A vehicle obstacle avoidance path planning method, device and electronic equipment

Abstract

Embodiments of the present application disclose a vehicle obstacle avoidance path planning method and device and electronic equipment, and relate to the technical field of automatic driving. The method comprises the following steps: determining a target obstacle in a current cruising lane; predicting a running state of the target obstacle at each discrete time point within a predetermined time length; determining, based on a current execution path of the ego vehicle, a running state of the ego vehicle at two or more predicted positions of the ego vehicle in a lateral direction of the current cruising lane at each discrete time point; calculating, according to the running state of the target obstacle at each discrete time point, the running state of the ego vehicle at each predicted position of the ego vehicle at each discrete time point, and a preset dangerous degree model, a dangerous degree of the target obstacle to each predicted position of the ego vehicle at each discrete time point; and determining, based on the dangerous degree of the target obstacle to each predicted position of the ego vehicle, an initial obstacle avoidance route of the ego vehicle to the target obstacle within the predetermined time length. The present application is suitable for obstacle avoidance.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a vehicle obstacle avoidance path planning method, device, electronic device, and readable storage medium. Background Technology

[0002] Obstacle avoidance capability is a crucial aspect reflecting the level of autonomous driving technology, and in-lane obstacle avoidance behavior is a vital component of building this capability. Compared to other behaviors, in-lane obstacle avoidance is triggered more frequently in urban scenarios, requiring accurate assessment of the obstacle's impact on the vehicle's movement and precise calculation of the avoidance radius to successfully avoid obstacles within confined spaces. With the continuous expansion of intelligent driving applications, achieving efficient and accurate obstacle avoidance behavior has become a key research focus.

[0003] In existing technologies, the influence range of obstacles on the driving lane is determined by analysis, so as to determine the filling value of each area under the influence of obstacles. This facilitates the analysis of the filling value and road traffic impact parameters to determine the traffic efficiency of the lane. Ultimately, the efficiency of vehicle obstacle avoidance is improved based on the traffic efficiency. However, the vehicle obstacle avoidance efficiency is still relatively low. Summary of the Invention

[0004] In view of this, embodiments of this application provide a vehicle obstacle avoidance path planning method, apparatus, electronic device, and readable storage medium, which can improve the obstacle avoidance efficiency of a vehicle.

[0005] In a first aspect, embodiments of this application provide a vehicle obstacle avoidance path planning method applied to a self-driving vehicle. The method includes: determining a target obstacle within the current cruising lane; predicting the running state of the target obstacle at discrete time points within a predetermined time length based on the running state of the target obstacle at the current moment; determining the running state of the self-driving vehicle at two or more predicted positions along the lateral direction of the current cruising lane at each discrete time point based on the current execution path of the self-driving vehicle; calculating the degree of danger of the target obstacle to each predicted position of the self-driving vehicle at each discrete time point based on the running state of the target obstacle at each discrete time point, the running state of the self-driving vehicle at each predicted position at each discrete time point, and a preset danger level model; and determining an initial obstacle avoidance route for the self-driving vehicle to the target obstacle within the predetermined time length based on the degree of danger of the target obstacle to each predicted position of the self-driving vehicle.

[0006] According to a specific implementation of an embodiment of this application, determining the target obstacle in the current cruise lane includes: determining a decision area for an obstacle avoidance route; detecting each obstacle appearing in the decision area; determining the running intention of each obstacle in the current cruise lane; classifying each obstacle according to its different running intentions; and determining the obstacle corresponding to the specified type of running intention as the target obstacle.

[0007] According to a specific implementation of an embodiment of this application, the decision region includes a longitudinal range and a lateral range; the decision region for determining the obstacle avoidance route includes:

[0008] The longitudinal range of the decision region is determined as follows:

[0009] X beg = -max{v cruise_speed T1, D1}

[0010] X end = max{v cruise_speed T2, D2}

[0011] Among them, X beg X represents the first longitudinal boundary behind the vehicle at the current moment. end v represents the second longitudinal boundary in front of the vehicle at the current moment. cruise_speed T1 is the maximum cruising speed of the vehicle at the current moment, T2 is the preset detection time threshold behind the vehicle, D1 is the preset detection time threshold in front of the vehicle, and D2 is the minimum detection distance behind the vehicle and the minimum detection distance in front of the vehicle.

[0012] The horizontal range of the decision-making region is determined as follows:

[0013] Y beg = -min{D right_max D right_edge}

[0014] Y end = min{D left_max D left_edge}

[0015] Among them, Y beg Y represents the first lateral boundary to the right of the vehicle at the current moment. end D represents the second lateral boundary to the left of the vehicle at the current moment. right_max D represents the maximum horizontal detection distance on the right side. left_max D represents the maximum horizontal detection distance on the left. right_edgeD represents the distance between the lane centerline and the right-hand non-crossable curb. left_edge The distances between the lane centerline and the non-crossable curb on the left are respectively.

[0016] According to a specific implementation of this application, determining the operating state of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point based on the vehicle's current execution path includes: determining the operating state of the vehicle on the current execution path at each discrete time point based on the vehicle's current execution path; and determining the operating state of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point based on the vehicle's operating state on the current execution path at each discrete time point.

[0017] According to a specific implementation of this application, determining the running state of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point based on the vehicle's running state on the current execution path at each discrete time point includes: determining the running state of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point based on the vehicle's longitudinal running state on the current execution path at each discrete time point and two or more lateral positions along the lateral direction of the current cruise lane.

[0018] According to a specific implementation of an embodiment of this application, each discrete time point includes a first time point, and each predicted position of the vehicle at the first time point includes a first predicted position; wherein, calculating the degree of danger of the target obstacle to each predicted position of the vehicle at each discrete time point includes: calculating the degree of danger of the target obstacle to the first predicted position of the vehicle at the first time point based on the following danger level model:

[0019]

[0020] in, The degree of threat posed by the target obstacle to the vehicle's first predicted position; This is the first weighting coefficient; This is the second weighting coefficient; The threat coefficient corresponding to the size type of the target obstacle; This is the first predicted position of the vehicle; The position of the target obstacle within the current cruise lane at the first time point; The adjustment value for the position of the target obstacle is based on the acceleration of the target obstacle and the acceleration of the vehicle. for and The difference matrix, where, This is a matrix consisting of the relative distances between the vehicle and the target obstacle at their x-coordinates and the relative distances between the vehicle and the target obstacle at their y-coordinates. To maintain a safe distance; For safe distance ;

[0021] Among them, safe distance Calculate using the following formula:

[0022]

[0023] Among them, D x For lateral safety distance; D y For longitudinal safety distance; The lateral boundary of the target obstacle; The longitudinal boundary of the target obstacle; The speed of the vehicle. The longitudinal relative velocity between the vehicle and the target obstacle is... The lateral relative velocity between the vehicle and the target obstacle. The preset safety time threshold, The preset comfort acceleration.

[0024] According to a specific implementation of this application, after determining the initial obstacle avoidance route of the vehicle to the target obstacle within the predetermined time length, the method further includes: calculating a first lateral offset of the initial obstacle avoidance route relative to the centerline of the current cruise lane; calculating a second lateral offset of the obstacle avoidance path currently executed by the vehicle relative to the centerline of the current cruise lane; calculating the difference between the first lateral offset and the second lateral offset; if the difference is greater than 0, then incrementing the planning count by 1; if the planning count is equal to a preset count, then planning the optimal obstacle avoidance route of the vehicle based on the initial obstacle avoidance route.

