Obstacle risk field environment modeling method and device, and related products

Abstract

The present invention discloses a method for modeling an obstacle risk field environment, a device thereof, and related products. The method comprises: establishing a current-frame ST grid map corresponding to a dynamic obstacle in a Frenet coordinate system with the current position of the vehicle as the origin; determining the current-frame occupied area of ​​the dynamic obstacle in the current-frame ST grid map and the predicted trajectory probability of each grid based on the predicted risk area of ​​each dynamic obstacle and the current-frame occupancy probability of each grid, and the intended risk area and the current-frame occupancy probability of each grid based on the current-frame occupancy probability of each dynamic obstacle in the current-frame ST grid map; and calculating the current-frame occupancy probability of each occupied grid in the current-frame ST grid map based on the occupied area of ​​each dynamic obstacle in the current-frame ST grid map and the predicted trajectory probability of each grid to obtain the current-frame ST risk field. The method of the present invention facilitates "early prediction" of the acceleration and deceleration of dynamic obstacles, improves the response speed of the autonomous driving system, and rationalizes the influence of other high-speed traffic participants on the driving behavior of the vehicle in interactive scenarios.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving environment modeling, and in particular to a method for modeling an obstacle risk field environment, a device for modeling an obstacle risk field environment, a storage medium, a computer program product containing instructions, an electronic device, and a mobile tool. Background Art

[0002] The autonomous driving system is a comprehensive system that integrates environmental perception, decision-making control, and action execution. It is a system that fully considers the coordinated planning of vehicles and traffic environment, and is also an important part of the future intelligent transportation system.

[0003] In autonomous driving systems, establishing a risk field for dynamic obstacles in the environment is crucial. The stability and accuracy of the risk field model directly affect the behavior of the autonomous driving system when dealing with dynamic obstacles. For example, in an unlit intersection, in order to reasonably give way to other traffic participants, it is necessary to establish a good risk field.

[0004] The current obstacle risk field environment modeling method is mainly aimed at unstructured road conditions. It directly performs local path planning by processing the perceived surrounding environment obstacles in the global coordinate system or the vehicle coordinate system based on grid maps or artificial potential fields.

[0005] On structured roads, the vehicle and obstacles are constrained by traffic rules and signs, making it impossible to directly apply existing grid maps or artificial potential field methods to model the obstacle risk field environment. The inventors have discovered that existing obstacle risk field modeling methods can be applied to structured road obstacle risk field modeling. However, when correlating obstacle risk fields at different times, they fail to account for changes in the obstacle's own speed or intention. In other words, these methods assume that the obstacle's speed will not fluctuate significantly in the short term. While these methods are effective for modeling the risk field for low-speed obstacles such as pedestrians and non-motorized vehicles, they may exhibit a certain degree of hysteresis when modeling the risk field for high-speed obstacles such as vehicles, failing to "predict" the movement changes of other obstacles in advance.

[0006] Therefore, the above obstacle risk field environment modeling method needs to be further improved. Summary of the Invention

[0007] The embodiments of the present invention aim to solve at least one of the above technical problems.

[0008] In a first aspect, an embodiment of the present invention provides a method for modeling an obstacle risk field environment, including:

[0009] The current frame ST grid map corresponding to the dynamic obstacle is established in the Frenet coordinate system with the current position of the vehicle as the origin;

[0010] According to the predicted trajectory and reference acceleration of the dynamic obstacle, the predicted risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid are determined;

[0011] Determine the target acceleration based on the baseline acceleration of the dynamic obstacle, and determine the intended risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid based on the predicted trajectory of the dynamic obstacle and the target acceleration;

[0012] Determine the current frame occupied area of ​​the dynamic obstacle in the ST grid map and the current frame predicted trajectory probability of each grid based on the predicted risk area of ​​the dynamic obstacle and the current frame occupied probability of each grid, the intended risk area and the current frame occupied probability of each grid;

[0013] According to the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid, the current frame occupancy probability of each occupied grid in the current frame ST grid map is calculated to obtain the current frame ST risk field.

[0014] In a second aspect, an embodiment of the present invention provides an obstacle risk field environment modeling device, comprising:

[0015] The mapping module is used to build the current frame ST grid map corresponding to the dynamic obstacles in the Frenet coordinate system with the current position of the vehicle as the origin;

[0016] A predicted risk area determination module is used to determine the predicted risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid;

[0017] An intention risk area determination module is used to determine a target acceleration based on the baseline acceleration of the dynamic obstacle, and to determine the intention risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid based on the predicted trajectory and target acceleration of the dynamic obstacle;

[0018] The current frame predicted trajectory probability determination module is used to determine the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid based on the predicted risk area of ​​the dynamic obstacle and the current frame occupancy probability of each grid, the intended risk area and the current frame occupancy probability of each grid;

[0019] The obstacle risk field construction module is used to calculate the current frame occupancy probability of each occupied grid in the current frame ST grid map based on the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid to obtain the current frame ST risk field.

[0020] In a third aspect, an embodiment of the present invention provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the method provided in the first aspect.

[0021] In a fourth aspect, an embodiment of the present invention further provides a computer program product, which includes a computer program stored on a storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer performs the steps of the method provided according to the first aspect of the present invention.

[0022] In a fifth aspect, an embodiment of the present invention provides an electronic device comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the method provided according to the first aspect of the present invention.

[0023] In a sixth aspect, an embodiment of the present invention provides a mobile tool, characterized in that it includes the electronic device shown in the fifth aspect.

