Trajectory planning method, device, computing device, mobile object and storage medium
By using the potential field function to construct a repulsive potential field in the autonomous driving system, the speed planning problem caused by the uncertainty of traffic participants' trajectory prediction in trajectory planning is solved, and the driving efficiency and safety of autonomous driving vehicles are improved.
Patent Information
- Application Number
- CN202011454433.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-12-10
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2040-12-10
AI Technical Summary
When facing the uncertainty of traffic participants' trajectory prediction results, the speed planning is too conservative, resulting in reduced driving efficiency and even the problem of stagnation of autonomous vehicles.
The potential field function is used to construct the speed obstacle block, and the repulsive potential field is constructed based on the probability and time lapse of the predicted trajectory, and the vehicle speed is adjusted to adapt to the uncertainty of the predicted trajectory of traffic participants, reduce the impact on the vehicle speed, and improve driving efficiency.
Through this method, vehicles can plan their speed more reasonably in uncertain environments, avoid overconservatism, improve mobility efficiency, and ensure driving safety and efficiency.
Smart Images

Figure CN114620058B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of autonomous driving, specifically to the field of trajectory planning, and in particular to a trajectory planning method, apparatus, computing device, mobile body and storage medium. Background Art
[0002] At present, vehicle autonomous driving technology has attracted much attention. In the vehicle's autonomous driving system, the trajectory planning module plays a vital role in the vehicle's driving safety and traffic efficiency. Trajectory planning includes the vehicle's position and speed, which belong to the spatial and temporal dimensions respectively. Trajectory planning can be divided into two sub-problems: path planning and speed planning. In trajectory planning, the speed of the planned path is mainly planned and the speed of the vehicle is adjusted to avoid dynamic traffic participants.
[0003] One of the upstream information sources for trajectory planning is the vehicle's environmental vehicle trajectory prediction module, which makes real-time predictions on the trajectories of other traffic participants such as environmental vehicles in the vehicle's driving environment, and performs speed planning based on the prediction results.
[0004] For example, in the prior art, there is a real-time trajectory planning method for unmanned vehicles based on speed barriers, which reasonably and effectively performs real-time trajectory planning of the vehicle in the configuration space of the vehicle's three-dimensional position and heading, combines smooth controllable primitives that conform to the vehicle's dynamic constraints with the concept of speed barriers, and performs heuristic search by setting the optimal speed to solve the collision-free estimation of any two configuration spaces of the unmanned vehicle.
[0005] This method only considers the current speed of the unmanned vehicle and the obstacle, but in fact, the future speed of both the unmanned vehicle and the obstacle is not necessarily uniform. The real-time trajectory planned by this method is not predictable, and the trajectory of the vehicle over time is not the best and most reasonable.
[0006] In addition, in the prior art, there is also a method for selecting the optimal reference trajectory and performing dynamic trajectory planning based on the different positions of obstacle vehicles on the road and their corresponding working conditions. By analyzing the generation of the lane change intention of the unmanned vehicle and the conditions under which the lane change can be executed, and based on the prediction of the position and speed of the surrounding obstacle vehicles, the local optimal lane change trajectory is fitted at the initial moment of determining the obstacle avoidance lane change, and then this optimal trajectory is used as the local reference trajectory to generate a trajectory cluster that the unmanned vehicle can drive, and the designed speed distance cost function is combined with the loss function, and the nonlinear model predictive control is used to screen out the optimal trajectory in the trajectory cluster.
[0007] In this method, the optimal trajectory is selected from the trajectory cluster as the final trajectory of the unmanned vehicle. However, in this trajectory cluster, each trajectory has a constant speed, and for the same path, the set speed is also unique, which couples the spatial and temporal solution spaces, loses many feasible solutions, and cannot achieve global optimality.
[0008] In the process of creating this application, the inventor discovered that, at certain times, the driving intentions of various road participants are unclear, and the prediction results of their trajectories are uncertain. Specifically, there may be multiple predicted trajectories for each road participant, and the prediction results of each predicted trajectory present a probability distribution. However, the above-mentioned prior art does not take this probability distribution into consideration. If each predicted trajectory is treated equally, the speed of the vehicle will be too conservative, and the driving efficiency will be greatly reduced on complex urban roads, and the problem of autonomous driving vehicles stagnating may even occur. Summary of the invention
[0009] In view of the above problems in the prior art, the present application provides a trajectory planning method, apparatus, computing device, mobile body and storage medium, which can adapt to the uncertainty of trajectory prediction results of traffic participants, and perform speed planning of the mobile body taking into account different prediction probabilities so that the speed of the mobile body will not be too conservative, thereby improving mobility efficiency.
[0010] To achieve the above-mentioned purpose, the first aspect of the present application provides a trajectory planning method, which is applied to a first mobile body capable of autonomous movement, and includes a path planning step of planning the moving path of the first mobile body and a speed planning step of planning the moving speed of the first mobile body on the planned path. In the speed planning step: obtaining the predicted trajectory of the second mobile body and the probability that the second mobile body moves according to the predicted trajectory, wherein the predicted trajectory includes the predicted path and predicted speed of the second mobile body; when there is an intersection path between the planned path and the predicted path, constructing a speed obstacle block according to the planned path and the predicted trajectory, wherein the speed obstacle block is characterized by the corresponding relationship between the position and time of the second mobile body on the intersection path; using a potential field function to construct a repulsive potential field of the speed obstacle block, wherein the potential field function takes the position and time on the planned path as independent variables, takes the probability as a constant, takes the strength of the repulsive potential field as a dependent variable, and the greater the value of the probability, the stronger the strength of the repulsive potential field; according to the repulsive potential field and the planned path, planning the moving speed of the first mobile body on the planned path. The position as the independent variable of the potential field function is, for example, the accumulated length along the planned path of the first moving body.
[0011] The first moving object may be, for example, an autonomous vehicle, a drone, or a robot, etc. The second moving object may be, for example, a vehicle, a person, etc. Here, the second moving object mainly refers to a person or an object that can move at a certain speed, but is not limited thereto, and may also include an object that is stationary.
[0012] The predicted trajectory of the second moving object and its probability can be estimated based on the state information of the second moving object, the road network structure near the second moving object, etc. The state information of the second moving object may include the current position, direction, speed, etc.
[0013] There may be multiple predicted trajectories for the second moving object. For example, at a location such as an intersection, the predicted trajectory of the second moving object may be to go straight, turn left, or turn right. In this case, the predicted trajectory of the second moving object has uncertainty, and probability is used to characterize the possibility of the second moving object moving according to each predicted trajectory.
[0014] Using the trajectory planning method provided in the first aspect of the present application, the strength of the repulsive potential field of the second mobile body on the first mobile body is related to the probability that the second mobile body moves according to the predicted trajectory. The larger the value of the probability, the stronger the strength of the repulsive potential field. Therefore, it can better adapt to the uncertainty of the predicted trajectory of the second mobile body, reduce the impact of the predicted trajectory of the second mobile body with a smaller probability on the moving speed of the first mobile body, so that the speed of the first mobile body will not be too conservative, thereby improving the moving efficiency of the first mobile body.
[0015] As a possible implementation method of the first aspect of the present application, the "planning the moving speed of the first moving object on the planned path" is to plan the correspondence between the position and time of the first moving object on the planned path according to the repulsive potential field and the planned path, and calculate the moving speed of the first moving object according to the correspondence.
