sensor device. The predetermined set of characteristics can be interpreted as control parameters applied to generate a smooth path without unnecessary harsh acceleration or jolts, i.e., criteria that are more relevant to the user experience than physical vehicle constraints. The predetermined set of characteristics can, for example, include at least one of the path smoothness level, the distance to the lane center, and the path length. On the other hand, the predetermined set of constraints can further include at least one of the minimum turning radius of vehicle 1, the (longitudinal) length of vehicle 1, the (lateral) width of vehicle 1, the (vertical) height of vehicle 1, the ground clearance of vehicle 1, and at least one drivable area boundary. In this context, a set is to be interpreted as a number equal to or greater than one.

[0042] Regarding the drivable area boundary constraint, it should be interpreted as not allowing the vehicle to drive into areas that are not suitable for driving (e.g., lawns, sidewalks, etc.). Of course, the constraint set may further include other constraints such as avoiding collisions with external objects (e.g., other vehicles, road signs, pedestrians, etc.).

[0043] Furthermore, at least one constraint based on the current posture (position and travel angle) of the vehicle includes two (virtual) side boundaries 7, 8 extending from the vehicle 1 in the direction of travel of the vehicle 1 (indicated by arrows traveling along the vehicle 1). Figure 2 The example embodiment depicted in does not include side boundaries 7, 8 extending both from the front and the rear of the vehicle 1. In more detail, the area 26 between the two side boundaries defines an allowed region 26 within which at least a portion of the path of the time step t is allowed to form.

[0044] The side boundaries 7, 8 extend parallel to the longitudinal center axis 21 of the vehicle 1. The longitudinal center axis is an imaginary axis extending along the length of the vehicle 1 through the center point 5 of the rear axle. Moreover, the side boundaries 7, 8 are set at predetermined lateral distances 24, 25 from the longitudinal center axis 21 of the vehicle 1. The side boundaries 7, 8 can be equidistantly spaced from the center axis 21. Alternatively, the first side boundary 7 can be set at a first lateral distance 24 from the longitudinal center axis 21, and the second side boundary 8 can be set at a second lateral distance 24 from the longitudinal center axis 21, the first distance and the second distance being different. The spacing between the side boundaries 7, 8 and the longitudinal center axis 21 can depend on external factors such as the geographical location of the vehicle 1, predetermined user preferences, etc.

[0045] Furthermore, the distances 24, 25 between the longitudinal center axis 21 of the vehicle 1 and the side boundaries 7, 8 can be dynamic. More specifically, if a portion of the path 2 at time step t cannot be formed within the allowable region between the two side boundaries 7, 8, the distances 24, 25 can be increased to a predetermined maximum distance position or until a portion of the path can be formed within the allowable region.

[0046] Furthermore, the constraint set on which the updated path is based may further include at least one of the vehicle's minimum turning radius, the vehicle's (longitudinal) length, the vehicle's (lateral) width, the vehicle's (vertical) height, and the vehicle's ground clearance. In other words, the updated path is optimized so that it does not violate the physical properties of vehicle 1.

[0047] Accordingly, the generated path 2 will take into account the current posture of the vehicle 1, which will give the subsequent vehicle trajectory a higher chance of merging with the generated path 2 in a smoother path, thereby improving stability and comfort. In other words, by introducing this constraint (side boundaries 7, 8), the generated path 2 is aligned with the vehicle posture. Therefore, when the generated path 2 is tracked by the controller, smooth vehicle motion can be achieved. This concept will be referred to in detail in the following text. Figure 3 (a) to Figure 3 (d) further clarify that Figure 3 (a) to Figure 3(d) shows a comparison between the path planning solution proposed according to an embodiment of the present disclosure and a traditional path planning system in a valet parking application.

[0048] Figure 3 (a) is a schematic top view of a vehicle 1 having a control device for path planning according to an exemplary embodiment of the present disclosure, and Figure 3 (b) shows the vehicle trajectory 3 during the path convergence. A vehicle 1 is located in a parking lot 4, and a path planning system is configured to generate a path 2 for parking the vehicle in a designated parking space 33 in a safe and comfortable manner. Figure 3 (c) and Figure 3 (d) is a comparative top view of a vehicle 1' with a conventional path planning solution, wherein Figure 3 (c) shows the resulting path 2', and Figure 3 (d) shows the vehicle trajectory 3' during the path convergence in the parking lot 4'.