[0024] The embodiments of the present invention introduce predicted risk areas and intended risk areas for dynamic obstacles. That is, for obstacles with high speed and regular movement, such as motor vehicles, the impact of the obstacle's own speed or intention changes is taken into account, reducing the impact of front-end input errors caused by predicted trajectory errors or delays, etc., which helps to "predict high-speed obstacles in advance", improve the response speed of the autonomous driving system, and rationalize the impact of other high-speed traffic participants on the vehicle's driving behavior in interactive scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following is a brief introduction to the drawings used in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0026] Figure 1 This is a flow chart of a method for modeling an obstacle risk field environment provided by an embodiment of the present invention;

[0027] Figure 2 yes Figure 1 Schematic diagram of the sub-process of step S12;

[0028] Figure 3 1 is a flow chart of a method for determining a target acceleration based on a baseline acceleration of a dynamic obstacle provided by an embodiment of the present invention;

[0029] Figure 4A Schematic diagram of a probability density model of acceleration provided by an embodiment of the present invention;

[0030] Figure 4B Schematic diagram of the predicted risk area of ​​a dynamic obstacle in the ST grid map of the current frame provided by an embodiment of the present invention;

[0031] Figure 4C This is a schematic diagram of the intended risk area of ​​a dynamic obstacle in the ST grid map of the current frame provided by an embodiment of the present invention;

[0032] Figure 5 yes Figure 1 Schematic diagram of the sub-process of step S13;

[0033] Figure 6A Schematic diagram of the predicted risk area of ​​a dynamic obstacle in the ST grid map of the current frame and the occupancy probability of each grid in the current frame provided by an embodiment of the present invention;

[0034] Figure 6B Schematic diagram of the intended risk area of ​​a dynamic obstacle in the ST grid map of the current frame and the occupancy probability of each grid in the current frame provided by an embodiment of the present invention;

[0035] Figure 6C yes Figure 6A The predicted risk area and the current frame occupancy probability of each grid are shown as well as Figure 6B Schematic diagram of the intention risk area and the current frame occupancy probability of each grid superimposed.

[0036] Figure 7 1 is a flow chart of a method for calculating and updating, for each dynamic obstacle, the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map, provided by an embodiment of the present invention;

[0037] Figure 8 1 is a flow chart of a method for calculating and updating the current frame occupancy probability of each occupied grid in the current frame ST grid map after the updated dynamic obstacle occupies the current frame area in the current frame ST grid map in an embodiment of the present invention;

[0038] Figure 9 This is a principle block diagram of an obstacle risk field environment modeling device provided by one embodiment of the present invention;

[0039] Figure 10 This is a principle block diagram of an obstacle risk field environment modeling device provided by another embodiment of the present invention;

[0040] Figure 11 This is a principle block diagram of an obstacle risk field environment modeling device provided by another embodiment of the present invention;

[0041] Figure 12 yes Figure 11 Principle block diagram of the target acceleration determination unit 301;

[0042] Figure 13 yes Figure 11 Principle block diagram of the central risk area determination unit 302;

[0043] Figure 14 This is a principle block diagram of an obstacle risk field environment modeling device provided by another embodiment of the present invention;

[0044] Figure 15 yes Figure 14 Schematic diagram of the current frame occupied area updating unit 501;

[0045] Figure 16 yes Figure 14 Schematic diagram of the current frame occupancy probability updating unit 502;

[0046] Figure 17 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0048] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0049] The present invention may be described in the general context of computer-executable instructions, such as program modules, executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media, including storage devices.

[0050] In the present invention, "module", "device", "system" and the like refer to related entities applied to a computer, such as hardware, a combination of hardware and software, software or software in execution, etc. Specifically, for example, an element can be, but is not limited to, a process running on a processor, a processor, an object, an executable element, an execution thread, a program and / or a computer. In addition, an application or script program running on a server, or a server can all be an element. One or more elements can be in an execution process and / or thread, and an element can be localized on a computer and / or distributed between two or more computers, and can be run by various computer-readable media. An element can also communicate through local and / or remote processes based on a signal having one or more data packets, for example, a signal from a data packet interacting with another element in a local system, a distributed system, and / or a signal from a network on the Internet that interacts with other systems via signals.

[0051] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include" and "comprise" include not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article or device. In the absence of further limitations, the elements defined by the phrase "include..." do not exclude the presence of other identical elements in the process, method, article or device that includes the elements.

[0052] The obstacle risk field modeling method in the embodiment of the present invention can be applied to electronic devices, such as but not limited to smart phones, smart tablets, personal PCs, computers, cloud servers, controllers, etc.

[0053] The technical solution of this invention primarily implements environmental modeling of the obstacle risk field on structured roads, primarily using a grid representation based on the Frenet coordinate system (also known as the road coordinate system). This involves using a grid to represent the spatial position of obstacles within the grid map, i.e., their position within the structured road. The obstacle risk field defined herein can be represented as the relationship between the spatial position of each grid point within the grid map and its corresponding occupancy probability.

[0054] Figure 1 The obstacle risk field environment modeling method provided by the embodiment of the present invention is schematically shown. The method can be applied to any mobile tool that can realize autonomous driving, such as autonomous driving vehicles (passenger cars, buses, minibuses, trucks, off-road vehicles, sanitation vehicles, sweepers, floor washing vehicles and vacuum trucks, etc.), sweeping robots, etc., and the present invention does not limit this. Figure 1As shown, the method includes:

[0055] S11: Create the current frame ST grid map corresponding to the dynamic obstacle in the Frenet coordinate system with the current position of the vehicle as the origin;

[0056] In this embodiment, the origin of the Frenet coordinate system is set to the current position of the vehicle, that is, the projection point of the vehicle on the center line of the structured road.

[0057] In the Frenet coordinate system, the centerline of the structured road is used as the reference line (i.e., the reference line for the ego vehicle). Dynamic obstacles are described using the S and T parameters in the Frenet coordinate system, where T is the obstacle time relative to the current time, and S is the longitudinal distance projected relative to the reference line at time T. This creates a dynamic obstacle ST grid map. This method combines perceived obstacles with high-precision maps.