[0016] By adopting this method, the moving speed of the first moving body in each period can be planned in real time with a small amount of calculation, thereby improving the moving efficiency of the first moving body.
[0017] As a possible implementation of the first aspect of the present application, the "position on the intersection path" and the "position on the planned path" are characterized by the distance along the planned path to the position of the first moving body at the current moment.
[0018] As a possible implementation manner of the first aspect of the present application, the potential field strength decreases as the time relative to the current moment increases.
[0019] By adopting this method, the influence of the predicted trajectory of the second moving body in the future on the speed planning of the first moving body can be reduced, and the influence of the predicted trajectory with lower confidence can be reduced, so that the first moving body can be kept at the current speed as much as possible, and then accelerate or decelerate when the moving environment gradually becomes clear, thereby reflecting the interaction with other traffic participants to a greater extent and improving the mobility efficiency.
[0020] As a possible implementation of the first aspect of the present application, when the time lapse exceeds a time threshold, the potential field strength is zero. The time threshold may be determined according to a maximum duration of the predicted trajectory of the first moving object.
[0021] By adopting this method, when the time threshold is exceeded, the influence of the predicted trajectory of the second moving body on the speed planning of the first moving body is not considered. Therefore, the influence of the predicted trajectory of the second moving body whose confidence decreases significantly with the increase of time can be avoided, thereby improving the movement efficiency of the first moving body.
[0022] As a possible implementation of the first aspect of the present application, the time threshold is set according to the type of the second moving object. The second moving object may be an object with different movement patterns, such as a pedestrian or a vehicle.
[0023] In this way, the time threshold is set according to the type of the second moving object, so that the moving speed of the first moving object can be more reasonably planned to adapt to objects with different movement laws.
[0024] As a possible implementation of the first aspect of the present application, the potential field function is:
[0025]
[0026] Among them, a prob Represents the probability of the predicted trajectory, with a value range of (0,1]; d centor (s, t) represents the distance from (s, t) to the center of the speed obstacle block; d edge (s, t) represents the distance from (s, t) to the edge of the speed obstacle block; Indicates the boundary of the influence area of the edge of the speed barrier block. When it exceeds this boundary, that is, When p v (s, t) = 0, that is, the vehicle 10 is not affected by the speed obstacle block; t max Indicates the maximum duration of the predicted trajectory. When this maximum duration is exceeded, t>t max When p v (s, t)=0, the vehicle 10 is not affected by the speed obstacle block.
[0027] In this way, the moving speed of the first moving object can be planned more reasonably by taking into account the predicted trajectory probability of the second moving object, the distance to the obstacle, and the passage of time relative to the current time.
[0028] To achieve the above-mentioned purpose, the second aspect of the present application provides a trajectory planning device, which is applied to a first mobile body capable of autonomous movement, and includes a path planning module for planning the moving path of the first mobile body and a speed planning module for planning the moving speed of the first mobile body on the planned path. The speed planning module performs the following processing: obtaining the predicted trajectory of the second mobile body and the probability that the second mobile body moves according to the predicted trajectory, wherein the predicted trajectory includes the predicted path and predicted speed of the second mobile body; when there is an intersection path between the planned path and the predicted path, constructing a speed obstacle block according to the planned path and the predicted trajectory, wherein the speed obstacle block is characterized by the corresponding relationship between the position and time of the second mobile body on the intersection path; using a potential field function to construct a repulsive potential field of the speed obstacle block, wherein the potential field function takes the position and time on the planned path as independent variables, takes the probability as a constant, takes the strength of the repulsive potential field as a dependent variable, and the greater the value of the probability, the stronger the strength of the repulsive potential field; and planning the moving speed of the first mobile body on the planned path according to the repulsive potential field and the planned path.
[0029] The first moving object may be, for example, an autonomous vehicle, a drone, or a robot, etc. The second moving object may be, for example, a vehicle, a person, etc. Here, the second moving object mainly refers to a person or an object that can move at a certain speed, but is not limited thereto, and may also include an object that is stationary.
[0030] The predicted trajectory of the second moving object and its probability can be estimated based on the state information of the second moving object, the road network structure near the second moving object, etc. The state information of the second moving object may include the current position, direction, speed, etc.
[0031] There may be multiple predicted trajectories for the second moving object. For example, at a location such as an intersection, the predicted trajectory of the second moving object may be to go straight, turn left, or turn right. In this case, the predicted trajectory of the second moving object has uncertainty, and probability is used to characterize the possibility of the second moving object moving according to each predicted trajectory.
[0032] Using the trajectory planning device provided in the second aspect of the present application, the strength of the repulsive potential field of the second mobile body on the first mobile body is related to the probability that the second mobile body moves according to the predicted trajectory. The larger the value of the probability, the stronger the strength of the repulsive potential field. Therefore, it can better adapt to the uncertainty of the predicted trajectory of the second mobile body, reduce the impact of the predicted trajectory of the second mobile body with a smaller probability on the moving speed of the first mobile body, so that the speed of the first mobile body will not be too conservative, thereby improving the moving efficiency of the first mobile body.
[0033] As a possible implementation method of the second aspect of the present application, the "planning the moving speed of the first moving object on the planned path" is to plan the correspondence between the position and time of the first moving object on the planned path according to the repulsive potential field and the planned path, and calculate the moving speed of the first moving object according to the correspondence.
[0034] By adopting this method, the moving speed of the first moving body in each period can be planned in real time with a small amount of calculation, thereby improving the moving efficiency of the first moving body.
[0035] As a possible implementation of the second aspect of the present application, the "position on the intersection path" and the "position on the planned path" are characterized by the distance along the planned path to the position of the first moving body at the current moment.
[0036] As a possible implementation manner of the second aspect of the present application, the potential field strength decreases as the time relative to the current moment increases.
[0037] By adopting this method, the influence of the predicted trajectory of the second moving body in the future on the speed planning of the first moving body can be reduced, and the influence of the predicted trajectory with lower confidence can be reduced, so that the first moving body can be kept at the current speed as much as possible, and then accelerate or decelerate when the moving environment gradually becomes clear, thereby reflecting the interaction with other traffic participants to a greater extent and improving the mobility efficiency.
[0038] As a possible implementation of the second aspect of the present application, when the time lapse exceeds a time threshold, the potential field strength is zero. The time threshold may be determined according to a maximum duration of the predicted trajectory of the first moving object.
[0039] By adopting this method, when the time threshold is exceeded, the influence of the predicted trajectory of the second moving body on the speed planning of the first moving body is not considered. Therefore, the influence of the predicted trajectory of the second moving body whose confidence decreases significantly with the increase of time can be avoided, thereby improving the movement efficiency of the first moving body.
[0040] As a possible implementation of the second aspect of the present application, the time threshold is set according to the type of the second moving object. The second moving object may be an object with different movement patterns, such as a pedestrian or a vehicle.
[0041] In this way, the time threshold is set according to the type of the second moving object, so that the moving speed of the first moving object can be more reasonably planned to adapt to objects with different movement laws.
[0042] In order to achieve the above-mentioned purpose, the third aspect of the present application provides a mobile body, which has a track device as described in any one of the above-mentioned second aspect and its possible implementation methods.