[0049] The resulting path 2 is consistent with the vehicle pose due to the constraints imposed by the side boundaries 7, 8. Therefore, when it is subsequently tracked by the vehicle controller (by Figure 3 (b) shows the trajectory 3), which allows for smooth vehicle motion. Further, constraints can be imposed on the path, such as a minimum distance to external objects, here in the form of parked vehicles 32. Figure 3 (c) and Figure 3 (d) shows the generated path 2' and vehicle trajectory 3', (i.e., without considering the vehicle pose) but with the constraints imposed by the side boundaries. The generated path 2' is instead optimized around the center 31' of the drivable area 4. Because the initial vehicle offset (the difference between the vehicle pose and the target path 2') is large in the prior art example, the controller tends to overshoot the path 2', and further because the length of the target path 2' is relatively short, the path convergence may fail (as in Figure 3 (d)).

[0050] Figure 4 400 is a schematic block diagram representation of a method for path planning for an autonomous or semi-autonomous vehicle according to an exemplary embodiment of the present disclosure. The block diagram 400 illustrates how the path generated / updated at each time step t (t=k, k=[1, 2...N]) depends on the paths generated at one or more previous time steps t (t=k-m, m=[1, 2...N-1], m≤k). The method includes obtaining 402a-402d a drivable area of ​​the vehicle's surroundings. The term "obtain" is to be interpreted broadly and includes receiving, retrieving, collecting, acquiring, and the like.

[0051] A drivable area is determined based on sensor data 401 retrieved from one or more sensor devices configured to monitor the vehicle's surroundings. As mentioned, the drivable area may include one or more lanes of a road segment, a parking lot, a one-way street, etc. Furthermore, the method includes generating 403a-403d paths within the drivable area for each time step t (t=k) based on a predetermined set of characteristics of the path and a predetermined set of constraints. As mentioned, the predetermined characteristics of the path are more related to comfort aspects (lane smoothness, distance to other targets, distance to boundaries, distance to lane center, etc.), i.e., the characteristics can be understood as optimization parameters. Moreover, the characteristics can be dynamic (i.e., depending on current user settings, time of day, etc.). Constraints, on the other hand, are more related to restrictions, such as physical limitations of the vehicle (e.g., turning radius), safety (e.g., traffic monitoring to avoid collisions), and, in particular, restrictions imposed by the vehicle's current posture as described above.

[0052] Constraints can be interpreted to some extent as static restrictions that should not be violated (e.g., the vehicle cannot turn with a radius greater than its minimum turning radius, or the vehicle cannot pass through an area narrower than the width of the vehicle). However, in some exemplary embodiments, the constraints based on the current posture of the vehicle are soft constraints, i.e., they can be violated (e.g., to avoid a collision) if one or more other predetermined criteria are met. In other words, the updated path of time step t may be allowed to form on the side boundary (cf. Figure 2 7, 8) outside.

[0053] Figure 5 FIG2 is a schematic flow chart representation of a method 500 for path planning for an autonomous or semi-autonomous vehicle according to an embodiment of the present disclosure. The following description will focus on the path generation step 502. However, as previously illustrated, it can be said that the path is generated at least in part based on sensor data 501, which can be provided, for example, by a perception system of the vehicle.

[0054] Next, path generation 502 is treated as an optimization problem with a cost function and a set of constraints. The cost function represents the user's desired path characteristics, and the constraints include restrictions on the drivable area in which the path is allowed to exist. The optimal path is then found as the one that minimizes the cost function.

[0055] In many cases, and in practical implementations, the problem is formulated as a quadratic programming problem:

[0056] Minimize the cost function:

[0057]

[0058] Subject to:

[0059] Aw≤b (2)

[0060] Where w is the parameter vector to be optimized, Q is the weight matrix, c is the weight vector, A is the constraint matrix, and b is the constraint vector.

[0061] The parameter vector determines the shape of the target path. For example, when the path is modeled by a polynomial, the vector is the polynomial coefficient. The advantage of this type of approach is that it uses an efficient algorithm that can find the optimal solution in real time. Figure 5 Examples of specific implementations of quadratic programming problems are discussed below.