[0058] In one example, lowercase s and t are specified to represent a grid in the ST grid map, the leading superscript represents the grid number, and the leading subscript represents the ID of the dynamic obstacle. For example, the zeroth occupied grid point in the occupied area of ​​a dynamic obstacle with an ID of 5 can be represented as ( 0 5s, 0 5t). The maximum value of s is represented by max_s, and the minimum value of s is represented by min_s; the maximum value of t is represented by max_t, and the minimum value of t is represented by min_t. Because the min_t and max_t of the same dynamic obstacle have different s ranges (min_s, max_s) in the same time coordinate system, they are distinguished by superscripts, for example, {(min_s, max_s), min_t} and {(min_s', max_s'), max_t}.

[0059] S12: Determine the predicted risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid in the current frame according to the predicted trajectory and reference acceleration of the dynamic obstacle;

[0060] Assuming that a dynamic obstacle will accelerate uniformly at a baseline acceleration within a short period of time, the predicted risk region for the dynamic obstacle in the ST grid map of the current frame can be determined, typically represented as RST. Since the RST calculation result for a single dynamic obstacle is the occupied area in the ST grid map, which contains multiple grid points after rasterization, uppercase S and T are used to represent the collection of all grid points for a single dynamic obstacle. The leading subscript represents the obstacle ID. For example, the RST region for an obstacle ID of 5 can be represented as (5RS, 5RT).

[0061] Exemplarily, the predicted trajectory of the dynamic obstacle is output by a prediction module. The prediction module is an upstream module of the known environment modeling of the present invention. The prediction module generally provides two prediction models, namely a uniform speed prediction model and a variable speed prediction model.

[0062] In the present invention, the data related to the predicted trajectory usually includes the starting speed v0 of the starting point p0 of the predicted trajectory line, which represents the collision starting point p in the process of collision between the dynamic obstacle and the vehicle. min The corresponding collision starting point velocity v pmin and the time min_t of the corresponding collision start point, the collision end point p during the collision between the dynamic obstacle and the vehicle max The corresponding collision end point velocity v pmax And the corresponding data such as max_t of the collision end point.

[0063] In some embodiments, the starting velocity v0 of the starting point p0 of the predicted trajectory can be output by the prediction module. In an alternative embodiment, the starting velocity v0 can be output by the perception module. Like the prediction module, the perception module is also an upstream module of the environmental modeling of the present invention.

[0064] When the prediction module provides a variable speed prediction model, the baseline acceleration can be directly output by the prediction module, with each dynamic obstacle having a corresponding baseline acceleration. If the prediction module does not provide this baseline acceleration, it can be estimated based on data related to the predicted trajectory (e.g., the speed of each point on the predicted trajectory, or the position of each point and the corresponding time).

[0065] In some embodiments, the collision start time t (i.e., min_t) and s (i.e., min_s) of the ego vehicle and the dynamic obstacle, as well as the collision end time t (i.e., max_t) and s (i.e., max_s) are determined based on the dynamic obstacle's current frame predicted trajectory, the dynamic obstacle's baseline acceleration, the dynamic obstacle's size, the ego vehicle's reference line, and the ego vehicle's size. The ST region where the dynamic obstacle and the ego vehicle collide is determined based on (min_t, min_s) and (max_t, max_s). This method can be used to determine the collision start time min_t and the collision end time max_t.

[0066] For example, according to the starting speed v0 of the predicted trajectory of the dynamic obstacle, the time min_t of the collision start point between the dynamic obstacle and the vehicle, and the speed v pmin , the time max_t of the collision end point and the speed v of the collision end point pmax , the baseline acceleration a of the dynamic obstacle can be estimated p , which can be expressed as:

[0067]

[0068] When the reference acceleration is output by the predicted trajectory data, the collision start point velocity v can also be estimated accordingly based on the reference acceleration and the time min_t of the collision start point and the time max_t of the collision end point. pmin and the velocity v at the end point of collision pmax .

[0069] S13: Determine the target acceleration based on the baseline acceleration of the dynamic obstacle, and determine the intended risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid thereof based on the predicted trajectory of the dynamic obstacle and the target acceleration;

[0070] When considering the influence of the dynamic obstacle's own speed or intention change, it is necessary to determine the risk area of ​​the dynamic obstacle in the current frame ST grid map based on the predicted trajectory of the dynamic obstacle and the acceleration of each target, that is, the intention risk area, usually expressed as PST, and calculate the current occupancy probability of each grid in the PST.

[0071] Similar to the predicted risk area RST, the calculation result of a single dynamic obstacle PST is also the occupied area in the ST grid map. After rasterization, it contains multiple grids. For example, the PST area of ​​the obstacle ID 5 can be expressed as (5PS, 5PT).

[0072] S14: Determine the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid based on the predicted risk area of ​​the dynamic obstacle and the current frame occupied probability of each grid, the intended risk area and the current frame occupied probability of each grid;

[0073] For example, the predicted risk area of ​​the dynamic obstacle and the current frame occupancy probability of each grid thereof can be directly added to the intended risk area and the current frame occupancy probability of each grid thereof to obtain the current frame occupancy area of ​​the dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid thereof.

[0074] It should be noted that each dynamic obstacle needs to execute the above steps S12 to S14 to obtain the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid.

[0075] S15: According to the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid, the current frame occupancy probability of each occupied grid in the current frame ST grid map is calculated to obtain the current frame ST risk field.

[0076] In this embodiment, the risk field of a dynamic obstacle is represented as an ST risk field, which mainly describes the relationship between each occupied grid point (s, t) of a dynamic obstacle and the occupancy probability P, denoted as P(s, t).