[0025] Secondly, embodiments of this application provide a vehicle obstacle avoidance path planning device applied to a self-driving vehicle. The device includes: a first determining module for determining a target obstacle within the current cruising lane; a prediction module for predicting the running state of the target obstacle at each discrete time point within a predetermined time length based on the running state of the target obstacle at the current moment; a second determining module for determining the running state of the self-driving vehicle at two or more predicted positions along the lateral direction of the current cruising lane at each discrete time point based on the current execution path of the self-driving vehicle; a calculation module for calculating the degree of danger of the target obstacle to each predicted position of the self-driving vehicle at each discrete time point based on the running state of the target obstacle at each discrete time point, the running state of the self-driving vehicle at each predicted position at each discrete time point, and a preset danger level model; and a third determining module for determining the initial obstacle avoidance route of the self-driving vehicle to the target obstacle within the predetermined time length based on the degree of danger of the target obstacle to each predicted position of the self-driving vehicle.

[0026] According to a specific implementation of an embodiment of this application, the first determining module includes: a first determining submodule, used to determine the decision area of ​​the obstacle avoidance route; a detection submodule, used to detect each obstacle appearing in the decision area; a second determining submodule, used to determine the running intention of each obstacle in the current cruise lane; a classification submodule, used to classify each obstacle according to different running intentions; and a third determining submodule, used to determine the obstacle corresponding to the specified type of running intention as the target obstacle.

[0027] According to a specific implementation of an embodiment of this application, the decision region includes a vertical range and a horizontal range; the first determining submodule is specifically used for:

[0028] The longitudinal range of the decision region is determined as follows:

[0029] X beg =- max{v cruise_speed T1, D1}

[0030] X end = max{v cruise_speed T2, D2}

[0031] Among them, X beg X represents the first longitudinal boundary behind the vehicle at the current moment. end v represents the second longitudinal boundary in front of the vehicle at the current moment. cruise_speedT1 is the maximum cruising speed of the vehicle at the current moment, T2 is the preset detection time threshold behind the vehicle, D1 is the preset detection time threshold in front of the vehicle, and D2 is the minimum detection distance behind the vehicle and the minimum detection distance in front of the vehicle.

[0032] The horizontal range of the decision-making region is determined as follows:

[0033] Y beg =- min{D right_max D right_edge}

[0034] Y end = min{D left_max D left_edge}

[0035] Among them, Y beg Y represents the first lateral boundary to the right of the vehicle at the current moment. end D represents the second lateral boundary to the left of the vehicle at the current moment. right_max D represents the maximum horizontal detection distance on the right side. left_max D represents the maximum horizontal detection distance on the left. right_edge The distances between the lane centerline and the right-hand, non-crossable curb. Dleft_edge The distances between the lane centerline and the non-crossable curb on the left are respectively.

[0036] According to a specific implementation of an embodiment of this application, the second determining module includes: a fourth determining submodule, used to determine the running state of the vehicle on the current execution path at each discrete time point based on the current execution path of the vehicle; and a fifth determining submodule, used to determine the running state of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point based on the running state of the vehicle on the current execution path at each discrete time point.

[0037] According to a specific implementation of an embodiment of this application, the fifth determining submodule is specifically used to: determine the running state of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point based on the longitudinal running state of the vehicle on the current execution path at each discrete time point and two or more lateral positions in the lateral direction of the current cruise lane.

[0038] According to a specific implementation of an embodiment of this application, each discrete time point includes a first time point, and each predicted position of the vehicle at the first time point includes a first predicted position.

[0039] Specifically, the calculation module is used to: calculate the degree of danger posed by the target obstacle to the first predicted position of the vehicle at the first time point based on the following danger level model:

[0040]

[0041] in, The degree of threat posed by the target obstacle to the vehicle's first predicted position; This is the first weighting coefficient; This is the second weighting coefficient; The threat coefficient corresponding to the size type of the target obstacle; This is the first predicted position of the vehicle; The position of the target obstacle within the current cruise lane at the first time point; The adjustment value for the position of the target obstacle is based on the acceleration of the target obstacle and the acceleration of the vehicle. for and The difference matrix, where, This is a matrix consisting of the relative distances between the vehicle and the target obstacle at their x-coordinates and the relative distances between the vehicle and the target obstacle at their y-coordinates. To maintain a safe distance; For safe distance ;

[0042] Among them, safe distance Calculate using the following formula:

[0043]

[0044] Among them, D x For lateral safety distance; D y For longitudinal safety distance; The lateral boundary of the target obstacle; The longitudinal boundary of the target obstacle; The speed of the vehicle. The longitudinal relative velocity between the vehicle and the target obstacle is... The lateral relative velocity between the vehicle and the target obstacle. The preset safety time threshold, The preset comfort acceleration.

[0045] According to a specific implementation of an embodiment of this application, the device is further configured to: calculate a first lateral offset of the initial obstacle avoidance route relative to the centerline of the current cruise lane after the third determining module determines the initial obstacle avoidance route of the vehicle within the predetermined time length; calculate a second lateral offset of the obstacle avoidance path currently executed by the vehicle relative to the centerline of the current cruise lane; calculate the difference between the first lateral offset and the second lateral offset; if the difference is greater than 0, increment the planning count by 1; if the planning count is equal to a preset count, plan the optimal obstacle avoidance route of the vehicle based on the initial obstacle avoidance route.

[0046] Thirdly, embodiments of this application provide an electronic device, which includes: a housing, a processor, a memory, a circuit board, and a power supply circuit, wherein the circuit board is disposed inside the space enclosed by the housing, and the processor and the memory are disposed on the circuit board; the power supply circuit is used to supply power to various circuits or devices of the above-mentioned electronic device; the memory is used to store executable program code; the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, for executing the vehicle obstacle avoidance path planning method described in any of the foregoing implementations.

[0047] Fourthly, embodiments of this application provide a computer-readable storage medium storing one or more programs, which can be executed by one or more processors to implement the vehicle obstacle avoidance path planning method described in any of the foregoing implementations.

[0048] The vehicle obstacle avoidance path planning method, device, electronic device, and readable storage medium of this embodiment calculate the degree of danger of the target obstacle to the vehicle at each predicted position at each discrete time point based on the running state of the target obstacle at each discrete time point, the running state of the vehicle at each predicted position at each discrete time point, and a preset danger level model. Then, based on the degree of danger of the target obstacle to the vehicle at each predicted position, the initial obstacle avoidance route of the vehicle to the target obstacle within a predetermined time length is determined. In the process of determining the initial obstacle avoidance route of the vehicle to the target obstacle within the predetermined time length, the running state of the target obstacle at each discrete time point and the running state of the vehicle at each predicted position at each discrete time point are used, and combined with the preset danger level model, the initial obstacle avoidance route of the vehicle to the target obstacle within the predetermined time length is determined. In this way, the obstacle avoidance efficiency of the vehicle can be improved. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart illustrating a vehicle obstacle avoidance path planning method provided in an embodiment of this application.

[0051] Figure 2 An architecture diagram of a vehicle obstacle avoidance path planning scheme provided in a specific embodiment of this application;

[0052] Figure 3 This is a schematic diagram of an obstacle avoidance space according to an embodiment of this application;

[0053] Figure 4 This is a schematic diagram of the lateral predicted position in a specific embodiment of this application;

[0054] Figure 5 This is a schematic diagram of the output of the activation function according to a specific embodiment of this application;

[0055] Figure 6 This is a schematic diagram of a practical application scenario of a self-driving vehicle in a specific embodiment of this application;

[0056] Figure 7 A schematic diagram of the structure of a vehicle obstacle avoidance path planning device provided in an embodiment of this application;

[0057] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0058] The embodiments of this application will now be described in detail with reference to the accompanying drawings. It should be understood that the described embodiments are merely some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0059] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0060] To enable those skilled in the art to better understand the technical concept, implementation scheme and beneficial effects of the embodiments of this application, detailed descriptions are provided below through specific embodiments.