[0062] In order to add the constraint based on the current posture of the vehicle, the above optimization problem is modified. More specifically, equations (1) and (2) are updated as follows:

[0063]

[0064] A a w a ≤b a (4)

[0065] Among them, w a is the parameter vector to be optimized, and it is enhanced from w in equations (1) and (2):

[0066]

[0067] Q a and c a are the cost matrix enhanced by Q from equation (1) and the cost vector enhanced by c from equation (1), respectively:

[0068]

[0069] Where ρ is the weight of the (soft) constraint based on the current posture of the vehicle. a and b a are the constraint matrix enhanced from A in equation (2) and the constraint vector enhanced from b in equation (2), respectively:

[0070]

[0071] Next, method 500 includes calculating 503 a cost function matrix and calculating a cost function vector. In this context, a cost function can be understood as a mathematical representation of desired target path characteristics. For example, these characteristics can be a specific "smoothness" of the path, proximity to lane centers, total length of the path, etc. To calculate the matrix Q and vector c of the quadratic programming problem described above, the cost function values ​​are numerically calculated in discrete space (or time).

[0072] In more detail, the step of calculating 503 the cost function matrix and the cost function vector comprises calculating 504 a “base” matrix and a “base” vector (Q, c), and then forming an enhanced matrix and an enhanced vector (Q a , c a ).

[0073] The method further includes calculating 506 a set of constraints, wherein the set of constraints further includes at least one constraint based on the boundaries of the drivable area and at least one constraint based on vehicle characteristics. More specifically, a constraint matrix and a constraint vector are calculated 506. Some constraints are derived from the boundaries of the drivable area to ensure that the optimal path lies within the drivable area (e.g., within a parking lot and within permitted lanes within the parking lot). Additionally, the curvature of the path may be required to be below a certain threshold (e.g., a minimum turning radius of the vehicle).

[0074] Furthermore, the step of calculating 506 the constraint matrix and the constraint vector further includes deriving 507 at least one constraint based on the current vehicle posture. This constraint may also be referred to as, for example Figure 2 , lateral boundaries 7 and 8 are shown in FIG. Therefore, method 500 further includes obtaining or forming 508 lateral boundaries based on the current posture of vehicle 1. In other words, a narrow constraint is generated / obtained that extends from the longitudinal center axis of the vehicle through the front and rear of the vehicle by a predetermined length. Next, the lateral boundaries are offset 509 to the left and right, respectively, by a predetermined distance from the longitudinal center axis of the vehicle.

[0075] Once the "pose" constraints are derived 507, an enhanced constraint matrix and an enhanced constraint vector (A) are formed 510. a 、b a ). Next, an optimization problem is solved 511. This can be understood as minimizing a cost function given a predetermined set of characteristics and a predetermined set of constraints. More specifically, this step 511 solves the quadratic programming problem discussed above. For this task, efficient and numerically stable traditional algorithms can also be used.

[0076] Further, a target path is constructed 512 from the solution 511. Depending on how the quadratic programming problem is formulated, the output of optimization 511 can be a set of polynomial coefficients, a sequence of control inputs to the ego vehicle motion model, or a set of weights that form the shape of the path. Once the path is generated 502 or constructed 512, method 500 can include controlling the vehicle to execute the generated / constructed path.

[0077] Optionally, executable instructions for performing these functions are included in a non-transitory computer-readable storage medium or other computer program products configured to be executed by one or more processors.

[0078] Figure 6is a schematic side view of a vehicle with a control device 10 for path planning for an autonomous or semi-autonomous vehicle 1. Vehicle 1 further includes a perception system 66, a positioning system 65, and an inertial measurement unit 69, all in communication with the control device. The perception system 66 in this context can be understood as a system responsible for acquiring raw sensor data from sensors 67a, 67b, and 67c, such as cameras, lidars, radars, and ultrasonic sensors, and converting this raw data into scene understanding. The positioning system 65 is configured to monitor the vehicle's geographic location and movement and can take the form of a global navigation satellite system (GNSS) such as GPS. However, to improve accuracy, the positioning system can alternatively be implemented as a real-time kinematic (RTK) GPS. An inertial measurement unit (IMU) is understood to be an electronic device that uses a combination of accelerometers, gyroscopes, and sometimes magnetometers to measure and report the specific thrust, angular velocity, and sometimes the orientation of the vehicle. Thus, the vehicle's attitude can be retrieved from any one or a combination of these peripheral systems 65, 67, and 69.