[0077] The embodiments of the present invention introduce predicted risk areas and intended risk areas for dynamic obstacles. That is, for obstacles with high speed and regular movement, such as motor vehicles, the impact of changes in the speed or intention of the dynamic obstacles themselves is taken into account, reducing the impact of front-end input errors caused by predicted trajectory errors or delays, etc., which helps to "predict high-speed obstacles in advance", improve the response speed of the autonomous driving system, and rationalize the impact of other high-speed traffic participants on the driving behavior of the vehicle in interactive scenarios.

[0078] Figure 2 The method for determining the predicted risk area of ​​a dynamic obstacle in the current frame ST grid map is schematically shown, and the method includes:

[0079] S121: Determining a ST region where the dynamic obstacle and the ego vehicle collide based on the predicted trajectory of the dynamic obstacle, the reference acceleration, and the reference line of the ego vehicle;

[0080] S122: Discretize the ST region grid into the current frame ST grid map to obtain the predicted risk region of the dynamic obstacle in the current frame ST grid map.

[0081] In this embodiment, the ST region where a dynamic obstacle collides with the ego vehicle can be determined using methods known in the art, which are not limited in the present invention. For example, as described above, the time t (i.e., min_t) and s (i.e., min_s) of the collision start point, as well as the time t (i.e., max_t) and s (i.e., max_s) of the collision end point, are determined based on the dynamic obstacle's current frame predicted trajectory, the dynamic obstacle's baseline acceleration, the dynamic obstacle's size, the ego vehicle's reference line, and the ego vehicle's size. The ST region where the dynamic obstacle collides with the ego vehicle is determined based on (min_t, min_s) and (max_t, max_s). This ST region is not rasterized. For example, a first virtual frame is generated based on the size of the dynamic obstacle to represent the dynamic obstacle, and a second virtual frame is generated based on the size of the ego vehicle to represent the ego vehicle. Each trajectory point in the predicted trajectory of the dynamic obstacle is used as the center point of the first virtual frame, and each waypoint on the ego vehicle reference line is used as the center point of the second virtual frame. The first virtual frame is simulated to move along the predicted trajectory, and the second virtual frame is simulated to move along the reference route. The time points when the two start and end the collision and their corresponding s coordinate values ​​are recorded.

[0082] For example, the ST region where the dynamic obstacle collides with the ego-vehicle can be represented as {(min_s, max_s), min_t} and {(min_s', max_s'), max_t}. The grids occupied by obstacles are marked as Id[]. Since the predicted obstacle trajectories may overlap, the ID of each grid occupied by obstacles is theoretically not unique, so it is recorded as Id[].

[0083] It should also be noted that the output predicted trajectory varies depending on the prediction model provided by the prediction module. However, the calculation of the occupied area in the ST grid diagram of the obstacle prediction risk area RST (i.e., the ST area where the dynamic obstacle and the ego vehicle collide) in the variable speed prediction model is essentially the same as the calculation of the occupied area in the ST grid diagram in the uniform speed prediction model. The difference is that the prediction module outputs different predicted trajectory lengths under different models, and the predicted point positions are different at the same prediction time t. Therefore, the calculation results (min_s, max_s) and (min_s', max_s') are the same, but min_t and max_t are different.

[0084] like Figure 4B As shown, it shows the predicted risk area of ​​the dynamic obstacle in the ST grid map of the current frame, where the gray grid is the RST area occupation grid.

[0085] In some embodiments, as Figure 3 As shown in FIG, the target acceleration is determined according to the baseline acceleration of the dynamic obstacle, which can be specifically implemented as follows:

[0086] S131A: Establish an acceleration probability density model with the reference acceleration as the mean and the preset variance value as the variance;

[0087] Among them, the reference acceleration can be expressed as a p , the preset variance value can be expressed as o', and the acceleration probability density model can be expressed as f(a). The acceleration probability density model f(a) can be established using a probability density distribution, such as a Gaussian distribution. It is understood that other probability density distributions can be used to establish the acceleration probability density model f(a), and the present invention is not limited thereto.

[0088] Exemplarily, the probability density model f(a) of the acceleration may be determined by Gaussian distribution, for example, as follows:

[0089]

[0090] In some embodiments, the preset variance value o' is determined based on changes in perceptible acceleration, for example, based on an acceleration range that is comfortable for the driver, while also being limited by factors such as lane speed limits and the current speed of the vehicle.

[0091] S131B: setting an acceleration discrete interval in the acceleration probability density model according to a preset interval value;

[0092] For example, the preset variance value o' can be used as the preset interval value to set the discrete interval of acceleration [a p -mo',a p +mo'], where m is a constant, for example, m=2.

[0093] Reference Figure 4A , which shows a schematic diagram of the probability density model of acceleration.

[0094] S131C: Select multiple discrete points from the acceleration discrete interval as target accelerations.

[0095] For example, according to the predicted trajectory starting speed v0, the time min_t of the collision starting point and the collision starting point speed v pmin , it can be estimated from the starting point p0 to the collision starting point p min The minimum distance D min , and according to the collision start point speed v0, the collision end point time max_t and the collision end point speed v pmax , it can be estimated that the distance from the starting point p0 to the collision end point p max The maximum distance D max , where the minimum distance D min and the maximum distance D max Can be expressed as:

[0096]

[0097]

[0098] Furthermore, based on the starting speed v0 of the predicted trajectory starting point of the obstacle, the minimum distance D from the predicted trajectory starting point to the above-mentioned collision starting point and collision ending point is min and the maximum distance D max , the determined discrete interval corresponding to [a p -2o',a p +2o'] (interval is o') the acceleration of each target a i The minimum time function min_ti to reach the collision starting point and the target acceleration a corresponding to each discrete interval i The maximum time function max_ti to reach the end point of the collision can be expressed as:

[0099]

[0100]

[0101] In some implementations, the intended risk area of ​​each target acceleration in the ST grid map of the current frame may be determined based on the minimum time function and the maximum time function.