[0061] Figure 1 This is a flowchart illustrating a vehicle obstacle avoidance path planning method provided in an embodiment of this application, as shown below. Figure 1 As shown, the vehicle obstacle avoidance path planning method in this embodiment is applied to a self-driving vehicle.

[0062] In practical applications, it can be used in urban traffic scenarios.

[0063] The method in this embodiment may include:

[0064] S101. Identify the target obstacle within the current cruise lane.

[0065] The current cruise lane is the lane where the vehicle is located. The target obstacle can be a stationary object or a moving object.

[0066] S102. Based on the current running state of the target obstacle, predict the running state of the target obstacle at each discrete time point within a predetermined time length.

[0067] Operating status can include position, velocity, and / or acceleration.

[0068] A module can be installed on the vehicle to predict the running status of the target obstacle at multiple discrete time points over a future period of time, based on the obstacle's current running status.

[0069] The process of predicting the running state of a target obstacle at discrete time points within a predetermined time length can adopt existing methods for predicting the running state of an object, which will not be elaborated here.

[0070] S103. Based on the current execution path of the vehicle, determine the running status of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point.

[0071] The current execution path can be the path planned by the vehicle's path planning module, and the vehicle is currently traveling along this path. The current execution path can include various points in time within a future period, including the vehicle's position, speed, and acceleration.

[0072] Each discrete time point in this step corresponds one-to-one with each discrete time point in step S102. For example, if the discrete time points are t1, t2, t3, ..., tn, then each discrete time point in this step is also t1, t2, t3, ..., tn.

[0073] In this embodiment, at each discrete time point, there are more than two predicted positions along the lateral direction of the current cruise lane, and the running status of each predicted position at each discrete time point is determined according to the current execution path of the vehicle.

[0074] The predicted position in this step may also include position, velocity and / or acceleration, where position includes lateral position and longitudinal position.

[0075] S104. Based on the running status of the target obstacle at each discrete time point, the running status of the vehicle at each predicted position at each discrete time point, and the preset danger level model, calculate the danger level of the target obstacle to each predicted position of the vehicle at each discrete time point.

[0076] At any given time point in each discrete time point, the operational status of the target obstacle is known. At that time point, the operational status of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane is also known. Thus, by combining these two with a preset danger level model, the danger level of the target obstacle to the vehicle at each of the two or more predicted positions at that time point can be calculated.

[0077] Based on the above process, the degree of danger posed by the target obstacle to each predicted position of the vehicle at each discrete time point can be calculated.

[0078] S105. Based on the degree of danger of the target obstacle to each predicted position of the vehicle, determine the initial obstacle avoidance route of the vehicle to the target obstacle within a predetermined time length.

[0079] Based on the degree of danger posed by the target obstacle to each vehicle position at each discrete time point, the initial obstacle avoidance route of the vehicle to the target obstacle within a predetermined time length can be determined, and the vehicle can avoid the target obstacle according to the initial obstacle avoidance route.

[0080] In this embodiment, the interaction between the vehicle and the target obstacle can be simulated over a period of time in the future, and then the optimal strategy at the current moment can be output based on the result of the interaction.

[0081] In this embodiment, based on the operating status of the target obstacle at each discrete time point, the operating status of the vehicle at each predicted position at each discrete time point, and a preset hazard level model, the hazard level of the target obstacle to each predicted position of the vehicle at each discrete time point is calculated. Then, based on the hazard level of the target obstacle to each predicted position of the vehicle, the initial obstacle avoidance route of the vehicle to the target obstacle within a predetermined time length is determined. In determining the initial obstacle avoidance route of the vehicle to the target obstacle within the predetermined time length, the operating status of the target obstacle at each discrete time point and the operating status of the vehicle at each predicted position at each discrete time point are used, combined with the preset hazard level model, to determine the initial obstacle avoidance route of the vehicle to the target obstacle within the predetermined time length. This can improve the obstacle avoidance efficiency of the vehicle. In addition, this embodiment can evaluate the operating status of the vehicle and the operating status of the target obstacle in the same dimension through the preset hazard level model, so that common traffic participants in urban scenarios can be evaluated in the same dimension through the model, thereby making the avoidance decision result more comprehensive and accurate.

[0082] In some examples, identifying a target obstacle within the current cruise lane (S101) may include:

[0083] S101a, Decision area for determining obstacle avoidance routes.

[0084] In some cases, the decision area can be determined based on the vehicle's current location and the actual road conditions.

[0085] S101b, Detect all obstacles appearing within the decision-making area.

[0086] The system detects obstacles within the decision area, and the target obstacle can be selected from among them.

[0087] S101c, Determine the operational intent of each obstacle within the current cruise lane.

[0088] In some examples, the intention to drive may include static, lane follow, cut in, cut out, and cross.

[0089] Based on the operational status of each obstacle at discrete time points within a predetermined time period, the operational intention of each obstacle within the current cruise lane can be determined.

[0090] S101d. Classify obstacles according to their different operational intentions.

[0091] S101e: Identify the obstacle corresponding to the specified type of running intent as the target obstacle.

[0092] For example, obstacles whose intended movement is stationary or straight can be identified as target obstacles.

[0093] In this embodiment, the impact of dynamic and static traffic participants is considered hierarchically within the constructed decision area.

[0094] In some examples, the decision area includes a longitudinal range and a lateral range; the decision area for determining the obstacle avoidance route may include:

[0095] The longitudinal range of the decision region is determined as follows:

[0096] X beg = -max{v cruise_speed T1, D1}

[0097] X end = max{v cruise_speed T2, D2}

[0098] Among them, X beg X represents the first longitudinal boundary behind the vehicle at the current moment. end v represents the second longitudinal boundary in front of the vehicle at the current moment. cruise_speed T1 is the maximum cruising speed of the vehicle at the current moment, T2 is the preset detection time threshold behind the vehicle, D1 is the preset detection time threshold in front of the vehicle, and D2 is the minimum detection distance behind the vehicle and the minimum detection distance in front of the vehicle.

[0099] The horizontal range of the decision-making region is determined as follows:

[0100] Y beg =- min{D right_max D right_edge}

[0101] Y end = min{D left_max D left_edge}

[0102] Among them, Y beg Y represents the first lateral boundary to the right of the vehicle at the current moment. end D represents the second lateral boundary to the left of the vehicle at the current moment. right_max D represents the maximum horizontal detection distance on the right side. left_max D represents the maximum horizontal detection distance on the left. right_edge The distances between the lane centerline and the right-hand, non-crossable curb. Dleft_edge The distances between the lane centerline and the non-crossable curb on the left are respectively.

[0103] In some examples, determining the vehicle's operating state (S103) at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point, based on the vehicle's current execution path, may include:

[0104] S103a. Based on the current execution path of the vehicle, determine the running status of the vehicle on the current execution path at each discrete time point.

[0105] It can collect the running status corresponding to each discrete time point on the current execution path.

[0106] S103b: Based on the vehicle's running status on the current execution path at each discrete time point, determine the running status of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point.