[0079] The control device 10 includes one or more processors 11, a memory 12, a sensor interface 13, and a communication interface 14. The processor 11 may also be referred to as a control circuit 11 or a control circuit device 11. The control circuit 11 is configured to execute instructions stored in the memory 12 to perform a method for path planning for an autonomous or semi-autonomous vehicle according to any of the embodiments disclosed herein. In other words, the memory 12 of the control device 10 may include one or more (non-transitory) computer-readable storage media for storing computer-executable instructions, for example, when the instructions are executed by one or more computer processors 11, the computer processor 11 may perform the technology described herein. The memory 12 optionally includes a high-speed random access memory (such as DRAM, SRAM, DDR RAM or other random access solid-state memory devices), and optionally includes a non-volatile memory (such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices or other non-volatile solid-state storage devices).

[0080] In more detail, the control circuit 11 is configured to obtain (via the sensor interface 13 or the communication interface 14) a drivable area of ​​the surrounding environment of the vehicle 1. The drivable area can be defined, for example, by road boundaries, which are detected and determined by the perception system 66 and associated sensor devices 66a-66c. However, in an alternative embodiment (not shown), the control device 10 can receive raw data directly from one or more sensor devices 66a-66c, and the determination of the drivable area can be performed by a dedicated module of the control device 10 based on the received sensor data. The control circuit 11 is further configured to generate a path within the drivable area for time step t based on a predetermined set of constraints and a predetermined set of characteristics of the path. The constraint set includes at least one constraint based on the current position of the vehicle. The current posture of the vehicle can be obtained from any suitable peripheral system of the vehicle 1 (such as, the positioning system 65 or the IMU 69).

[0081] Furthermore, the vehicle 1 can be connected to an external network 62 (e.g. for retrieving map data) via, for example, a wireless link. The same or some other wireless links can be used to communicate with other vehicles in the vicinity of the vehicle or with local infrastructure elements. Cellular communication technology can be used for long-range communication (e.g. to an external network) and, if the cellular communication technology used has low latency, it can also be used for communication between vehicles, vehicle-to-vehicle (V2V) and / or vehicle-to-infrastructure (V2X). Examples of cellular radio technologies are GSM, GPRS, EDGE, LTE, 5G, 5G NR, etc., also including future cellular solutions. However, in some solutions, medium and short-range communication technologies such as wireless local area networks (LANs), e.g. solutions based on IEEE802.11, are used. ETSI is working on cellular standards for vehicle communications and, for example, 5G is considered a suitable solution due to its low latency and high bandwidth and efficient handling of communication channels.

[0082] In summary, to mitigate the problem of failed convergence to the vehicle's target path, a constraint depending on the vehicle's current pose is introduced. More specifically, the "vehicle pose constraint" has the following properties.

[0083] A narrow constraint is formed that extends toward the front and rear of the vehicle along the vehicle's longitudinal center axis (i.e., the axis intersecting the vehicle's center point, such as the center of the rear axle). Furthermore, as an option, this constraint can only be violated if no optimal path can be found without violating it. Alternatively, the spacing between the side boundaries (defining the constraints) can be increased until a solution is found. By introducing this constraint, the optimal target path is consistent with the vehicle's pose, thereby achieving smooth vehicle motion when tracked by the controller.

[0084] Furthermore, the proposed path optimization method, which directly optimizes a mathematical function representing the path (independent of the control input sequence), is computationally efficient because the number of parameters to be optimized is smaller. This method is robust and can initiate autonomous driving mode from any initial ego vehicle pose while achieving smooth vehicle motion.

[0085] The present disclosure has been presented above with reference to specific embodiments. However, other embodiments than those described above are possible and are within the scope of the present disclosure. Different method steps of the method performed by hardware or software other than those described above may be provided within the scope of the present disclosure. Thus, according to an exemplary embodiment, a non-transitory computer-readable storage medium is provided having stored thereon one or more programs configured to be executed by one or more processors of a vehicle control system, the one or more programs including instructions for performing a method according to any of the above-described embodiments. Alternatively, according to another exemplary embodiment, a cloud computing system may be configured to perform any of the methods presented herein. The cloud computing system may include distributed cloud computing resources that collectively perform the methods presented herein under the control of one or more computer program products.