[0102] Specifically, first, based on the minimum time function and the maximum time function, the T interval occupied by each target acceleration in the current frame ST grid map is determined; then, based on the determined T interval and the corresponding S interval occupied by the dynamic obstacle in the current frame ST grid map under each target acceleration, the occupied areas of each target acceleration in the current frame ST grid map within the above discrete intervals, namely, each intended risk area PST, are determined.

[0103] Figure 4C The figure shows a schematic diagram of the dynamic obstacle intention risk area, where the gray grid is the PST area occupation grid. Figure 4C The ST grid diagram shown shows the target acceleration a i The T interval occupied by the ST grid diagram and the S interval occupied by the dynamic obstacle under each target acceleration in the ST grid diagram.

[0104] Therefore, the target acceleration a i The corresponding PST regions are {(min_s,max_s),min_ti} and {(min_s',max_s'),max_ti}.

[0105] In some embodiments, as Figure 5 As shown, according to the predicted trajectory and target acceleration of the dynamic obstacle, the current frame occupancy probability of each grid in the intention risk area in the ST grid map of the current frame is determined, which can be specifically implemented as follows:

[0106] S132A: Determine the normalized acceleration probability density of the target acceleration according to the acceleration probability density model;

[0107] S132B: Determine the current frame occupancy probability of each grid in the intended risk area according to the current frame occupancy probability of each grid in the predicted risk area of ​​the dynamic obstacle and the normalized acceleration probability density of the target acceleration.

[0108] For example, assume that the discrete interval [a p -mO′,a p +mO′] in which m=2, that is, the discrete interval is [a p -2o′, a p +2o′], when the interval is o′, the probability P(ai) of the target acceleration ai can be obtained by integrating the acceleration probability density model f(a) in step S131A:

[0109]

[0110] Since the probability sum after the restricted interval and discretization will be less than 1, the probability P(ai) corresponding to each target acceleration ai is normalized to obtain the normalized acceleration probability density P1(ai) as the acceleration discrete probability of each target acceleration:

[0111]

[0112] For example, the current frame occupancy probability P(ai, (S, T)) of each grid in the intention risk area can be obtained by the following formula:

[0113] P(a i , (S, T)) = P (S, T) × P1 (a i ) (9)

[0114] Among them, P(S, T) represents the current frame occupancy probability P(S, T) of each grid in the predicted risk area of ​​the ST grid map of the current frame. Since the ST areas corresponding to different target accelerations may overlap, the grid point of obstacle k ( x k s, x k The probability P(t) x k s, x k t) is:

[0115]

[0116] Figure 6A The predicted risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid are shown.

[0117] Figure 6B The probability of a dynamic obstacle occupying each area under each target acceleration in the current frame ST grid map is shown. For simplicity, the figure shows only the probability that one of the target accelerations is in the corresponding occupied area.

[0118] like Figure 6C As shown, it shows Figure 6A The predicted risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid are shown in Figure 6B Schematic diagram of the current frame occupied area of ​​a dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid determined by superimposing the intention risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid.

[0119] In the above embodiment, only the ST risk field of the dynamic obstacle in the current frame, ie, a single frame, is considered, and historical frame data is not considered.

[0120] The technical solution of the embodiment of the present invention, from the perspective of a single frame, allows a certain error within the sensor accuracy range, and transforms the original "danger / no danger" result expression into a "0-1" probabilistic risk description, thereby improving the applicability of the data.

[0121] In some implementations, it is also necessary to superimpose the ST risk field of the obstacle in the current frame with the ST risk field accumulated in the historical frames to calculate and update the ST risk field of the obstacle in the current frame.

[0122] For example, Figure 7 As shown, for each dynamic obstacle, the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map is calculated and updated, which can be specifically implemented as follows:

[0123] S151A: Determine the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map as a first estimated occupied area;

[0124] S151B: Project the occupied area of ​​the dynamic obstacle in the previous frame onto the ST grid map of the current frame to obtain a second estimated occupied area.

[0125] Exemplarily, the area occupied by the dynamic obstacle in the previous frame is displaced according to the time change and the vehicle position change between the previous frame and the current frame to obtain a second estimated area occupied by the dynamic obstacle in the ST grid map of the current frame.

[0126] For example, the coordinates of the second estimated occupied area obtained by projecting the occupied area of ​​the previous frame onto the ST grid map of the current frame can be expressed as:

[0127] s m =s m m-1 -△s; (11)

[0128] t m =t m m-1 -△t; (12)

[0129] Here, s m Indicates the vertical coordinate of the current frame corresponding to the moment, s m m-1 is the calculation result of the previous frame, that is, the vertical coordinate of the previous frame at the corresponding time. △s is the movement distance of the vehicle along the path between the current frame and the previous frame, which can be obtained by calculating the change in the vehicle positioning. △t is the time difference between the two frames, for example, 0.1 second.

[0130] In some embodiments, the time interval (ie, update period) between the current frame and the previous frame can be set as needed, for example, based on the hardware frequency of the sensor, etc. The present invention is not limited thereto.

[0131] S151C: Determine the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map based on the first estimated occupied area and the second estimated occupied area.

[0132] For example, the first estimated occupied area and the second estimated occupied area may be filtered to obtain the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid image. Filtering methods such as Bayesian filtering and Kalman filtering are not limited in the present invention.

[0133] Alternatively, the first estimated occupied area and the second estimated occupied area may be directly superimposed together, that is, the union of the first estimated occupied area and the second estimated occupied area may be obtained.