[0107] In this embodiment, based on the vehicle's running status at each discrete time point on the current execution path, and according to a preset strategy, the running status of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point can be determined.

[0108] In some examples, determining the vehicle's operating state at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point, based on the vehicle's operating state on the current execution path at each discrete time point (S103b), may include:

[0109] A1. Based on the longitudinal running state of the vehicle on the current execution path at each discrete time point, and two or more lateral positions in the lateral direction of the current cruise lane, determine the running state of the vehicle at two or more predicted positions in the lateral direction of the current cruise lane at each discrete time point.

[0110] For any one of the discrete time points, the corresponding longitudinal running state can be determined based on the vehicle's running state on the current execution path at that time point. This longitudinal running state is then combined with two or more lateral positions in the lateral direction of the current cruise lane to obtain the running state of the vehicle at two or more predicted positions in the lateral direction of the current cruise lane at that time point.

[0111] Two or more lateral positions in the current cruise lane can be determined by spacing the current cruise lane laterally at equal or unequal intervals. These positions can be used as lateral positions. For example, if the current cruise lane is 5m wide, the lateral range can be determined as 0-5m. In this case, 0m, 2.5m, and 5m can be determined as lateral positions. These can be combined with the longitudinal running status to obtain the predicted position. 0.5m, 3.5m, and 4.5m can also be determined as lateral positions.

[0112] Understandably, in practice, if the current cruise lane is incorporated into a pre-established coordinate system, the lateral range of the current cruise lane can be determined based on its specific position in the coordinate system. For example, if the lateral range of the current cruise lane is symmetrical with respect to the X-axis of the coordinate system, then the lateral range is -2.5m to +2.5m.

[0113] The vehicle's operating status at two or more predicted positions along the lateral direction of the current cruise lane at other times can be determined based on the above process.

[0114] When a vehicle avoids obstacles, it does so by moving the distance laterally. Therefore, the obstacle avoidance strategy in this embodiment is manifested as the distance of lateral movement. By predicting the lateral position and the longitudinal running state, multiple running states that the vehicle may exist at a given moment are determined. The running states include position, speed, and acceleration. In the running state of each predicted position, the position is different, specifically the lateral position is different.

[0115] In some examples, each discrete time point may include a first time point, and each predicted position of the vehicle at the first time point includes the first predicted position; wherein, calculating the degree of danger of the target obstacle to each predicted position of the vehicle at each discrete time point (S104) may include:

[0116] S104a. Based on the following hazard level model, calculate the hazard level of the target obstacle to the vehicle's first predicted position at the first time point:

[0117]

[0118] in, The degree of threat posed by the obstacle to the vehicle's first predicted position; This is the first weighting coefficient; This is the second weighting coefficient; Threat coefficient corresponding to the size and type of the target obstacle; The first predicted position of the vehicle; The position of the target obstacle within the current cruise lane at the first point in time; This is the adjustment value for the position of the target obstacle based on the acceleration of the target obstacle and the acceleration of the vehicle; for and The difference matrix, where, This is a matrix consisting of the relative distances between the vehicle and the target obstacle at their x-coordinates and y-coordinates. To maintain a safe distance; For safe distance ;

[0119] Among them, safe distance Calculate using the following formula:

[0120]

[0121] Among them, D x For lateral safety distance; D y For longitudinal safety distance; The lateral boundary of the target obstacle; The longitudinal boundary of the target obstacle; For the speed of the vehicle, The longitudinal relative velocity between the vehicle and the target obstacle. The lateral relative velocity between the vehicle and the target obstacle. The preset safety time threshold, The preset comfort acceleration.

[0122] For multiple target obstacles, the degree of danger to the vehicle can be calculated separately according to the above process, and the degree of danger of each obstacle can be added together to obtain the total degree of danger value.

[0123] After determining the initial obstacle avoidance path of the vehicle towards the target obstacle within the predetermined time length, the method of this embodiment may further include:

[0124] S106. Calculate the first lateral offset of the initial obstacle avoidance route relative to the centerline of the current cruise lane.

[0125] S107. Calculate the second lateral offset of the obstacle avoidance path currently executed by the vehicle relative to the centerline of the current cruise lane.

[0126] S108. Calculate the difference between the first lateral offset and the second lateral offset.

[0127] S109. If the difference is greater than 0, then increase the planning time by 1.

[0128] S110. If the number of planning attempts equals the preset number of attempts, then based on the initial obstacle avoidance route, plan the optimal obstacle avoidance path for the vehicle.

[0129] When the number of planning attempts reaches the preset number, the optimal obstacle avoidance path for the vehicle is then planned, thereby improving decision-making stability and preventing the vehicle from shaking significantly.

[0130] The following detailed description of the solution in this application is based on a specific embodiment.

[0131] The architecture diagram of the vehicle obstacle avoidance path planning scheme in this embodiment is as follows: Figure 2As shown, the obstacle avoidance behavior in the lane is initiated by the obstacle avoidance decision module, which outputs a recommended strategy based on the vehicle's status, target obstacle information, and lane information. The trajectory planning module then generates a trajectory based on this strategy to avoid the target obstacle.

[0132] The vehicle obstacle avoidance path planning method in this embodiment may include:

[0133] Step 1: Construct the decision region Ω for obstacle avoidance based on the vehicle's current ego position (x0, y0) in both the horizontal and vertical directions.

[0134] The longitudinal range of the decision-making area: from the rear X of the vehicle beg To the front position X of the car end ,in,

[0135] X beg = -max{v cruise_speed T1, D1}

[0136] X end = max{v cruise_speed T2, D2}

[0137] Among them, v cruise_speed The current maximum cruising speed is T1, the vehicle rear detection time threshold is T2, the vehicle front detection time threshold is D1, the minimum detection distance behind the vehicle is D2, and the minimum detection distance in front of the vehicle is D2.

[0138] The lateral range of the decision-making area: from the right side of the vehicle (Y) beg To the left of the car Y end ,in,

[0139] Y beg = -min{D right_max D right_edge}

[0140] Y end = min{D left_max D left_edge}

[0141] Among them, D right_max D represents the maximum horizontal detection distance on the right side. left_max D represents the maximum horizontal detection distance on the left. right_edge D is the distance between the lane centerline and the right-hand non-crossable curb. left_edge This refers to the distance from the curb to the left of the lane centerline that cannot be crossed.

[0142] Step 2: Obstacle avoidance screening and preprocessing:

[0143] 2.1 Match the position of the obstacle at the current time t0 with the decision region range in step 1 to obtain the set of obstacles Mobj = {Mi|i = 1, 2, 3, ..., n} within the decision region.

[0144] 2.2 Considering the driving state of obstacle Mobj within the next time interval T, we define the position, velocity, and acceleration information of the obstacle at each moment as its state. The state sequence of the obstacle within time interval T can then be denoted as Objtraj_state={Oi|i=1, 2, 3, ..., n}. If the obstacle is a dynamic target, its predicted trajectory is sampled at equal intervals to obtain a state sequence of n moments {t0, ..., tn} containing the current moment. If the obstacle is a static target, its future driving state will not change, so its state within time interval T is always the current state O0 at t=0. Aligning the time sequence {t0, ..., tn} of the dynamic target sampling, we map O0 to the above time sequence to obtain the future driving state.