[0086] In general, computer-accessible media may include any tangible or non-transitory storage media or memory media, such as electronic, magnetic, or optical media, for example, a disk or CD / DVD ROM coupled to a computer system via a bus. As used herein, the terms "tangible" and "non-transitory" are intended to describe computer-readable storage media (or "memory") that exclude propagating electromagnetic signals, but are not intended to otherwise limit the types of physical computer-readable storage devices encompassed by the phrase computer-readable media or memory. For example, the terms "non-transitory computer-readable media" or "tangible memory" are intended to encompass various types of storage devices that do not necessarily store information permanently, including, for example, random access memory (RAM). Program instructions and data stored in non-transitory form on tangible computer-accessible storage media may be further transmitted by a transmission medium or signal, such as an electrical, electromagnetic, or digital signal, which may be carried via a communication medium such as a network and / or wireless link.

[0087] Processor 11 (associated with control device 10) can be or include any number of hardware components for performing data or signal processing or for executing computer code stored in memory 12. Device 10 has associated memory 12, and memory 12 can be one or more devices for storing data and / or computer code used to complete or facilitate the various methods described in this specification. The memory may include volatile memory or non-volatile memory. The memory 12 may include a database component, an object code component, a script component, or any other type of information structure for supporting the various actions of this specification. According to exemplary embodiments, any distributed or local memory device may be utilized in conjunction with the systems and methods of this specification. According to exemplary embodiments, the memory 12 is communicatively connected to the processor 11 (e.g., via circuitry or any other wired, wireless, or network connection) and includes computer code for performing one or more processes described herein.

[0088] It will be appreciated that the sensor interface 13 may also provide the possibility of acquiring sensor data directly or via a dedicated sensor control circuit 66 in the vehicle. The communication / antenna interface 14 may further provide the possibility of sending output to a remote location (e.g., a remote operator or a control center) by means of an antenna 68. Moreover, some sensors in the vehicle may be arranged to communicate with the control device 10 using a local network such as a CAN bus, I2C, Ethernet, fiber optics, etc. The communication interface 14 may be arranged to communicate with other control functions of the vehicle and may therefore also be considered a control interface; however, a separate control interface (not shown) may be provided. Local communication within the vehicle may also be of a wireless type using protocols such as WiFi, LoRa, Zigbee, Bluetooth, or similar medium / short range technologies.

[0089] Accordingly, it should be understood that portions of the described solutions may be implemented in the vehicle, in a system external to the vehicle, or in a combination of both internal and external locations; for example, in a server in communication with the vehicle, a so-called cloud solution. For example, sensor data may be sent to an external system, and that system may perform the steps of determining the drivable area of ​​vehicle 1's surroundings. The various features and steps of the embodiments may be combined in combinations other than those described.

[0090] Exemplary methods, computer-readable storage media, control devices, and vehicles are listed in the following:

[0091] 1. A method for path planning for an autonomous or semi-autonomous vehicle, the method comprising:

[0092] Obtaining a drivable area of ​​the vehicle's surrounding environment;

[0093] generating a path within the drivable area for time step t based on a predetermined set of constraints and a predetermined set of characteristics of the path;

[0094] The predetermined constraint set includes at least one constraint based on a current posture of the vehicle.

[0095] 2. The method of item 1, wherein the step of generating a path within the drivable area for the time step t comprises:

[0096] calculating a cost function, wherein the cost function is a mathematical representation of the predetermined set of characteristics of the path;

[0097] calculating the predetermined set of constraints, wherein the predetermined set of constraints further comprises: at least one constraint based on a boundary of the drivable area and at least one constraint based on vehicle characteristics;

[0098] solving an optimization problem based on the calculated cost function and the calculated set of predetermined constraints;

[0099] The path is formed based on the solved optimization problem.

[0100] 3. The method according to item 2, wherein the step of calculating the cost function comprises:

[0101] Calculate the cost function matrix Q and the cost function vector c;

[0102] Forming the enhanced cost function matrix Q a and the enhanced cost function vector c a .