[0043] By adopting the moving body provided in the third aspect of the present application, the speed of the moving body can be prevented from being too conservative, thereby improving the moving efficiency.
[0044] To achieve the above-mentioned objectives, the fourth aspect of the present application provides a computing device, comprising: at least one processor; and at least one memory storing program instructions, wherein the at least one processor executes the method described in the above-mentioned first aspect and any possible implementation thereof by executing the program instructions.
[0045] To achieve the above-mentioned purpose, the fifth aspect of the present application provides a computer-readable storage medium having program instructions stored thereon, and a computer executes the method described in any one of the above-mentioned first aspect and its possible implementation methods by executing the program instructions.
[0046] These and other aspects of the invention will become apparent from the following description of the embodiment(s). BRIEF DESCRIPTION OF THE DRAWINGS
[0047] The following further illustrates the various features of the present invention and the relationship between the various features with reference to the accompanying drawings. The accompanying drawings are all exemplary, some features are not shown in actual proportion, and some drawings may omit the conventional features in the field involved in the present application and are not necessary for the present application, or additionally show the features that are not necessary for the present application. The combination of the various features shown in the accompanying drawings is not intended to limit the present application. In addition, throughout this specification, the same figure numerals refer to the same content. The specific description of the drawings is as follows:
[0048] Figure 1 This is a schematic diagram of an application scenario of an embodiment of the present application;
[0049] Figure 2 is another application scenario schematic diagram of an embodiment of the present application;
[0050] Figure 3 is a structural schematic diagram of a vehicle provided in an embodiment of the present application;
[0051] Figure 4 is a structural schematic diagram of a trajectory planning device provided in an embodiment of the present application;
[0052] Figure 5 This is a schematic diagram of a path planning scenario provided by an embodiment of the present application;
[0053] Figure 6 The embodiment of the present application provides Figure 5 The corresponding St graph;
[0054] Figure 7 is a schematic flow chart of a trajectory planning method provided in an embodiment of the present application;
[0055] Figure 8 is a schematic diagram of a specific application example of the trajectory planning method provided in an embodiment of the present application;
[0056] Fig. 9 is with Figure 8 St diagram corresponding to the application example;
[0057] Fig.10 is with Figure 8 Schematic diagram of St curve corresponding to the application example;
[0058] Fig.11 is with Figure 8 Schematic diagram of Vt curve corresponding to the application example;
[0059] Fig.12 It is a structural diagram of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0060] In the following description, the numbers representing the steps, such as S10, S20, etc., do not necessarily mean that the steps must be executed in this manner. If permitted, the order of the previous and next steps can be interchanged, or the steps can be executed simultaneously.
[0061] The term "comprising" as used in the description and claims should not be interpreted as being limited to what is listed thereafter; it does not exclude other elements or steps. Therefore, it should be interpreted as specifying the presence of the features, integers, steps or components mentioned, but does not exclude the presence or addition of one or more other features, integers, steps or components and groups thereof. Therefore, the expression "a device comprising means A and B" should not be limited to a device consisting of components A and B only.
[0062] References to "one embodiment" or "an embodiment" in this specification mean that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least one embodiment of the present invention. Therefore, the phrases "in one embodiment" or "in an embodiment" appearing in various places in this specification do not necessarily refer to the same embodiment, but may refer to the same embodiment. In addition, in one or more embodiments, the particular features, structures, or characteristics can be combined in any appropriate manner, as would be apparent to one of ordinary skill in the art from this disclosure.
[0063] The technology provided in the embodiments of the present application is applied to, for example, autonomous driving vehicles, including fully automated unmanned vehicles and partially automated, conditionally automated, and highly automated vehicles. In addition, the trajectory planning method provided in the embodiments of the present application can also be applied to other autonomously moving objects such as robots and drones. The following is an example of an application to an autonomous driving vehicle.
[0064] In order to accurately describe the technical content of this application and to accurately understand this application, the following explanations or definitions are given for the terms used in this specification before a specific description is given.
[0065] Trajectory planning: planning the trajectory that the autonomous vehicle will travel in the future, including time and space information;
[0066] Prediction uncertainty: The trajectory prediction of the surrounding vehicle may disagree with its future trajectory. Figure 1 One-to-one correspondence presents a probability distribution, not a definite prediction;
[0067] Speed planning: The trajectory planning problem is divided into two sub-problems: path planning and speed planning. Speed planning refers to planning the speed curve when driving on this path;
[0068] Potential field: often used in the field of path planning, setting up potential field functions for obstacles to repel the vehicle to achieve the purpose of obstacle avoidance;
[0069] St graph: The vertical axis is displacement s (unit: m (meter)), which represents the cumulative length along the currently planned path; the horizontal axis is time t (unit: s (second)), which represents the passage of time relative to the current moment.
[0070] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present application. In case of any inconsistency, the meaning described in this specification or the meaning derived from the contents recorded in this specification shall prevail. In addition, the terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.
[0071] First, refer to Figure 1 , Figure 2 The application scenarios of the embodiments of the present application are described.
[0072] Figure 1 Schematic diagram of an application scenario of the embodiment of the present application. Figure 1 As shown, a vehicle 10 (hereinafter sometimes referred to as the self-driving vehicle 10) is traveling on a road. There is a fallen object 20 as a stationary object located in front of the lane where the self-driving vehicle 10 is traveling on the road, and another vehicle 30 (hereinafter sometimes referred to as the other vehicle 30) located in an adjacent lane as an environmental vehicle. The fallen object 20 becomes a static obstacle that may collide with the self-driving vehicle 10. When the self-driving vehicle 10 plans a planned path to partially enter the adjacent lane in order to avoid the fallen object 20, the other vehicle 30 becomes a dynamic obstacle that may collide with the self-driving vehicle 10. Therefore, when the self-driving vehicle 10 plans a trajectory to avoid collision with the fallen object 30, it is necessary to predict the driving trajectory of the other vehicle 30 to plan the driving speed of the self-driving vehicle 10 to avoid collision with the other vehicle 30.
[0073] exist Figure 1 In the scenario, the future driving trajectory of the other vehicle 30 predicted by the vehicle 10 is, for example, parking on the roadside as indicated by the dashed line with an arrow. Therefore, the vehicle 10 can plan a trajectory to avoid the left object 20 as indicated by the dashed line with an arrow based on the predicted driving trajectory of the other vehicle 30.
[0074] However, sometimes, Figure 1 Unlike the scenario shown, the driving intention of the other vehicle 30 on the road is unclear. Figure 2 Schematic diagram of another application scenario of the embodiment of the present application. Figure 2 In the scenario, the vehicle 10 is going to go straight along the path R0 at the intersection, and there is another vehicle 30 on the road that intersects the road where the vehicle 10 is traveling. As for the other vehicle 30, its driving intention at the intersection is uncertain. It may turn left, go straight, or turn right. The vehicle 10 has three predicted trajectories for the other vehicle 30, namely, the left turn predicted trajectory R1, the straight-ahead predicted trajectory R2, and the right turn predicted trajectory R3. Therefore, the predicted trajectory of the other vehicle 30 by the vehicle 10 has prediction uncertainty and is probabilistically distributed. Moreover, as the time relative to the current moment increases, the confidence of the predicted trajectory will be greatly reduced. If these predicted trajectories are treated equally during trajectory planning, the speed of the vehicle 10 will be too conservative, and the driving efficiency will be greatly reduced on complex urban roads, and there may even be a problem that the vehicle 10 cannot drive.