[0145] Based on the relationship between the motion state of the obstacle from time t0 to tn and the lane (circular lane) where the vehicle is located, the obstacle is divided into different intentions. The specific intentions are classified as {static, lane follow, cut in, cut out, cross}, etc. The two types of obstacles, static and lane follow, are retained. Mobj is filtered to obtain the target obstacle set MObj_pre={ Mi|i= 1, 2, 3, ..., n}, where static indicates that the obstacle is stationary and lane follow indicates that the obstacle is in the lane where the vehicle is located and is moving along the extension direction of the lane where the vehicle is located.

[0146] Step 3: Perform forward simulation based on obstacle avoidance strategy space

[0147] 3.1: The horizontal range [Y] of the decision region Ω in step 1. beg Y end The road structure is used to trim the data to generate a set of lateral strategies.

[0148] 3.1.1 Cutting Horizontal Range Y act_dir

[0149] The lateral range of motion includes the left and right boundaries; among which,

[0150] Left boundary point Y act_dir_left = min{D lane_width_dir_left +D buffer D solid_dir_left}–D half_ego ;

[0151] Right boundary point Y act_dir_right= min{D lane_width_dir_right +D buffer D solid_dir_right}–D half_ego ;

[0152] Among them, D lane_width_dir_left D is the distance from the center line of the lane where the vehicle is located to the left edge of that lane. lane_width_dir_right D is the distance from the center line of the lane where the vehicle is located to the right edge of that lane; buffer Compensation distance for lane boundaries; D solid_dir_left D is the distance between the center line of the lane where the vehicle is located and the solid line on the left side of the lane; solid_dir_right D is the distance between the center line of the lane where the vehicle is located and the solid line on the right side of the lane; half_ego It is half the width of the vehicle.

[0153] 3.1.2 When an obstacle avoidance strategy is executed by the vehicle, it manifests as the distance of lateral movement; therefore, there is a correspondence between the strategy and the lateral position during obstacle avoidance. The lateral range Yact_dir is sampled at equal intervals to obtain the lateral strategy Plat = {Pi|i=1, 2, 3, ..., n}.

[0154] 3.2: Generate a longitudinal strategy space based on the vehicle's current driving state (ego).

[0155] 3.2.1 Define the longitudinal position, longitudinal velocity, and longitudinal acceleration of the vehicle as the longitudinal state, and extract the state sequence of the trajectory executed by the vehicle at the previous moment as Slgt = {Si| i=1, 2, 3, ..., n}.

[0156] 3.2.2 The horizontal policy number n obtained in step 3.1.2 is used to expand Slgt to n dimensions, thereby obtaining the vertical policy Plgt = { Pi, j | i=1, 2, 3, ..., n; j=1, 2, 3, ..., n}.

[0157] 3.3: Coarse Trajectory under Simulated Obstacle Avoidance Strategy

[0158] The horizontal policy Plat from step 3.1 and the vertical policy Plgt from step 3.2 are matched according to the following rules: Traverse the horizontal policy Plat. Assume the currently processed horizontal policy is Pk, and bind it with all vertical policies to obtain the combined policy Pk_lgt = {Pk, j | j=1, 2, 3, ..., n}. This obtains all combinations of horizontal and vertical policies. Define coarse trajectory points Jraw{Ji, j | i=1, 2, 3, ..., n; j=1, 2, 3, ..., n} representing candidate decision policies, where I and j of each trajectory point represent the policy indices of Plat and Plgt, respectively.

[0159] Step 4: Output and evaluate candidate strategies based on the obstacle threat model

[0160] 4.1 Building a Real-Time Threat Model

[0161] Given information such as the position, orientation, speed, and type of the vehicle and the target obstacle, the real-time threat level of the obstacle to the vehicle (threat level index) can be modeled as follows:

[0162]

[0163] The above formula consists of two parts: α is the current position of the vehicle (x, y), aobs is a coefficient related to the size and type of the obstacle, μ1 is the obstacle position (x0, y0), and μ2 is the adjustment value (x1, y1) of the obstacle position based on the acceleration of the two vehicles. ∑1 is D diff and D safe The matrix of differences, where ∑2 is the safe distance D safe D diff For the ego of the car and the obstacle M Obj_pre Position, horizontal and vertical relative real-time distance, D safe For the ego of the car and the obstacle M Obj_pre The relative safe distances in the horizontal and vertical directions. D safe The definition is as follows:

[0164]

[0165] Among them, D x For lateral safety distance; D y For longitudinal safety distance; , For the horizontal and vertical boundaries of the obstacle, For the ego speed of the car, The longitudinal relative velocity, The lateral relative velocity, For a safe time threshold, For comfortable acceleration.

[0166] 4.2 Establishing a spatiotemporally distributed threat model

[0167] 4.2.1 The obstacle avoidance decision considers the future for a time length of T and a time interval of m, from which a time series tn can be obtained. Based on this time series, the trajectory of the obstacle to be avoided MObj_pre obtained in step 2 and the coarse trajectory Jraw corresponding to the obstacle avoidance strategy in step 3 are extracted.

[0168] 4.2.2 Constructing the vehicle's single-frame obstacle avoidance space

[0169] Compare the vehicle's ego at time t with the target obstacle M. Obj_preThe position, velocity, and acceleration state information are mapped to the obstacle avoidance decision space Ω obtained in step 1, allowing us to obtain the relative relationships between the vehicle and the obstacle's position, velocity, acceleration, and other states. For example... Figure 3 In the obstacle avoidance space shown, light-colored obstacles represent dynamic obstacles, while dark-colored obstacles represent static obstacles.

[0170] 4.2.3 Establishing a spatiotemporal joint obstacle avoidance model

[0171] Based on the time series {t0, ..., tn} in step 2.2, extract the state information of the target obstacle MObj_pre's position, velocity, and acceleration at each corresponding time point. Meanwhile, since the vehicle's longitudinal direction Plgt obtained in step 3.2 is time-bound, extract the vehicle's longitudinal state at the corresponding time points {t0, ..., tn}.

[0172] Following the method in step 4.2.2, an obstacle avoidance space is established for n future time intervals, extending from time t0 to time tn. This yields the obstacle avoidance space between the vehicle's ego and the target obstacle M when the vehicle executes its ego strategy. Obj_pre For a continuous spatiotemporal joint obstacle avoidance decision-making model, see [link to model]. Figure 3 .

[0173] 4.2.4 Evaluate the trajectory output in step 3

[0174] Based on the vehicle's ego and the target obstacle M obtained in step 4.2.2 Obj_pre The spatial distribution of the vehicle's ego within a continuous time period T is modeled, and all obstacle avoidance strategies in step 3 are modeled. The threat level of the vehicle's ego when it executes the specified strategy is calculated using the threat assessment formula in step 4.1.

[0175] Based on the spatiotemporal joint obstacle avoidance model in step 4.2.3, the strategy with the minimum threat level in each frame is calculated by traversing from t0 to tn, which is the optimal obstacle avoidance strategy for a single frame, and the coarse trajectory Jraw evaluation results are obtained.

[0176] like Figure 4 In the illustrated embodiment, taking time t1 as an example, the obstacle avoidance strategy is executed as follows: the vehicle ego starts from t0 and queries the lateral strategy that matches the longitudinal strategy at time t1. At this time, the lateral obstacle avoidance strategy that the vehicle can execute at time t1 is represented by the dots within the rectangle. After the vehicle ego executes the action, each point reached represents the state of the vehicle ego, including information such as the vehicle's position and speed, which are the coarse trajectory points that can be executed.

[0177] 4.3 Output the optimal trajectory

[0178] Using the vehicle's current ego position as the initial value for lateral decision-making, and considering the constraints of vehicle kinematics, the lateral movement is limited, that is, the maximum lateral movement distance in each step is no greater than Dveh. The time series t0-tn is selected sequentially, and the coarse trajectory with the smallest influence from surrounding obstacles corresponding to each frame is searched as the optimal strategy Jbest.