[0103] 4. The method according to item 2 or item 3, wherein the step of calculating the predetermined constraint set comprises:

[0104] Calculate the constraint matrix A and constraint vector b;

[0105] deriving the at least one constraint based on the current pose of the vehicle;

[0106] Form the enhanced constraint matrix A a and the enhanced constraint vector b a .

[0107] 5. The method according to any one of the preceding clauses 1, wherein the vehicle has a longitudinal axis intersecting a predetermined center point of the vehicle, and

[0108] The at least one constraint based on the current posture of the vehicle includes two side boundaries extending parallel to the longitudinal axis of the vehicle at a first predetermined lateral distance from the longitudinal axis.

[0109] 6. A method according to item 5, wherein the area between the two side boundaries defines an allowed area within which at least a portion of the path of time step t is allowed to form.

[0110] 7. The method according to item 6, further comprising:

[0111] If the portion of the path at time step t cannot be formed within the allowable area between the two side boundaries, the first predetermined distance is increased until the portion of the path can be formed within the allowable area.

[0112] 8. The method according to item 5 or item 6, wherein the two lateral boundaries have a longitudinal extension within a second predetermined distance from at least one outer edge of the vehicle.

[0113] 9. A method according to any of the preceding items, wherein the predetermined set of characteristics includes at least one of a level of path smoothness, a distance to a lane centre and a length of the path.

[0114] 10. The method according to any one of the preceding items, wherein the predetermined set of constraints further includes at least one of: a minimum turning radius of the vehicle, a length of the vehicle, a width of the vehicle, a height of the vehicle, a ground clearance of the vehicle, and at least one drivable area boundary.

[0115] 11. A computer-readable storage medium storing one or more programs, the one or more programs being configured to be executed by one or more processors of a vehicle control system, the one or more programs comprising instructions for executing the method according to any one of the preceding items.

[0116] 12. A control device for path planning of an autonomous or semi-autonomous vehicle, the control device comprising a control circuit configured to:

[0117] Obtaining a drivable area of ​​the vehicle's surrounding environment;

[0118] generating a path within the drivable area for time step t based on a predetermined set of constraints and a predetermined set of characteristics of the path;

[0119] The predetermined constraint set includes at least one constraint based on a current posture of the vehicle.

[0120] 13. The control device according to item 12, wherein the control circuit is configured to generate a path for a time step t within the drivable area by:

[0121] calculating a cost function comprising the predetermined set of characteristics of the path;

[0122] calculating the predetermined set of constraints, wherein the predetermined set of constraints further comprises: at least one constraint based on a boundary of the drivable area and at least one constraint based on vehicle characteristics;

[0123] solving an optimization problem based on the calculated cost function and the calculated set of predetermined constraints;

[0124] The path is formed based on the solved optimization problem.

[0125] 14. A control device according to item 12 or item 13, wherein the vehicle has a longitudinal axis intersecting with a predetermined center point of the vehicle, wherein the at least one constraint based on the current posture of the vehicle includes two side boundaries extending parallel to the longitudinal axis at a first predetermined distance from the longitudinal axis of the vehicle.

[0126] 15. A control device according to item 14, wherein the area between the two side boundaries defines an allowed area within which at least a portion of the path of time step t is allowed to form.

[0127] 16. A vehicle comprising:

[0128] a perception system comprising at least one sensor for monitoring the surrounding environment of the vehicle;

[0129] A control device according to any of items 12-15.

[0130] It should be noted that the word "comprising" does not exclude the presence of other elements or steps than those listed, and the word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. It should further be noted that any reference numerals do not limit the scope of the claims, the present disclosure may be implemented at least in part by means of both hardware and software, and several "means" or "units" may be represented by the same item of hardware.

[0131] Although each figure can illustrate the specific order of method steps, the order of steps may be different from the order described. In addition, two or more steps can be performed simultaneously or partially simultaneously. This variation will depend on selected software and hardware systems and depend on the designer's choice. All of these variations are within the scope of the present disclosure. Similarly, software implementation can adopt the standard programming techniques with rule-based logic or other logic that is used to complete various connection steps, processing steps, comparison steps and decision steps to complete. The above-mentioned and described embodiments are only given as examples, and should not limit the present disclosure. Other solutions, purposes, targets and functions within the scope of the present disclosure as claimed in the patent embodiments described below should be apparent to those skilled in the art.