[0134] Further, if Figure 8 As shown, after the updated dynamic obstacle obtained in the above method occupies the area in the current frame in the current frame ST grid map, the current frame occupancy probability of each occupied grid in the current frame ST grid map is calculated, which can be specifically implemented as follows:

[0135] S152A: Determine at least one target dynamic obstacle corresponding to each occupied grid based on the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid; determine the current frame estimated occupancy probability of each occupied grid based on the current frame predicted trajectory probability of each target dynamic obstacle occupying the occupied grid and a preset maximum occupancy probability;

[0136] Because each dynamic obstacle has a corresponding predicted trajectory, there may be intersections or overlaps between the predicted trajectories, resulting in some occupied grids in the current frame ST grid map being occupied by multiple dynamic obstacles at the same time. Therefore, when calculating the current frame estimated occupancy probability of these occupied grids, it is necessary to consider the current frame predicted trajectory probabilities of the corresponding multiple dynamic obstacles in the occupied grid. Therefore, it is first necessary to determine which target dynamic obstacles occupy each occupied grid. Exemplarily, for example, it can be determined whether the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map contains the occupied grid. If so, it is determined that the dynamic obstacle is the target dynamic obstacle occupying the occupied grid.

[0137] For example, the dynamic obstacle ID is represented by i, the occupied grid is represented by j, and the current frame predicted trajectory probability of a single target dynamically occupying the occupied grid j is expressed as i P m = i k mThe estimated occupancy probability of the current frame of the occupancy grid j can be expressed as:

[0138] P m =min(max_p,∑ i k m ) (13)

[0139] in i k m is the predicted trajectory probability of the target dynamic obstacle with value i in Id[] in the occupancy grid j, and max_p is the preset maximum occupancy probability.

[0140] S152B: Calculate the current frame occupancy probability of the occupied grid according to the current frame estimated occupancy probability of the occupied grid and the previous frame occupancy probability of the occupied grid.

[0141] Exemplarily, the current frame occupancy probability of the occupied grid may be calculated based on a Bayesian filter formula. The current frame occupancy probability of the occupied grid is calculated using the following Bayesian filter formula.

[0142]

[0143]

[0144]

[0145] Among them, m represents the current moment of the current coordinate system where the current frame is located, L is the intermediate variable, and according to the physical meaning represented by the obstacle st, when t m m-1 When <0, clear the obstacle P(s,t) risk field.

[0146] This allows us to calculate and update the risk field P(s,t) for each occupied grid point. A high probability value for P(s,t) indicates high stability in the temporal dimension and high repeatability in the spatial dimension. This value can be used to define threshold intervals and determine the risk level of the obstacle's intended trajectory and its corresponding predicted trajectory.

[0147] Alternatively, the above-mentioned Bayesian filter can also be replaced by other filtering formulas, such as Kalman filtering.

[0148] The method of the embodiment of the present invention realizes the simultaneous superposition of multiple frames of continuous results, improves the stability of the data, avoids the influence of single frame jumps, and further improves the accuracy of obstacle risk assessment.

[0149] Figure 9 The present invention provides an obstacle risk field environment modeling device according to an embodiment of the present invention. Figure 9 As shown: The device includes:

[0150] Mapping module 100: used to create a current frame ST grid map corresponding to dynamic obstacles in the Frenet coordinate system with the current position of the vehicle as the origin;

[0151] Predicted risk area determination module 200: used to determine the predicted risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid;

[0152] The intention risk area determination module 300 is used to determine the target acceleration based on the baseline acceleration of the dynamic obstacle, and determine the intention risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid based on the predicted trajectory of the dynamic obstacle and the target acceleration;

[0153] Current frame predicted trajectory probability determination module 400: used to determine the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid based on the predicted risk area of ​​the dynamic obstacle and the current frame occupancy probability of each grid, the intended risk area and the current frame occupancy probability of each grid;

[0154] Obstacle risk field construction module 500: used to calculate the current frame occupancy probability of each occupied grid in the current frame ST grid map based on the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid to obtain the current frame ST risk field.

[0155] Figure 10 This is an obstacle risk field environment modeling device provided by another embodiment of the present invention. Figure 10 As shown: In the device, the predicted risk area determination module 200 specifically includes:

[0156] The ST region determination unit 201 is used to determine the ST region where the dynamic obstacle collides with the ego vehicle based on the predicted trajectory of the dynamic obstacle, the baseline acceleration, and the reference line of the ego vehicle. The predicted risk region determination unit 202 is used to discretize the ST region grid into the current frame ST grid map to obtain the predicted risk region of the dynamic obstacle in the current frame ST grid map.

[0157] Figure 11 This is an obstacle risk field environment modeling device provided according to another embodiment of the present invention. Figure 11 As shown: In the device, the intention risk area determination module 300 specifically includes:

[0158] a target acceleration determining unit 301, configured to determine a target acceleration according to a reference acceleration of a dynamic obstacle; and

[0159] The intention risk area determination unit 302 is configured to determine the current frame occupancy probability of each grid in the intention risk area in the current frame ST grid map according to the predicted trajectory and target acceleration of the dynamic obstacle.

[0160] Further, such as Figure 12 As shown, the target acceleration determination unit 301 further includes:

[0161] The acceleration probability density model construction subunit 3011 is used to establish an acceleration probability density model with a reference acceleration as the mean and a preset variance value as the variance;

[0162] The acceleration discrete interval setting subunit 3012 is used to set the acceleration discrete interval in the acceleration probability density model according to a preset interval value; and

[0163] The target acceleration determination subunit 3013 is configured to select a plurality of discrete points from the acceleration discrete interval as the target acceleration.