[0179] In this embodiment, based on a spatiotemporal joint approach, the obstacle avoidance strategy of the candidate vehicle is simulated in the forward direction to achieve interaction with the target obstacle, and then the optimal strategy at the current moment is output based on the interaction result.

[0180] Step 5: Output stable results

[0181] Design an activation function to output a stable bias decision. .

[0182]

[0183] Where drequest is the number of planning iterations incremented by 1 when there is a difference between the first lateral offset of the optimal obstacle avoidance path calculated in this instance and the second lateral offset of the currently executed obstacle avoidance path and the centerline of the vehicle's lane, and n is the threshold value for responding to changes.

[0184] Taking the initial activation as an example, such as Figure 5 As shown, the current avoidance behavior is not activated. Before the number of consecutive planning reaches n, the avoidance behavior is suppressed, and the avoidance erroneous triggers caused by external input jitter such as perception are filtered to improve decision stability and prevent the vehicle from shaking significantly.

[0185] When the number of consecutive planning iterations reaches n, the result is sent to the trajectory planning module as a stable decision, which generates the coarse trajectory of strategy Jbest into a trajectory that can be executed by the vehicle.

[0186] In this embodiment, the interaction between the vehicle and the target obstacle can be simulated over a period of time in the future. Then, based on the result of the interaction, the optimal strategy at the current moment is output so that the vehicle can safely and comfortably complete the detour and avoidance of the target obstacle.

[0187] It is understandable that if there is a difference between the first lateral offset of the optimal obstacle avoidance path calculated in this instance and the second lateral offset of the currently executed obstacle avoidance path and the centerline of the vehicle's lane, and the number of planning attempts is less than n, then return to step 1 and determine the obstacle avoidance path again according to the above process.

[0188] If the number of planning attempts is less than n, and there is no difference between the first lateral offset of the obstacle avoidance path determined according to the above scheme from the center line of the lane where the vehicle is located and the second lateral offset of the currently executed obstacle avoidance path from the center line of the lane where the vehicle is located, the number of planning attempts will be reset to zero.

[0189] In some examples, the driving scenario of the vehicle is as follows: Figure 6 As shown, assume vehicle ego (the vehicle itself) is traveling in lane M. The target obstacle F is a non-motorized vehicle that is continuously driving on the right lane line within lane M.

[0190] The state of each vehicle is defined by its lateral coordinate (l), longitudinal coordinate (s), and lateral velocity (l / s) in the Frenet coordinate system. ), longitudinal velocity ( Quadruple <l,s, >Construction. In this example, it is assumed that the vehicle's ego state is <0, 50, 0.0, 12.0>, its length is 5m, its width is 2m, and its cruising speed is 12m / s. The vehicle F's state is <-1.5, 70.0, 0.0, 6.0>, its length is 2m, and its width is 1m. Lane M is a dashed lane with a width of 3.8m, and there are passable lanes R and L on both sides. The lateral range on one side is 4m, the shortest forward range is 30m, and the shortest backward distance is 10m. The buffer is 30cm. The future time considered is t=5s.

[0191] The decision region is defined by the horizontal coordinate (l) and vertical coordinate (s) in the Frenet coordinate system. <l,s, Therefore, according to step 1, the decision region P<-10.0,250.0,-5.0, 5.0> for this embodiment can be obtained.

[0192] According to step 2, since vehicle F's current position is within range P, it is initially screened into obstacles. The predicted trajectory sequence taj={} of F is extracted from the vehicle state described above. Since l and t0 are within range, F is identified as a lane-keep type target, selected as a lateral avoidance target, and the obstacle set obj{F} is output.

[0193] Based on step 3, calculate the lateral action range Lat_act_range=[-1.2,1.2], and obtain the lateral policy set lat={-1.2, -0.8, -0.4, 0, 0.4, 0.8, 1.2}.

[0194] Define the longitudinal state of the vehicle, consisting of the longitudinal coordinate (s) and the longitudinal velocity (s). It consists of longitudinal acceleration. Assuming the vehicle continues to move at a constant speed, the longitudinal state sequence of the vehicle is extracted based on the trajectory executed by the vehicle at the previous moment: [50, 12, 0], [62, 12, 0], [74, 12, 0], [86, 12, 0], [98, 12, 0], [110, 12, 0].

[0195] The vehicle's longitudinal state is extended to 7 dimensions based on the number of lateral policies (lat). The lateral and longitudinal policies are combined to obtain a coarse trajectory matrix of candidate obstacle avoidance policies. The policy consists of the longitudinal coordinate (s), velocity (v), and acceleration (...). ), Horizontal coordinate (l) quadruple <s, Composition.

[0196]

[0197] According to step 4, obtain the future driving sequence of obstacle F: <-1.5, 70.0, 0.0, 6.0>, <-1.5, 76.0, 0.0, 6.0>, <-1.5, 82.0, 0.0, 6.0>, <-1.5, 88.0, 0.0, 6.0>, <-1.5, 96.0, 0.0, 6.0>, <-1.5, 104.0, 0.0, 6.0>.

[0198] Based on the formulas in the steps, calculate the risk assessment of the obstacle avoidance strategy corresponding to the obstacle driving state, output the strategy evaluation results, and compare them in a table.

[0199]

[0200] Considering the vehicle's lateral movement capability, the obstacle avoidance strategy with the lowest cost is extracted. The current optimal obstacle avoidance strategy is:

[0201] <20, 12, 0, 0>、<32, 12, 0, 0.4>、<44, 12, 0, 0.8>、<56, 12, 0, 0.8>、<68, 12, 0, 0.8>、<80, 12, 0, 0.8>.

[0202] Understandably, when determining an obstacle avoidance strategy, the vehicle's lateral movement capability is considered first, and the value with the lowest degree of danger is selected. The corresponding operating state is then used as an operating point for the obstacle avoidance strategy.

[0203] According to step 5, the current moment is the initial activation of obstacle avoidance. After three consecutive responses, the vehicle will begin to plan the obstacle avoidance path.

[0204] Figure 7 This is a schematic diagram of the structure of a vehicle obstacle avoidance path planning device provided in an embodiment of this application, as shown below. Figure 7As shown, the vehicle obstacle avoidance path planning device of this embodiment is applied to a self-driving vehicle. The device includes: a first determining module 11, used to determine a target obstacle in the current cruise lane; a prediction module 12, used to predict the running state of the target obstacle at each discrete time point within a predetermined time length based on the running state of the target obstacle at the current moment; a second determining module 13, used to determine the running state of the self-driving vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point based on the current execution path of the self-driving vehicle; a calculation module 14, used to calculate the degree of danger of the target obstacle to each predicted position of the self-driving vehicle at each discrete time point based on the running state of the target obstacle at each discrete time point, the running state of the self-driving vehicle at each predicted position at each discrete time point, and a preset danger level model; and a third determining module 15, used to determine the initial obstacle avoidance route of the self-driving vehicle to the target obstacle within the predetermined time length based on the degree of danger of the target obstacle to each predicted position of the self-driving vehicle.

[0205] The apparatus of this embodiment can be used to perform Figure 1 The technical solutions of the method embodiments shown are similar in principle and in effect, and will not be described again here.