[0075] Therefore, in the present application, when planning the trajectory of the own vehicle 10 based on the predicted trajectory of the other vehicle 30 with prediction uncertainty, the probability distribution of each predicted trajectory of the other vehicle 30 and the length of time relative to the current moment are considered to reduce the impact of the predicted trajectory with a lower probability on the speed planning of the own vehicle 10, thereby improving driving efficiency.
[0076] Below, refer to Figure 3-12 The trajectory planning method, trajectory planning device and computing device provided in the embodiments of the present application are described in detail.
[0077] The trajectory planning method provided in the embodiment of the present application is executed by, for example, a trajectory planning device 500 mounted on a vehicle 10 serving as an autonomous driving vehicle.
[0078] Figure 3 It is a schematic diagram of the structure of a vehicle 10 equipped with a trajectory planning device 500 provided in an embodiment of the present application.
[0079] like Figure 1 As shown, the vehicle 10 has a perception device 200 , a prediction device 300 , a decision device 400 , a trajectory planning device 500 and a motion control device 600 .
[0080] The sensing device 200 is used to obtain the driving environment information and the vehicle status information of the vehicle 10. The sensing device 200 includes, for example, a camera, a radar, a millimeter-wave radar, a laser radar, an accelerometer, a gyroscope and other sensing devices. The sensing device 200 can use, for example, various sensors in the advanced driving assistance system installed on the vehicle 10.
[0081] The driving environment information includes, for example, information about static objects and dynamic objects on the road on which the vehicle 10 is traveling. For example, the static object information includes information about stationary objects such as objects that have fallen on the road and traffic signs. For example, the dynamic object information includes information about traffic participants other than the vehicle 10, such as pedestrians and other vehicles on the road. The driving environment information specifically includes information such as the direction, size, speed, and position of each object. The driving environment information acquired by the perception device 200 can be sent to the prediction device 300, the decision device 400, and the trajectory planning device 500.
[0082] The vehicle state information includes speed information and position information of the vehicle 10. The speed information may include, for example, angular velocity, linear velocity, and linear acceleration. The position information may include, for example, information on the coordinates of the vehicle 10 in the earth coordinate system. The vehicle state information acquired by the perception device 200 may be sent to the decision device 400 and the trajectory planning device 500.
[0083] The prediction device 300 is used to predict the future predicted trajectory of the obstacle based on the driving environment information obtained by the perception device 00. The obstacle here refers to an object that may collide with the vehicle 10. The predicted trajectory includes information such as the predicted path and predicted speed of the obstacle.
[0084] In the case where there are multiple predicted trajectories of the obstacle, the prediction device 300 also predicts the probability distribution of each predicted trajectory. The prediction device 300 can use a known method to predict the probability of the obstacle moving along each predicted trajectory. For example, a multi-layer perceptron can be used to input the current state information of the obstacle and the road network structure near it, and through a neural network, the weights of all possible trajectories of the obstacle are output, and after normalization, the probability distribution of the predicted trajectory is obtained. The current state of the obstacle includes, for example, direction, size, speed, and position. The predicted trajectory and probability information of the obstacle predicted by the prediction device 300 can be sent to the decision device 40 and the trajectory planning device 500.
[0085] The decision device 400 is used to make decisions such as lane change, overtaking, deceleration and avoidance based on information such as driving environment information and vehicle status information. Information related to lane change, overtaking, deceleration and avoidance is called decision information. The decision information can be sent to the trajectory planning device 500.
[0086] The trajectory planning device 500 is used to plan a trajectory for the vehicle 10 to travel to a destination. The trajectory planning device 500 sends the planned trajectory to the motion control device 600. The detailed structure of the trajectory planning device 500 will be described later.
[0087] The motion control device 600 is used to control the vehicle 10 to travel along the trajectory planned by the trajectory planning device 500 .
[0088] Refer to the following Figure 4 The detailed structure of the trajectory planning device 500 provided in the embodiment of the present application is described.
[0089] Figure 4 : is a schematic diagram of the structure of the trajectory planning device 500 provided in an embodiment of the present application. The trajectory planning device 500 may be, for example, composed of a computing device having a processor, a memory, a communication interface, and a bus. The computing device performs the function of the trajectory planning device 500 by executing a program stored in a storage unit by the processor, and executes the trajectory planning method provided in an embodiment of the present application. The computing device may be, for example, an on-board computing platform of the vehicle 10.
[0090] like Figure 4As shown, the trajectory planning device 500 includes a path planning module 520, a speed planning module 530 and a trajectory synthesis module 540. The speed planning module 530 includes a prediction information mapping unit 532, a speed obstacle potential field construction unit 534, a St curve calculation unit 536 and a Vt curve calculation unit 538.
[0091] The path planning module 520 is used to plan the driving path of the vehicle 10, including global path planning and local path planning. When the vehicle 10 sets out from the starting point, the path planning module 520 plans the global planning path to the destination. When the vehicle 10 is driving to the destination, the path planning module 520 plans the local planning path according to the global planning path, real-time driving environment information, and decision-making information in a prescribed cycle. The local planning path is roughly planned along the global planning path, but in cases where it is necessary to avoid obstacles, overtake, etc., it will deviate from the global planning path.
[0092] Figure 5 It is a schematic diagram of the path planning scenario provided in an embodiment of the present application. Figure 5 In the example, the vehicle 10 is traveling in the lane 58R, and there is another vehicle 30a in front of it. In the lane 58L adjacent to the lane 58R, there is another vehicle 30b traveling in the opposite direction of the vehicle 10. Here, if the current speed of the vehicle 10 is higher than that of the other vehicle 30a, the vehicle 10 will collide with the other vehicle 30a in the future. Therefore, in order to avoid the collision with the other vehicle 30a, the path planning module 520 of the vehicle 10 plans a local planning path R that uses the lane 58L to overtake the other vehicle 30a.
[0093] The speed planning module 530 is used to plan the driving speed of the vehicle 10 on the local planned path according to the local planned path at the current moment (current cycle) and the predicted trajectory of the obstacle. Specifically, the speed planning module 530 obtains the St map based on the local planned path at the current moment, maps the speed obstacle blocks described later that will affect the driving speed of the vehicle 10 in the St map, and constructs repulsive potential fields of different strengths for each speed obstacle block in the St map according to the probability of the predicted trajectory of the obstacle corresponding to the speed obstacle block, and then calculates the St curve of the vehicle 10 according to the repulsive potential field, and obtains the Vt curve of the vehicle 10 by differentiating the St curve.