[0164] Further, such as Figure 13 As shown, the intention risk area determination unit 302 further includes:

[0165] The acceleration probability density normalization processing subunit 3021 is configured to determine the normalized acceleration probability density of the target acceleration according to the acceleration probability density model; and

[0166] The occupancy probability determination subunit 3022 is used to determine the current frame occupancy probability of each grid in the intended risk area according to the current frame occupancy probability of each grid in the predicted risk area of ​​the dynamic obstacle and the normalized acceleration probability density of the target acceleration.

[0167] Figure 14 This is an obstacle risk field environment modeling device provided according to another embodiment of the present invention. Figure 14 As shown: In the device, the obstacle risk field construction module 500 specifically includes:

[0168] The current frame occupied area updating unit 501 is used to calculate and update the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map; and

[0169] The current frame occupancy probability updating unit 502 is configured to calculate the current frame occupancy probability of each occupied grid in the current frame ST grid map.

[0170] In some embodiments, as Figure 15 As shown, the current frame occupied area updating unit 501 further includes:

[0171] The first estimated occupied area determining subunit 5011 is configured to determine the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map as the first estimated occupied area;

[0172] The second estimated occupied area determining subunit 5012 is configured to project the occupied area of ​​the dynamic obstacle in the previous frame into the ST grid map of the current frame to obtain a second estimated occupied area; and

[0173] The current frame occupied area updating subunit 5013 determines the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map according to the first estimated occupied area and the second estimated occupied area.

[0174] In some embodiments, as Figure 16 As shown, the current frame occupancy probability updating unit 502 further includes:

[0175] The current frame estimated occupancy probability determination subunit 5021 is configured to determine at least one target dynamic obstacle corresponding to each occupied grid based on the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid; determine the current frame estimated occupancy probability of each occupied grid based on the current frame predicted trajectory probability of each target dynamic obstacle occupying the occupied grid and the preset maximum occupancy probability; and

[0176] The current frame occupancy probability updating subunit 5022 is configured to calculate the current frame occupancy probability of the occupied grid according to the current frame estimated occupancy probability of the occupied grid and the previous frame occupancy probability of the occupied grid.

[0177] It should be noted that the specific implementation processes and principles of the various modules and units of the obstacle risk field environment modeling device in the embodiments of the present invention can be found in the corresponding descriptions of the corresponding method embodiments above, and therefore will not be repeated here. For example, the obstacle risk field environment modeling device in the embodiments of the present invention can be implemented in any electronic device with a processor, including but not limited to smartphones, smart tablets, personal computers, computers, cloud servers, controllers, etc.

[0178] An embodiment of the present invention further provides a non-volatile computer storage medium storing computer executable instructions, which can execute the obstacle risk field environment modeling method in any of the above embodiments;

[0179] As an embodiment, the non-volatile computer storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:

[0180] The current frame ST grid map corresponding to the dynamic obstacle is established in the Frenet coordinate system with the current position of the vehicle as the origin;

[0181] According to the predicted trajectory and reference acceleration of the dynamic obstacle, the predicted risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid are determined;

[0182] Determine the target acceleration based on the baseline acceleration of the dynamic obstacle, and determine the intended risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid based on the predicted trajectory of the dynamic obstacle and the target acceleration;

[0183] Determine the current frame occupied area of ​​the dynamic obstacle in the ST grid map and the current frame predicted trajectory probability of each grid based on the predicted risk area of ​​the dynamic obstacle and the current frame occupied probability of each grid, the intended risk area and the current frame occupied probability of each grid;

[0184] According to the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid, the current frame occupancy probability of each occupied grid in the current frame ST grid map is calculated to obtain the current frame ST risk field.

[0185] A non-volatile computer-readable storage medium can be used to store non-volatile software programs, non-volatile computer executable programs, and modules, such as the program instructions / modules corresponding to the methods described in the embodiments of the present invention. One or more program instructions stored in the non-volatile computer-readable storage medium, when executed by a processor, perform the obstacle risk field environment modeling method described in any of the aforementioned method embodiments.

[0186] The non-volatile computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the device, etc. In addition, the non-volatile computer-readable storage medium may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state memory device. In some embodiments, the non-volatile computer-readable storage medium may optionally include a memory remotely located relative to the processor, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0187] In some embodiments, an embodiment of the present invention further provides a computer program product, which includes a computer program stored on a non-volatile computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes any one of the above-mentioned obstacle risk field environment modeling methods.

[0188] In some embodiments, an embodiment of the present invention also provides an electronic device, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the obstacle risk field environment modeling method of any embodiment of the present invention.

[0189] Figure 17 FIG. 1 is a schematic diagram of the hardware structure of an electronic device for executing an obstacle risk field environment modeling method provided by another embodiment of the present invention, such as Figure 17 As shown, the device includes:

[0190] One or more processors 1710 and memory 1720, Figure 17 A processor 1710 is taken as an example.

[0191] The device for executing the obstacle risk field environment modeling method may further include: an input device 1730 and an output device 1740 .

[0192] The processor 1710, the memory 1720, the input device 1730 and the output device 1740 may be connected via a bus or other means. Figure 17 The bus connection is taken as an example.

[0193] Memory 1720, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the obstacle risk field environment modeling method in the embodiments of the present invention. Processor 1710 executes the non-volatile software programs, instructions, and modules stored in memory 1720 to execute various server functional applications and data processing, thereby implementing the obstacle risk field environment modeling method of the aforementioned method embodiment.

[0194] Memory 1720 may include a program storage area and a data storage area. The program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the obstacle risk field environment modeling device, etc. In addition, memory 1720 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, memory 1720 may optionally include memory remotely located relative to processor 1710. These remote memories may be connected to the obstacle risk field environment modeling device via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0195] The input device 1730 can receive input digital or character information and generate signals related to user settings and function control of the obstacle risk field environment modeling device. The output device 740 can include a display device such as a display screen.

[0196] The one or more modules are stored in the memory 1720 and, when executed by the one or more processors 1710 , perform the obstacle risk field environment modeling method in any of the above method embodiments.