[0206] The device in this embodiment calculates the degree of danger of the target obstacle to each predicted position of the vehicle at each discrete time point based on the running state of the target obstacle at each discrete time point, the running state of the vehicle at each predicted position at each discrete time point, and a preset danger level model. Then, based on the degree of danger of the target obstacle to each predicted position of the vehicle, it determines the initial obstacle avoidance route of the vehicle to the target obstacle within a predetermined time length. In the process of determining the initial obstacle avoidance route of the vehicle to the target obstacle within the predetermined time length, the running state of the target obstacle at each discrete time point and the running state of the vehicle at each predicted position at each discrete time point are used, and combined with the preset danger level model, the initial obstacle avoidance route of the vehicle to the target obstacle within the predetermined time length is determined. In this way, the obstacle avoidance efficiency of the vehicle can be improved.

[0207] As an optional implementation, the first determining module includes: a first determining submodule for determining the decision area of ​​the obstacle avoidance route; a detection submodule for detecting obstacles appearing in the decision area; a second determining submodule for determining the running intention of each obstacle in the current cruise lane; a classification submodule for classifying each obstacle according to different running intentions; and a third determining submodule for determining obstacles corresponding to a specified type of running intention as target obstacles.

[0208] As an optional implementation, the decision region includes a vertical range and a horizontal range; the first determining submodule is specifically used for:

[0209] The longitudinal range of the decision region is determined as follows:

[0210] X beg = -max{v cruise_speed T1, D1}

[0211] X end = max{v cruise_speed T2, D2}

[0212] Among them, X beg X represents the first longitudinal boundary behind the vehicle at the current moment. end v represents the second longitudinal boundary in front of the vehicle at the current moment. cruise_speed T1 is the maximum cruising speed of the vehicle at the current moment, T2 is the preset detection time threshold behind the vehicle, D1 is the preset detection time threshold in front of the vehicle, and D2 is the minimum detection distance behind the vehicle and the minimum detection distance in front of the vehicle.

[0213] The horizontal range of the decision-making region is determined as follows:

[0214] Y beg = -min{D right_max D right_edge}

[0215] Y end = min{D left_max D left_edge}

[0216] Among them, Y beg Y represents the first lateral boundary to the right of the vehicle at the current moment. end D represents the second lateral boundary to the left of the vehicle at the current moment. right_max D represents the maximum horizontal detection distance on the right side. left_max D represents the maximum horizontal detection distance on the left. right_edge The distances between the lane centerline and the right-hand, non-crossable curb. Dleft_edge The distances between the lane centerline and the non-crossable curb on the left are respectively.

[0217] As an optional implementation, the second determining module includes: a fourth determining submodule, used to determine the running status of the vehicle on the current execution path at each discrete time point based on the current execution path of the vehicle; and a fifth determining submodule, used to determine the running status of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point based on the running status of the vehicle on the current execution path at each discrete time point.

[0218] As an optional implementation, the fifth determining submodule is specifically used to: determine the running state of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point based on the longitudinal running state of the vehicle on the current execution path at each discrete time point and two or more lateral positions in the lateral direction of the current cruise lane.

[0219] As an optional implementation, each discrete time point includes a first time point, and each predicted position of the vehicle at the first time point includes the first predicted position.

[0220] Specifically, the calculation module is used to: calculate the degree of danger posed by the target obstacle to the first predicted position of the vehicle at the first time point based on the following danger level model:

[0221]

[0222] in, The degree of threat posed by the target obstacle to the vehicle's first predicted position; This is the first weighting coefficient; This is the second weighting coefficient; The threat coefficient corresponding to the size type of the target obstacle; This is the first predicted position of the vehicle; The position of the target obstacle within the current cruise lane at the first time point; The adjustment value for the position of the target obstacle is based on the acceleration of the target obstacle and the acceleration of the vehicle. for and The difference matrix, where, This is a matrix consisting of the relative distances between the vehicle and the target obstacle at their x-coordinates and the relative distances between the vehicle and the target obstacle at their y-coordinates. To maintain a safe distance; For safe distance ;

[0223] Among them, safe distance Calculate using the following formula:

[0224]

[0225] in, The lateral boundary of the target obstacle; The longitudinal boundary of the target obstacle; The speed of the vehicle. The longitudinal relative velocity between the vehicle and the target obstacle is... The lateral relative velocity between the vehicle and the target obstacle. The preset safety time threshold, The preset comfort acceleration.

[0226] As an optional implementation, the device is further configured to: calculate a first lateral offset of the initial obstacle avoidance route relative to the centerline of the current cruise lane after the third determining module determines the initial obstacle avoidance route of the vehicle to the target obstacle within the predetermined time length; calculate a second lateral offset of the obstacle avoidance path currently executed by the vehicle relative to the centerline of the current cruise lane; calculate the difference between the first lateral offset and the second lateral offset; if the difference is greater than 0, increment the planning count by 1; if the planning count is equal to a preset count, plan the optimal obstacle avoidance route of the vehicle based on the initial obstacle avoidance route.

[0227] The apparatus described in the above embodiments can be used to execute the technical solutions of the above method embodiments. The implementation principle and technical effects are similar, and will not be repeated here.

[0228] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application, as shown below. Figure 8 As shown, the device may include: a housing 81, a processor 82, a memory 83, a circuit board 84, and a power supply circuit 85. The circuit board 84 is disposed inside the space enclosed by the housing 81, and the processor 82 and the memory 83 are disposed on the circuit board 84. The power supply circuit 85 is used to supply power to the various circuits or devices of the above-mentioned electronic device. The memory 83 is used to store executable program code. The processor 82 runs the program corresponding to the executable program code by reading the executable program code stored in the memory 83, and is used to execute any of the vehicle obstacle avoidance path planning methods provided in the foregoing embodiments. Therefore, it can also achieve the corresponding beneficial technical effects, which have been described in detail above and will not be repeated here.

[0229] The aforementioned electronic devices exist in various forms, including but not limited to:

[0230] (1) Mobile communication devices: These devices are characterized by their mobile communication capabilities and are primarily designed to provide voice and data communication. These terminals include smartphones (such as iPhones), multimedia phones, feature phones, and low-end phones.

[0231] (2) Ultra-mobile personal computer devices: These devices fall under the category of personal computers, possessing computing and processing capabilities, and generally also have mobile internet access features. These terminals include PDAs, MIDs, and UMPCs, such as the iPad.

[0232] (3) Server: A device that provides computing services. The components of a server include a processor, hard disk, memory, system bus, etc. Servers are similar to general computer architectures, but because they need to provide highly reliable services, they have higher requirements in terms of processing power, stability, reliability, security, scalability, and manageability.

[0233] (4) Other electronic devices with data interaction functions.

[0234] Accordingly, embodiments of this application also provide a computer-readable storage medium storing one or more programs, which can be executed by one or more processors to implement any of the vehicle obstacle avoidance path planning methods provided in the foregoing embodiments, thus achieving the corresponding technical effects. This has been described in detail above and will not be repeated here.

[0235] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0236] The various embodiments in this specification are described in a related manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0237] In particular, the device embodiment is basically similar to the method embodiment, so the description is relatively simple. For relevant details, please refer to the description of the method embodiment.

[0238] For ease of description, the above apparatus is described by dividing it into various functional units / modules. Of course, in implementing this application, the functions of each unit / module can be implemented in one or more software and / or hardware.