[0094] The construction of the speed barrier block is explained. As described above, the predicted trajectory of the obstacle includes the predicted path and predicted speed of the obstacle. When there is an intersection path between the predicted path of the obstacle and the local planned path of the vehicle 10 at the current moment, the speed barrier block is constructed according to the intersection path. The case of "intersecting paths" here is, for example, the situation where the predicted path of the obstacle and the local planned path of the vehicle 10 intersect, at least partially overlap, etc. And the "intersecting path" can be understood, for example, as the part where the predicted path of the obstacle and the local planned path of the vehicle 10 overlap each other. In this case, at least part of the obstacle is on the local planned path of the vehicle 10 in a certain time period. The speed barrier block is calculated based on the time period when the volume of obstacles such as other vehicles overlaps with the local planned path of the vehicle 10. The speed barrier block can be characterized, for example, by the corresponding relationship between the position and time of the obstacle on the intersection path. Here, it should be noted that: even when there is only one obstacle, the number of speed obstacle blocks may be more than one, and the number of speed obstacle blocks depends on the number of predicted trajectories of all predicted trajectories of the obstacle that have intersection paths with the planned path of the vehicle 10.
[0095] As described above, the speed planning module 530 has a prediction information mapping unit 532, a speed obstacle potential field construction unit 534, an St curve calculation unit 536, and a Vt curve calculation unit 538. The prediction information mapping unit 532 is used to map the speed obstacle block in the St map according to the local planning path and the predicted trajectory information of the obstacle; the speed obstacle potential field construction unit 534 is used to construct a repulsive potential field for each speed obstacle block according to the probability of the predicted trajectory of each obstacle; the St curve calculation unit 536 is used to calculate the St curve of the vehicle 10 according to the repulsive potential field; and the Vt curve calculation unit 538 is used to calculate the Vt curve of the vehicle 10 according to the St curve.
[0096] Figure 6 is with Figure 5 The St graph corresponding to the scene. Figure 5 In the scenario of , the predicted trajectory of the other vehicle 30a overlaps with the planned path of the own vehicle 10 before and after the own vehicle 10 overtakes, that is, the other vehicle 30a becomes a speed obstacle block that affects the driving speed of the own vehicle 10. In addition, the predicted trajectory of the other vehicle 30b traveling in the lane 58L also overlaps with the planned path of the own vehicle 10, that is, the other vehicle 30b also becomes a speed obstacle block that affects the driving speed of the own vehicle 10. Therefore, Figure 6 As shown in FIG. 1 , in the St graph, two speed obstacle blocks B1 and B3 corresponding to the other vehicle 30a and a speed obstacle block B2 corresponding to the other vehicle 30b can be mapped. Based on the St graph with three speed obstacle blocks mapped, for example, the St curve of the vehicle 10 shown by the solid line in the figure can be calculated. Although not shown in the figure, by Figure 6 By differentiating the St curve in the figure, we can obtain the Vt curve of the vehicle 10 and plan the vehicle 10 in Figure 5 The driving speed on the local planning path R in .
[0097] The trajectory synthesis module 540 is used to synthesize the local planned path planned by the path planning module 520 and the driving speed on the local planned path planned by the speed planning module 532 to obtain the planned trajectory of the vehicle 10. In addition, the planned trajectory of the vehicle 10 is sent to the motion control device 600.
[0098] Below, refer to Figure 7 to Figure 11 The specific processing actions when the trajectory planning device 500 executes the trajectory planning method provided in the embodiment of the present application are described.
[0099] Figure 7 It is a schematic flow chart of the trajectory planning method provided in an embodiment of the present application. Figure 8 It is a schematic diagram of a specific application example of the trajectory planning method provided in an embodiment of the present application. Fig. 9 is with Figure 8 St diagram corresponding to the application example. Fig.10 is with Figure 8 Schematic diagram of St curve corresponding to the application example. Fig.11 is with Figure 8 Schematic diagram of the Vt curve corresponding to the application example.
[0100] exist Figure 8 In the example shown, the vehicle 10 is traveling in lane 78c of road 78 at a T-junction where roads 68 and 78 intersect, and is about to turn left into lane 68b of road 68. Another vehicle 30b is traveling in lane 68b, and another vehicle 30a is traveling in lane 68c.
[0101] Combine the following Figure 8 To illustrate with an example Figure 7 Processing of each step in the flowchart.
[0102] Step S10: The path planning module 520 performs local path planning at the current moment according to the driving environment information, the vehicle state information, the decision information, etc. In this step, some conventional path planning methods can be used, such as sampling-based path planning methods, search-based path planning methods, etc.
[0103] exist Figure 8 In the example of FIG. 1 , the path planning module 520 plans a path according to the situation that the vehicle 10 wants to turn left. Figure 8 The planned path R0 of the vehicle is to turn left from lane 78c into lane 68b.
[0104] Step S20: The prediction information mapping unit 532 of the speed planning module 530 maps the speed obstacle block to the St map based on the local planning path at the current moment according to the planned local planning path at the current moment and the predicted trajectory of the obstacle. Specifically, based on the local planning path at the current moment and the predicted trajectory of obstacles such as other vehicles, the time period in which the volume of the obstacle overlaps with the local planning path of the vehicle 10, i.e., the block, is calculated, and the speed obstacle block is formed by these blocks and mapped to the St map.
[0105] For example, time can be discretized, and for each discrete time point, the position of the obstacle at that time point, as well as its shape and size, can be found on the predicted trajectory. If the obstacle occupies a length on the planned path of the vehicle 10 at that discrete time point, the two endpoints of the length at this discrete time point are calculated. The envelope area formed by calculating the endpoints of all such lengths is the speed obstacle block.
[0106] exist Figure 8 In the example, there are two predicted trajectories of the other vehicle 30a, one is the predicted trajectory Ra1 of the other vehicle going straight on lane 68c, and the other is the predicted path Ra2 of the other vehicle turning right from lane 68c into lane 78b. In addition, the predicted trajectory of the other vehicle 30b is the predicted trajectory Rb of the other vehicle going straight on lane 68b. Among the three predicted trajectories of the other vehicle, there are two that overlap with the planned path R0 of the own vehicle, namely the predicted trajectories Ra1 and Rb of the other vehicle. Therefore, as shown in Figure 9, when the other vehicle 30a goes straight according to the predicted trajectory Ra1, it is mapped as a speed obstacle block B1 in the St map obtained based on the planned path R0 of the own vehicle 30 at the current moment. In addition, when the other vehicle 30b goes straight according to the predicted trajectory Rb of the other vehicle, it is mapped as a speed obstacle block B2 in the St map.
[0107] Step S30: The speed obstacle potential field construction unit 534 constructs speed obstacle potential fields of different strengths, namely, repulsion potential fields, for each speed obstacle block in the St graph according to the probability of the predicted trajectory of each obstacle.
[0108] Specifically, for each speed obstacle block, a repulsive potential field function is constructed according to the probability of the predicted trajectory of the corresponding obstacle. As the probability increases, the repulsive force of the repulsive potential field is enhanced, and as the time moves toward the future relative to the current moment, the repulsive force is weakened. In this embodiment, the designed potential field function is:
[0109]
[0110] Among them, a prob Represents the probability of the predicted trajectory, with a value range of (0,1]; d centor (s, t) represents the distance from (s, t) to the center of the speed obstacle block; dedge (s, t) represents the distance from (s, t) to the edge of the speed obstacle block; Indicates the boundary of the influence area of the edge of the speed barrier block. When it exceeds this boundary, that is, When p v (s, t) = 0, that is, the vehicle 10 is not affected by the speed obstacle block; t max Indicates the maximum duration of the predicted trajectory. When this maximum duration is exceeded, t>t max When p v (s, t)=0, the vehicle 10 is not affected by the speed obstacle block.