[0197] The above-mentioned product can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0198] The electronic devices according to the embodiments of the present invention may be implemented in various forms, including but not limited to:

[0199] (1) Mobile communication devices: These devices are characterized by their mobile communication capabilities and their primary purpose is to provide voice and data communications. These terminals include smartphones, multimedia phones, feature phones, and low-end phones.

[0200] (2) Ultra-mobile personal computer devices: These devices fall under the category of personal computers and have computing and processing capabilities, and generally also have mobile Internet access. These terminals include PDAs, MIDs, and UMPC devices, such as tablet computers.

[0201] (3) Portable entertainment devices: These devices can display and play multimedia content. They include audio and video players, handheld game consoles, e-books, smart toys, and portable car navigation devices.

[0202] (4) Server: A device that provides computing services. The server consists of a processor, hard disk, memory, system bus, etc. The server is similar to a general computer architecture, but because it needs to provide highly reliable services, it has higher requirements in terms of processing power, stability, reliability, security, scalability, and manageability.

[0203] (5) Other onboard electronic devices with data interaction functions, such as on-board devices installed in vehicles.

[0204] An embodiment of the present invention further provides a mobile tool including the electronic device described above. The mobile tool includes a vehicle capable of automatic driving (e.g., a passenger car, a sweeper, a sanitation vehicle, a bus, a minibus, a truck, a vacuum cleaner, a floor scrubber), a sweeping robot, etc.

[0205] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0206] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0207] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for modeling an obstacle risk field environment, comprising: The current frame ST grid map corresponding to the dynamic obstacle is established in the Frenet coordinate system with the current position of the vehicle as the origin; According to the predicted trajectory and reference acceleration of the dynamic obstacle, the predicted risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid are determined; Determine the target acceleration based on the baseline acceleration of the dynamic obstacle, and determine the intended risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid based on the predicted trajectory of the dynamic obstacle and the target acceleration; Determine the current frame occupied area of ​​the dynamic obstacle in the ST grid map and the current frame predicted trajectory probability of each grid based on the predicted risk area of ​​the dynamic obstacle and the current frame occupied probability of each grid, the intended risk area and the current frame occupied probability of each grid; According to the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid, the current frame occupancy probability of each occupied grid in the current frame ST grid map is calculated. The target acceleration is determined based on the baseline acceleration of the dynamic obstacle, specifically including: Establish an acceleration probability density model with the reference acceleration as the mean and the preset variance value as the variance; Setting an acceleration discrete interval in the acceleration probability density model according to a preset interval value; A plurality of discrete points are selected from the acceleration discrete interval as target accelerations.

2. The method according to claim 1, characterized in that Determine the current frame occupancy probability of each grid in the intention risk area of ​​the ST grid map of the current frame by the dynamic obstacle, specifically including: Determining a normalized acceleration probability density of the target acceleration according to the acceleration probability density model; The current frame occupancy probability of each grid in the intended risk area is determined according to the current frame occupancy probability of each grid in the predicted risk area of ​​the dynamic obstacle and the normalized acceleration probability density of the target acceleration.

3. The method according to claim 1, wherein determining the predicted risk area of ​​the dynamic obstacle in the ST grid image of the current frame based on the predicted trajectory of the dynamic obstacle and the reference acceleration comprises: Determining a ST region where the dynamic obstacle and the ego vehicle collide based on the predicted trajectory of the dynamic obstacle, the baseline acceleration, and the reference line of the ego vehicle; The ST region grid is discretized into the current frame ST grid map to obtain a predicted risk region of the dynamic obstacle in the current frame ST grid map.

4. The method according to claim 1, wherein Calculate the current frame occupancy probability of each occupied grid in the current frame ST grid map, specifically including: Determine at least one target dynamic obstacle corresponding to each occupied grid based on the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map; determine the current frame estimated occupancy probability of each occupied grid based on the current frame predicted trajectory probability of each target dynamic obstacle occupying the occupied grid and a preset maximum occupancy probability; The current frame occupancy probability of the occupied grid is calculated according to the current frame estimated occupancy probability of the occupied grid and the previous frame occupancy probability of the occupied grid.

5. An obstacle risk field environment modeling device, characterized in that: include: The mapping module is used to build the current frame ST grid map corresponding to the dynamic obstacles in the Frenet coordinate system with the current position of the vehicle as the origin; A predicted risk area determination module is used to determine the predicted risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid; An intention risk area determination module is used to determine a target acceleration based on the baseline acceleration of the dynamic obstacle, and to determine the intention risk area of ​​the dynamic obstacle in the current frame ST grid map and the current frame occupancy probability of each grid based on the predicted trajectory and target acceleration of the dynamic obstacle; The current frame predicted trajectory probability determination module is used to determine the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid based on the predicted risk area of ​​the dynamic obstacle and the current frame occupancy probability of each grid, the intended risk area and the current frame occupancy probability of each grid; The obstacle risk field construction module is used to calculate the current frame occupancy probability of each occupied grid in the current frame ST grid map based on the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map and the current frame predicted trajectory probability of each grid. The intention risk area determination module also includes: Establish an acceleration probability density model with the reference acceleration as the mean and the preset variance value as the variance; Setting an acceleration discrete interval in the acceleration probability density model according to a preset interval value; A plurality of discrete points are selected from the acceleration discrete interval as target accelerations.

6. A storage medium storing a computer program, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

7. A computer program product comprising instructions, characterized in that When the computer program product is run on a computer, the computer is enabled to perform the steps of the method according to any one of claims 1 to 4.

8. An electronic device comprising: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the method according to any one of claims 1 to 4.

9. A mobile tool, characterized in that: The electronic device comprising the electronic device according to claim 8.

Images