[0239] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0240] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A vehicle obstacle avoidance path planning method, characterized in that, Applied to bicycles, the method includes: Identify target obstacles within the current cruise lane; Based on the current running state of the target obstacle, predict the running state of the target obstacle at each discrete time point within a predetermined time length; Based on the current execution path of the vehicle, determine the running status of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point; Based on the operating status of the target obstacle at each discrete time point, the operating status of the vehicle at each predicted position at each discrete time point, and a preset danger level model, the danger level of the target obstacle to each predicted position of the vehicle at each discrete time point is calculated. Based on the degree of danger posed by the target obstacle to each predicted position of the vehicle, the initial obstacle avoidance route of the vehicle to the target obstacle is determined within the predetermined time length; Wherein, each discrete time point includes a first time point, and each predicted position of the vehicle at the first time point includes a first predicted position; The calculation of the degree of danger posed by the target obstacle to each predicted position of the vehicle at each discrete time point includes: Based on the following hazard level model, the hazard level of the target obstacle to the first predicted position of the vehicle at the first time point is calculated: ； in, The degree of threat posed by the target obstacle to the vehicle's first predicted position; This is the first weighting coefficient; This is the second weighting coefficient; The threat coefficient corresponding to the size type of the target obstacle; This is the first predicted position of the vehicle; The position of the target obstacle within the current cruise lane at the first time point; The adjustment value for the position of the target obstacle is based on the acceleration of the target obstacle and the acceleration of the vehicle. for and The difference matrix, where, This is a matrix consisting of the relative distances between the vehicle and the target obstacle at their x-coordinates and the relative distances between the vehicle and the target obstacle at their y-coordinates. To maintain a safe distance; For safe distance ; Among them, safe distance Calculate using the following formula: ； Among them, D x For lateral safety distance; D y For longitudinal safety distance; The lateral boundary of the target obstacle; The longitudinal boundary of the target obstacle; The speed of the vehicle. The longitudinal relative velocity between the vehicle and the target obstacle is... The lateral relative velocity between the vehicle and the target obstacle. The preset safety time threshold, The preset comfort acceleration.

2. The method according to claim 1, characterized in that, The process of determining target obstacles within the current cruise lane includes: Determine the decision area for obstacle avoidance routes; Detect all obstacles appearing within the decision area; Determine the operational intent of each obstacle within the current cruise lane; The obstacles are classified according to their different operational intentions; The obstacle corresponding to the specified type of running intent is identified as the target obstacle.

3. The method according to claim 2, characterized in that, The decision area includes a longitudinal range and a lateral range; the decision area for determining the obstacle avoidance route includes: The longitudinal range of the decision region is determined as follows: X beg =-max{v cruise_speed * T1, D1} X end = max{v cruise_speed * T2, D2} Among them, X beg X represents the first longitudinal boundary behind the vehicle at the current moment. end v represents the second longitudinal boundary in front of the vehicle at the current moment. cruise_speed T1 is the maximum cruising speed of the vehicle at the current moment, T2 is the preset detection time threshold behind the vehicle, D1 is the preset detection time threshold in front of the vehicle, and D2 is the minimum detection distance behind the vehicle and the minimum detection distance in front of the vehicle. The horizontal range of the decision-making region is determined as follows: Y beg =- min{D right_max ,D right_edge } Y end = min{D left_max ,D left_edge } Among them, Y beg Y represents the first lateral boundary to the right of the vehicle at the current moment. end D represents the second lateral boundary to the left of the vehicle at the current moment. right_max D represents the maximum horizontal detection distance on the right side. left_max D represents the maximum horizontal detection distance on the left. right_edge The distances between the lane centerline and the right-hand, non-crossable curb. Dleft_edge The distances between the lane centerline and the non-crossable curb on the left are respectively.

4. The method according to claim 2, characterized in that, The step of determining the operating status of the autonomous vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point, based on the vehicle's current execution path, includes: Based on the current execution path of the vehicle, determine the running status of the vehicle on the current execution path at each discrete time point; Based on the vehicle's operating status on the current execution path at each discrete time point, the operating status of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point is determined.

5. The method according to claim 4, characterized in that, The step of determining the operating status of the autonomous vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point based on the vehicle's operating status on the current execution path at each discrete time point includes: Based on the longitudinal running state of the vehicle on the current execution path at each discrete time point, and two or more lateral positions in the lateral direction of the current cruise lane, the running state of the vehicle at two or more predicted positions in the lateral direction of the current cruise lane at each discrete time point is determined.

6. The method according to claim 1, characterized in that, After determining the initial obstacle avoidance route of the vehicle towards the target obstacle within the predetermined time length, the method further includes: Calculate the first lateral offset of the initial obstacle avoidance path relative to the centerline of the current cruise lane; Calculate the second lateral offset of the obstacle avoidance path currently executed by the vehicle relative to the centerline of the current cruise lane; Calculate the difference between the first lateral offset and the second lateral offset; If the difference is greater than 0, the planning iteration count is incremented by 1; If the number of planning attempts equals the preset number of attempts, then based on the initial obstacle avoidance route, the optimal obstacle avoidance path for the vehicle is planned.

7. A vehicle obstacle avoidance path planning device, characterized in that, Applied to a bicycle, the device includes: The first determination module is used to determine target obstacles within the current cruise lane; The prediction module is used to predict the running state of the target obstacle at each discrete time point within a predetermined time length, based on the running state of the target obstacle at the current time. The second determining module is used to determine the running status of the vehicle at two or more predicted positions along the lateral direction of the current cruise lane at each discrete time point, based on the current execution path of the vehicle. The calculation module is used to calculate the degree of danger of the target obstacle to each predicted position of the vehicle at each discrete time point based on the running status of the target obstacle at each discrete time point, the running status of the vehicle at each predicted position at each discrete time point, and a preset danger level model. The third determining module is used to determine the initial obstacle avoidance route of the vehicle to the target obstacle within the predetermined time length based on the degree of danger of the target obstacle to each predicted position of the vehicle. Wherein, each discrete time point includes a first time point, and each predicted position of the vehicle at the first time point includes a first predicted position; The computing module is specifically used for: Based on the following hazard level model, the hazard level of the target obstacle to the first predicted position of the vehicle at the first time point is calculated: ； in, The degree of threat posed by the target obstacle to the vehicle's first predicted position; This is the first weighting coefficient; This is the second weighting coefficient; The threat coefficient corresponding to the size type of the target obstacle; This is the first predicted position of the vehicle; The position of the target obstacle within the current cruise lane at the first time point; The adjustment value for the position of the target obstacle is based on the acceleration of the target obstacle and the acceleration of the vehicle. for and The difference matrix, where, This is a matrix consisting of the relative distances between the vehicle and the target obstacle at their x-coordinates and the relative distances between the vehicle and the target obstacle at their y-coordinates. To maintain a safe distance; For safe distance ; Among them, safe distance Calculate using the following formula: ； Among them, D x For lateral safety distance; D y For longitudinal safety distance; The lateral boundary of the target obstacle; The longitudinal boundary of the target obstacle; The speed of the vehicle. The longitudinal relative velocity between the vehicle and the target obstacle is... The lateral relative velocity between the vehicle and the target obstacle. The preset safety time threshold, The preset comfort acceleration.

8. An electronic device, characterized in that, The electronic device includes: a housing, a processor, a memory, a circuit board, and a power supply circuit, wherein the circuit board is disposed inside the space enclosed by the housing, and the processor and the memory are disposed on the circuit board; the power supply circuit is used to supply power to various circuits or devices of the electronic device; the memory is used to store executable program code; the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, for executing the vehicle obstacle avoidance path planning method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, which can be executed by one or more processors to implement the vehicle obstacle avoidance path planning method according to any one of claims 1-6.

Images