[0111] According to formula (1), the entire potential field function p v (s, t) consists of three parts. The first part indicates that the repulsive potential field will change with the distance from the center of the speed barrier block; the second part indicates that the repulsive potential field will decay as the distance to the closest edge of the speed barrier block increases; the third part indicates that the potential field will decay as the time relative to the current moment increases, because the accuracy of the prediction is lower the further into the future. The probability of predicting the trajectory a prob As a constant of the potential field function, it will also affect the strength of the repulsive potential field. The greater the probability, the greater the strength of the repulsive potential field. In addition, in formula (1), t max It is equivalent to the time threshold, and is used to set the maximum time length that the vehicle 10 is affected by the speed obstacle block.
[0112] exist Figure 8 In the example, potential field functions are constructed for speed obstacle blocks B1 and B2 respectively. When constructing the potential field function for the speed obstacle block B1 corresponding to the predicted trajectory Ra1 of the other vehicle 30a, since there are two predicted trajectories of the other vehicle 30a, 1>a prob >0. In addition, when constructing the potential field function for the speed obstacle block B2 corresponding to the other vehicle 30b, since there is only one predicted trajectory of the other vehicle 30b, a prob =1. In addition, Figure 8 In the example, the origin is equivalent to the current moment.
[0113] Step S40: The St curve calculation unit 536 designs the objective function in the St graph by using numerical optimization, performs optimization and solves, and obtains the St curve of the vehicle at the current moment.
[0114] In this embodiment, the designed objective function is as follows:
[0115]
[0116] Among them, c p (s i,t i ) represents the potential energy term calculated according to the potential field function designed in step S30, c a (s i ,t i ) is the acceleration penalty term, c v (s i ,t i ) is the cruise term, ω1, ω2, ω3 are the weight coefficients of each term, i is the discrete point number of the discretized future time, and its value is 1, 2…n, t i is the moment of the ith discrete point, s i is the displacement at the ith discrete point.
[0117] According to formula (2), the designed objective function consists of three parts. The first part is the designed potential field function, which is used to avoid the speed obstacle block mapped in the St diagram; the second part is the acceleration term, which is used to ensure the ride comfort of the planned speed curve; the third part is the speed term, which is used to drive at the desired cruising speed as much as possible to ensure traffic efficiency. Therefore, each weight coefficient can be set according to different focus points. For example, if the focus is on ensuring ride comfort, ω2 can be set to a value larger than ω1 and ω3; if the focus is on effectively avoiding obstacles, ω1 can be set to a value larger than ω2 and ω3; if the focus is on ensuring traffic efficiency, ω3 can be set to a value larger than ω1 and ω2. In addition, ω1, ω2, and ω3 can be obtained through experimental debugging, for example, according to the type of vehicle and the performance of each functional module on the vehicle.
[0118] Based on the above designed objective function, the speed planning problem can be transformed into an optimization problem, and the optimal St curve can be solved by using nonlinear optimization methods such as the interior point method, or by using the gradient descent method.
[0119] exist Figure 8 In the example of , there are two speed obstacle blocks B1 and B2. In the embodiment of the present application, Fig.10 As shown, through the processing of step S40, the St curve shown by the solid line curve without a mark can be obtained. Fig.10 As a reference, the St curve shown by the solid line curve with a triangle mark below is also shown. The reference curve is the St curve obtained by other speed planning algorithms that do not take into account the prediction uncertainty of the predicted trajectory. By comparing the two St curves, it can be seen that the St curve obtained by the trajectory planning method provided by the embodiment of the present application can travel a longer distance in the same time, thereby improving driving efficiency.
[0120] In addition, you can also Fig.10The reference curve in is used as the initial value for the optimization solution in this step to make the iterative solution converge faster during real-time calculation, thereby reducing the computational load.
[0121] Step S50: The Vt curve calculation unit 538 differentiates the St curve to obtain the Vt curve of the vehicle.
[0122] exist Figure 8 In the example, by Fig.10 Differentiating the St curve shown by the unmarked solid line curve, we can obtain Fig.11 Vt curve shown.
[0123] Step S60 : the trajectory synthesis module 540 synthesizes the obtained Vt curve and the path planned in step S10 into a trajectory, and sends it to the motion control device 600 .
[0124] exist Figure 8 In the example of Fig.11 The Vt curve in Figure 8 The planned trajectory of the vehicle 10 including the time and space position is obtained by synthesizing the planned path R0 of the vehicle 10 in the figure.
[0125] After step S60, the current trajectory planning process ends.
[0126] also, Figure 7 The flowchart of is periodically executed at a certain frequency until the vehicle 10 arrives at the destination. Thus, the real-time performance of the trajectory planning can be guaranteed.
[0127] The trajectory planning method in this embodiment is described above, but the trajectory planning method provided in the embodiment of the present application is not limited thereto.
[0128] In some embodiments, other potential field functions may also be used in step S30. For example, the potential field function may be designed by not considering the magnitude of the time transition relative to the current moment to the future, but only considering the probability distribution of the predicted trajectory.
[0129] In some implementations, with respect to the potential field function p v Parameters in (s,t) and t max , different values can be set according to the type of obstacle. For example, when the obstacle is a person or a vehicle, the impact on the speed of the vehicle 10 is different due to the different moving speeds of the two. Therefore, the above parameters can be changed according to whether the obstacle is a person or a vehicle. and t max Thus, the driving efficiency of the vehicle 10 can be improved by adapting to the movement rules of pedestrians and vehicles as different traffic participants.
[0130] In some implementations, with respect to the potential field function p v Parameters in (s,t) and t max , the value of this parameter can be changed according to the characteristics of the prediction module, such as the confidence of the prediction module in predicting the trajectory.
[0131] In some embodiments, other objective functions may be used in step S40 to calculate the St curve of the vehicle 10. For example, the objective function may also be composed of two parts, one part is a potential energy term corresponding to the designed potential field function, and the other part is an acceleration penalty term or a cruise term.
[0132] In some embodiments, the above-mentioned trajectory planning method may not be executed by the trajectory planning device 500 set on the vehicle 10, but may be executed in a server or cloud connected to the vehicle 10 in a communicative manner, and then the trajectory planning result is sent to the vehicle 10 through the network.
[0133] The above embodiments are summarized as follows.
[0134] The present application embodiment provides a trajectory planning method, which is applied to a first mobile body capable of autonomous movement (such as the vehicle 10 in the above embodiment), and includes a path planning step (such as Figure 7 The step S10 in the above example is a step of planning the moving speed of the first moving body on the planned path (e.g. Figure 7 In the steps S20 to S50 of the above embodiment, in the speed planning step: obtaining the predicted trajectory of the second moving body (such as the other vehicles 30a and 30b in the above embodiment) and the probability that the second moving body moves according to the predicted trajectory, wherein the predicted trajectory includes the predicted path and predicted speed of the second moving body; when there is an intersection path between the planned path and the predicted path, constructing a speed barrier block according to the planned path and the predicted trajectory, wherein the speed barrier block is characterized by the corresponding relationship between the position and time of the second moving body on the intersection path; using a potential field function to construct a repulsive potential field of the speed barrier block, wherein the potential field function uses position and time as independent variables, the probability as a constant, and the strength of the repulsive potential field as a dependent variable, and the larger the value of the probability, the stronger the strength of the repulsive potential field; according to the repulsive potential field and the planned path, planning the moving speed of the first moving body on the planned path. In the above embodiment, "the accumulated length along the currently planned path" corresponds to the position as the independent variable of the potential field function.