Claims

1. A method (400, 500) for path planning of an autonomous or semi-autonomous vehicle, the method comprising: Obtaining ( 402 a - 402 d ) a drivable area of ​​the vehicle's surroundings; as well as generating (403a-403d, 502) a path within the drivable area for time step t based on a predetermined set of constraints and a predetermined set of characteristics of the path; wherein the predetermined constraint set includes at least one constraint based on the current posture of the vehicle, wherein the at least one constraint based on the current posture of the vehicle comprises: two side boundaries extending along and parallel to the longitudinal axis of the vehicle at a first predetermined lateral distance from the longitudinal axis of the vehicle; and Therein, the area between the two side boundaries defines an allowed area within which at least a portion of the path at time step t is allowed to form.

2. The method (400, 500) according to claim 1, wherein: The step of generating (403a-403d, 502) a path within the drivable area for the time step t comprises: calculating (503) a cost function, wherein the cost function is a mathematical representation of the predetermined set of characteristics of the path; calculating ( 506 ) the predetermined set of constraints, wherein the predetermined set of constraints further comprises: at least one constraint based on a boundary of the drivable area and at least one constraint based on vehicle characteristics; solving (511) the optimization problem based on the calculated cost function and the calculated predetermined constraint set; and The path is formed (512) based on the solved optimization problem.

3. The method (400, 500) according to claim 2, wherein: The steps of calculating (503) the cost function include: Calculating (504) a cost function matrix Q and a cost function vector c; and Form (505) the enhanced cost function matrix Q a and the enhanced cost function vector c a .

4. The method (400, 500) according to claim 2 or 3, wherein: The step of calculating (506) the predetermined set of constraints comprises: Calculate the constraint matrix A and constraint vector b; deriving (506) the at least one constraint based on the current pose of the vehicle; and Form (510) the enhanced constraint matrix A a and the enhanced constraint vector b a .

5. The method (400, 500) according to claim 1, further comprising: If the portion of the path at time step t cannot be formed within the allowed area between the two side boundaries, the first predetermined lateral distance is increased (509) until the portion of the path can be formed within the allowed area.

6. The method (400, 500) according to claim 1 or 5, wherein: The two side boundaries have a longitudinal extension within a second predetermined distance from at least one outer edge of the vehicle.

7. The method (400, 500) according to claim 1, wherein The predetermined set of characteristics includes at least one of a level of path smoothness, a distance to a lane center, and a length of the path.

8. The method (400, 500) according to claim 1, wherein The predetermined set of constraints further includes at least one of a minimum turning radius of the vehicle, a length of the vehicle, a width of the vehicle, a height of the vehicle, a ground clearance of the vehicle, and at least one drivable area boundary.

9. A non-transitory computer-readable storage medium storing one or more programs configured to be executed by one or more processors of a vehicle control system, the one or more programs including instructions for executing the method (400, 500) according to claim 1.

10. A control device (10) for path planning of an autonomous or semi-autonomous vehicle (1), the control device comprising a control circuit (11), the control circuit (11) being configured to: Obtaining a drivable area of ​​the surrounding environment (4) of the vehicle (1); and generating a path (2) within the drivable area for a time step t based on a predetermined set of constraints and a predetermined set of characteristics of the path (2); in, The predetermined set of constraints includes at least one constraint (7, 8) based on the current posture of the vehicle (1), wherein the at least one constraint (7, 8) based on the current posture of the vehicle comprises: two side boundaries (7, 8) extending along and parallel to the longitudinal axis (21) of the vehicle at a first predetermined lateral distance (24, 25); and Therein, the area between the two side boundaries defines an allowed area within which at least a portion of the path at time step t is allowed to form.

11. The control device (10) according to claim 10, wherein: The control circuit (11) is configured to generate a path (2) within the drivable area for a time step t by: calculating a cost function comprising said predetermined set of characteristics of said path (2); calculating the predetermined set of constraints, wherein the predetermined set of constraints further comprises: at least one constraint based on a boundary of the drivable area and at least one constraint based on vehicle characteristics; solving an optimization problem based on the calculated cost function and the calculated predetermined constraint set; and The path (2) is formed based on the solved optimization problem.

12. A vehicle (1), comprising: a perception system (67) comprising at least one sensor (66a-66c) for monitoring the vehicle's surroundings; and A control device (10) according to claim 10 or 11.

Images