[0135] The first moving object may be an autonomously moving vehicle, an autonomously moving drone, a robot, etc. The second moving object may be a person, etc., other than a vehicle.
[0136] By adopting this trajectory planning method, the strength of the repulsive potential field of the second mobile body on the first mobile body is related to the probability that the second mobile body moves according to the predicted trajectory. The larger the value of the probability is, the stronger the strength of the repulsive potential field is. Therefore, it can better adapt to the uncertainty of the predicted trajectory of the second mobile body, reduce the influence of the predicted trajectory of the second mobile body with a smaller probability on the moving speed of the first mobile body, so that the speed of the first mobile body will not be too conservative, and the moving efficiency of the first mobile body is improved.
[0137] Optionally, the “planning the moving speed of the first moving object on the planned path” is to plan the correspondence between the position and time of the first moving object on the planned path according to the repulsive potential field and the planned path, and calculate the moving speed of the first moving object according to the correspondence.
[0138] By adopting this method, the moving speed of the first moving body in each period can be planned in real time with a small amount of calculation, thereby improving the moving efficiency of the first moving body.
[0139] Optionally, the “position on the intersection path” is represented by the distance along the planned path to the position of the first moving body at the current moment.
[0140] Optionally, the potential field strength decreases as the time relative to the current moment increases.
[0141] By adopting this method, the influence of the predicted trajectory of the second moving body in the future on the speed planning of the first moving body can be reduced, and the influence of the predicted trajectory with lower confidence can be reduced, so that the first moving body can be kept at the current speed as much as possible, and then accelerate or decelerate when the moving environment gradually becomes clear, thereby reflecting the interaction with other traffic participants to a greater extent and improving the mobility efficiency.
[0142] Optionally, when the time lapse exceeds a time threshold, the potential field strength is zero. The time threshold may be determined according to a maximum duration of the predicted trajectory of the first moving object.
[0143] By adopting this method, when the time threshold is exceeded, the influence of the predicted trajectory of the second moving body on the speed planning of the first moving body is not considered. Therefore, the influence of the predicted trajectory of the second moving body whose confidence decreases significantly with the increase of time can be avoided, thereby improving the movement efficiency of the first moving body.
[0144] Optionally, the time threshold is set according to the type of the second moving object.
[0145] In this way, the time threshold is set according to the type of second moving objects with different movement laws, such as pedestrians and vehicles, so that the moving speed of the first moving object can be more reasonably planned to adapt to objects with different movement laws.
[0146] The present application also provides a trajectory planning device, which is applied to a first moving body capable of autonomous movement (e.g., the self-vehicle 10 in the above embodiment), and includes a path planning module (e.g., the path planning module 520 in the above embodiment) for planning the moving path of the first moving body and a speed planning module (e.g., the speed planning module 530 in the above embodiment) for planning the moving speed of the first moving body on the planned path, wherein the speed planning module performs the following processing: obtaining a predicted trajectory of a second moving body (e.g., the other vehicles 30a, 30b in the above embodiment) and the probability that the second moving body moves according to the predicted trajectory, wherein the predicted trajectory includes the first moving body; The predicted path and predicted speed of two moving bodies; when there is an intersection path between the planned path and the predicted path, a speed barrier block is constructed according to the planned path and the predicted trajectory, wherein the speed barrier block is characterized by the corresponding relationship between the position and time of the second moving body on the intersection path; a repulsive potential field of the speed barrier block is constructed using a potential field function, wherein the potential field function takes position and time as independent variables, the probability as a constant, and the strength of the repulsive potential field as a dependent variable, and the greater the value of the probability, the stronger the strength of the repulsive potential field; according to the repulsive potential field and the planned path, the moving speed of the first moving body on the planned path is planned. The position as the independent variable of the potential field function is, for example, the "accumulated length along the currently planned path" in the above embodiment.
[0147] By adopting this trajectory planning device, the strength of the repulsive potential field of the second mobile body on the first mobile body is related to the probability that the second mobile body moves according to the predicted trajectory. The larger the value of the probability is, the stronger the strength of the repulsive potential field is. Therefore, it can better adapt to the uncertainty of the predicted trajectory of the second mobile body, reduce the influence of the predicted trajectory of the second mobile body with a smaller probability on the moving speed of the first mobile body, so that the speed of the first mobile body will not be too conservative, and the moving efficiency of the first mobile body is improved.
[0148] Optionally, the “planning the moving speed of the first moving object on the planned path” is to plan the correspondence between the position and time of the first moving object on the planned path according to the repulsive potential field and the planned path, and calculate the moving speed of the first moving object according to the correspondence.
[0149] By adopting this method, the moving speed of the first moving body in each period can be planned in real time with a small amount of calculation, thereby improving the moving efficiency of the first moving body.
[0150] Optionally, the “position on the intersection path” is represented by the distance along the planned path to the position of the first moving body at the current moment.
[0151] Optionally, the potential field strength decreases as the time relative to the current moment increases.
[0152] By adopting this method, the influence of the predicted trajectory of the second moving body in the future on the speed planning of the first moving body can be reduced, and the first moving body can be kept at the current speed as much as possible. When the moving environment gradually becomes clear, the acceleration and deceleration response can be made, thereby reflecting the interaction with other traffic participants to a greater extent and improving the movement efficiency.
[0153] Optionally, when the time lapse exceeds a time threshold, the potential field strength is zero. The time threshold may be determined according to a maximum duration of the predicted trajectory of the first moving object.
[0154] By adopting this method, when the time threshold is exceeded, the influence of the predicted trajectory of the second moving body on the speed planning of the first moving body is not considered. Therefore, the influence of the predicted trajectory of the second moving body whose confidence decreases significantly with the increase of time can be avoided, thereby improving the movement efficiency of the first moving body.
[0155] Optionally, the time threshold is set according to the type of the second moving object.
[0156] In this way, the time threshold is set according to the type of second moving objects with different movement laws, such as pedestrians and vehicles, so that the moving speed of the first moving object can be more reasonably planned to adapt to objects with different movement laws.
[0157] An embodiment of the present application also provides a mobile object having the above-mentioned trajectory planning device.
[0158] By adopting this moving body, the speed of the moving body can be prevented from being too conservative, thereby improving the moving efficiency.
[0159] An embodiment of the present application also provides a computing device. Fig.12 2 is a schematic structural diagram of a computing device 2000 provided in an embodiment of the present application. The computing device 2000 includes: a processor 2100, a memory 2200, a communication interface 2300, and a bus 2400.
[0160] It should be understood that Fig.12 The communication interface 2300 in the computing device 2000 shown may be used to communicate with other devices.
[0161] The processor 2100 may be connected to the memory 2200. The memory 2200 may be used to store the program code and data. Therefore, the memory 2200 may be a storage unit inside the processor 2100, or an external storage unit independent of the processor 2100, or a component including a storage unit inside the processor 2100 and an external storage unit independent of the processor 210.
[0162] Optionally, the computing device 2000 may further include a bus 2400. The memory 2200 and the communication interface 2300 may be connected to the processor 2100 via the bus 2400. The bus 2400 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus 2400 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.12 The fact that only one line is used in the diagram does not mean that there is only one bus or only one type of bus.
[0163] It should be understood that in the embodiment of the present application, the processor 2100 may adopt a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. Alternatively, the processor 2100 may adopt one or more integrated circuits to execute relevant programs to implement the technical solutions provided in the embodiment of the present application.
[0164] The memory 2200 may include a read-only memory and a random access memory, and provides instructions and data to the processor 2100. A portion of the processor 2100 may also include a nonvolatile random access memory. For example, the processor 2100 may also store information on the device type.
[0165] When the computing device 2000 is running, the processor 2100 executes the computer execution instructions in the memory 2200 to perform the operation steps of the trajectory planning method provided in the embodiment of the present application.
[0166] It should be understood that the computing device 2000 according to the embodiment of the present application can be used as the corresponding subject for executing the trajectory planning method according to each embodiment of the present application, and the above-mentioned and other operations and / or functions of each module in the computing device 2000 are respectively for implementing the corresponding processes of each method of the present embodiment. For the sake of brevity, they will not be repeated here.
[0167] An embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored. The computer program is used to enable a computer to execute the trajectory planning method provided by each embodiment of the present application.
[0168] The computer storage medium of the embodiment of the present application can adopt any combination of one or more computer-readable media. Computer-readable media can be computer-readable signal media or computer-readable storage media. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination of the above. More specific examples (non-exhaustive lists) of computer-readable storage media include: electrical connections with one or more wires, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this document, computer-readable storage media can be any tangible medium containing or storing programs, which can be used by instruction execution systems, devices or devices or used in combination with them.
[0169] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.
[0170] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
[0171] Computer program code for performing the operations of the present application may be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0172] Note that the above are only preferred embodiments of the present application and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present application is described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may also include more other equivalent embodiments without departing from the concept of the present invention, all of which belong to the protection scope of the present invention.
Claims
1. A trajectory planning method, applied to a first moving object capable of autonomous movement, characterized in that: The method comprises a path planning step of planning a moving path of the first moving body and a speed planning step of planning a moving speed of the first moving body on the planned path. In the speed planning step: Acquire a predicted trajectory of a second moving object and a probability that the second moving object moves according to the predicted trajectory, wherein the predicted trajectory includes a predicted path and a predicted speed of the second moving object; When there is an intersection path between the planned path and the predicted path, constructing a speed obstacle block according to the planned path and the predicted trajectory, wherein the speed obstacle block is represented by a corresponding relationship between a position and time of the second moving body on the intersection path; Using a potential field function to construct a repulsive potential field of the speed obstacle block, wherein the potential field function takes position and time as independent variables, takes the probability as a constant, takes the strength of the repulsive potential field as a dependent variable, and the greater the value of the probability, the stronger the strength of the repulsive potential field; The moving speed of the first moving object on the planned path is planned according to the repulsive potential field and the planned path.
2. The trajectory planning method according to claim 1, characterized in that: The "planning the moving speed of the first moving body on the planned path" is to plan the correspondence between the position and time of the first moving body on the planned path according to the repulsive potential field and the planned path, and calculate the moving speed of the first moving body according to the correspondence.
3. The trajectory planning method according to claim 1 or 2, characterized in that: The “position on the intersection path” is represented by the distance along the planned path to the position of the first moving body at the current moment.
4. The trajectory planning method according to claim 1 or 2, characterized in that: The strength of the repulsive potential field decreases as the time relative to the current moment increases.
5. The trajectory planning method according to claim 3, characterized in that: The strength of the repulsive potential field decreases as the time relative to the current moment increases.
6. The trajectory planning method according to claim 4, characterized in that: When the time lapse exceeds a time threshold, the intensity of the repulsive potential field is zero.
7. The trajectory planning method according to claim 5, characterized in that: When the time lapse exceeds a time threshold, the intensity of the repulsive potential field is zero.
8. The trajectory planning method according to claim 3, characterized in that: The time threshold is set according to the type of the second moving object.
9. The trajectory planning method according to claim 4, characterized in that: The time threshold is set according to the type of the second moving object.
10. The trajectory planning method according to claim 5, characterized in that: The time threshold is set according to the type of the second moving object.
11. A trajectory planning device, applied to a first moving body capable of autonomous movement, characterized in that: comprising a path planning module for planning a moving path of the first moving body and a speed planning module for planning a moving speed of the first moving body on the planned path, The speed planning module performs the following processing: Acquire a predicted trajectory of a second moving object and a probability that the second moving object moves according to the predicted trajectory, wherein the predicted trajectory includes a predicted path and a predicted speed of the second moving object; When there is an intersection path between the planned path and the predicted path, constructing a speed obstacle block according to the planned path and the predicted trajectory, wherein the speed obstacle block is represented by a corresponding relationship between a position and time of the second moving body on the intersection path; Using a potential field function to construct a repulsive potential field of the speed obstacle block, wherein the potential field function takes position and time as independent variables, takes the probability as a constant, takes the strength of the repulsive potential field as a dependent variable, and the greater the value of the probability, the stronger the strength of the repulsive potential field; The moving speed of the first moving object on the planned path is planned according to the repulsive potential field and the planned path.
12. The trajectory planning device according to claim 11, characterized in that: The "planning the moving speed of the first moving body on the planned path" is to plan the correspondence between the position and time of the first moving body on the planned path according to the repulsive potential field and the planned path, and calculate the moving speed of the first moving body according to the correspondence.
13. The trajectory planning device according to claim 11 or 12, characterized in that: The “position on the intersection path” is represented by the distance along the planned path to the position of the first moving body at the current moment.
14. The trajectory planning device according to any one of claims 11 or 12, characterized in that: The strength of the repulsive potential field decreases as the time relative to the current moment increases.
15. The trajectory planning device according to claim 13, characterized in that: The strength of the repulsive potential field decreases as the time relative to the current moment increases.
16. The trajectory planning device according to claim 14, characterized in that: When the time lapse exceeds a time threshold, the intensity of the repulsive potential field is zero.
17. The trajectory planning device according to claim 15, characterized in that: When the time lapse exceeds a time threshold, the intensity of the repulsive potential field is zero.
18. The trajectory planning device according to claim 13, characterized in that: The time threshold is set according to the type of the second moving object.
19. The trajectory planning device according to claim 14, characterized in that: The time threshold is set according to the type of the second moving object.
20. The trajectory planning device according to claim 15, characterized in that: The time threshold is set according to the type of the second moving object.
21. A mobile object, characterized in that: A trajectory planning device according to any one of claims 11 to 20.
22. A computing device, characterized in that include: at least one processor; as well as at least one memory storing program instructions, The at least one processor performs the method of any one of claims 1 to 10 by executing the program instructions.
23. A computer-readable storage medium having program instructions stored thereon, characterized in that: The computer executes the method according to any one of claims 1 to 10 by executing the program instructions.
Citation Information
Patent Citations
Energy-saving path planning and obstacle avoiding method for drone
CN109343528A
Path optimization based on constrained smoothing spline for autonomous driving vehicles
CN109521763A
