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

Abstract

The present invention discloses an obstacle risk field environment modeling method and its device, and related products, the method comprising: establishing a current frame obstacle grid map in a Frenet coordinate system with the current position of the vehicle as the origin; determining the current frame occupied area of ​​the obstacle in the current frame obstacle grid map according to the current frame perception data of the obstacle; calculating and updating the current frame occupation probability of each grid in the current frame occupied area of ​​each obstacle in the current frame obstacle grid map, so as to obtain the current frame obstacle risk field. The environment modeling method of the present invention improves the stability and effectiveness of the obstacle risk field environment model.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving, 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 the autonomous driving system, the decision-making control module is a crucial component. Its upstream is the basic information output such as perception and positioning. Since the description of the environment by the perception sensor mainly relies on the vehicle coordinate system, and the description of the vehicle information by positioning mainly relies on the earth coordinate system, when the decision-making planning module uses the perception positioning information, it is necessary to perform environmental modeling and integrate the static map with the dynamic environmental information. The construction of the obstacle risk field is a key link in environmental modeling. The authenticity and stability of the risk field determine the functional effect of the horizontal and vertical obstacle interaction of the autonomous driving system.

[0004] At present, the 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] Under structured road conditions, the vehicle and obstacles are constrained by traffic rules and traffic signs, so it is impossible to directly apply existing grid maps or artificial potential field methods to model the obstacle risk field environment. Summary of the invention

[0006] The embodiments of the present invention aim to solve at least one of the above technical problems.

[0007] In a first aspect, an embodiment of the present invention provides a method for modeling an obstacle risk field environment, comprising:

[0008] The obstacle grid map of the current frame is established in the Frenet coordinate system with the current position of the vehicle as the origin;

[0009] Determine the current frame occupied area of ​​the obstacle in the current frame obstacle grid map according to the current frame perception data of the obstacle;

[0010] The current frame occupancy probability of each grid in the current frame occupancy area of ​​each obstacle in the current frame obstacle grid map is calculated and updated to obtain the current frame obstacle risk field.

[0011] In a second aspect, an embodiment of the present invention provides an obstacle risk field environment modeling device, comprising:

[0012] The mapping module is used to build the obstacle grid map of the current frame in the Frenet coordinate system with the current position of the vehicle as the origin;

[0013] An occupied area determination module, used to determine the current frame occupied area of ​​the obstacle in the current frame obstacle grid map according to the current frame perception data of the obstacle;

[0014] The obstacle risk field construction module is used to calculate and update the current frame occupancy probability of each grid in the current frame occupancy area of ​​each obstacle in the current frame obstacle grid map to obtain the current frame obstacle risk field.

[0015] 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.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the steps of the method provided in the first aspect.

[0017] 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 so that the at least one processor can perform the steps of the method provided in the first aspect of the present invention.

[0018] 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.

[0019] The embodiment of the present invention establishes a current frame obstacle grid map based on the frenet coordinate system. After determining the occupied area of ​​the obstacle in the current frame obstacle grid map, the occupation probability of each grid in the occupied area is calculated. That is, the obstacle risk field established by the technical solution of the present application is rasterized and refined to each grid with a corresponding occupancy probability, so that the obstacle risk field is more refined and accurate, and the effectiveness of the obstacle risk field is enhanced. At the same time, the time dimension and spatial dimension of the obstacle are effectively combined to improve the stability and effectiveness of the obstacle risk field environmental model, so that the SL and ST grid maps can be applied to the modeling of the obstacle risk field, providing accurate environmental model support for more reasonable decision-making and planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces 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 creative work.

[0021] Figure 1 is a schematic diagram of an obstacle risk field environment modeling method provided by an embodiment of the present invention;

[0022] FIG. 2A to FIG. 2C It is a schematic diagram of updating the SL coordinate system provided by an embodiment of the present invention;

[0023] FIG. 3A to FIG. 3C The embodiment of the present invention provides a schematic diagram of updating the ST coordinate system;

[0024] Figure 4 is a schematic diagram of a method for determining a current occupied area of ​​a static obstacle in an obstacle grid map of a current frame in an embodiment of the present invention;

[0025] Figure 5 is a schematic diagram of a method for determining a current occupied area of ​​a dynamic obstacle in an obstacle grid map of a current frame in an embodiment of the present invention;

[0026] Figure 6 is a schematic diagram of a method for determining a current frame ST risk field for a dynamic obstacle in an embodiment of the present invention;

[0027] Figure 7 yes Figure 6 A schematic diagram of a method for implementing step S132;

[0028] Figure 8 It is a principle block diagram of an obstacle risk field environment modeling device provided by an embodiment of the present invention;

[0029] Fig. 9 yes Figure 8 Schematic diagram of the modeling module 100;

[0030] Fig.10 is a principle block diagram of an obstacle risk field environment modeling device provided by another embodiment of the present invention;

[0031] Fig.11 yes Fig.10 Schematic diagram of the dynamic obstacle occupied area determination unit 220;

[0032] Fig.12 It is a principle block diagram of an obstacle risk field environment modeling device provided by another embodiment of the present invention;

[0033] Fig.13 yes Fig.12 The principle block diagram of the dynamic obstacle risk field construction unit 320;

[0034] Fig.14 It is a principle block diagram of an obstacle risk field environment modeling device provided by another embodiment of the present invention;

[0035] Fig.15 It is a schematic structural diagram of an embodiment of an electronic device of the present invention. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application may be combined with each other.

[0038] The present invention may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. 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 through a communications network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0039] In the present invention, "module", "device", "system" and the like refer to related entities applied to computers, such as hardware, a combination of hardware and software, software or software in execution, etc. In detail, for example, an element can be, but is not limited to, a process, a processor, an object, an executable element, an execution thread, a program and / or a computer running on a processor. In addition, an application or script program running on a server, 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 operated by various computer-readable media. An element can also communicate through local and / or remote processes according to a signal with one or more data packets, for example, a signal from a data that interacts with another element in a local system, a distributed system, and / or a network on the Internet through a signal to interact with other systems.

[0040] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include" and "comprise" include not only those elements, but also other elements that are not explicitly listed, or also include elements inherent to such processes, methods, articles or equipment. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the existence of other identical elements in the process, method, article or equipment that includes the elements.

[0041] 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.

[0042] The technical solution of the present invention mainly realizes the environmental modeling of the obstacle risk field under the structured road, which is mainly represented by the grid method based on the Frenet coordinate system (also called the road coordinate system). That is, the spatial position of the obstacle in the grid map, that is, the position in the structured road, is represented by the grid. The obstacle risk field defined in this article can be represented by the spatial position of each grid point in the grid map and the relationship between its corresponding occupancy probability.

[0043] Figure 1 is a schematic diagram of an obstacle risk field environment modeling method according to an embodiment of the present invention. 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 cleaners, etc.), sweeping robots, etc., and the present invention does not limit this. Figure 1 As shown, the method includes:

[0044] S11, establishing a current frame obstacle grid map in the Frenet coordinate system with the current position of the vehicle as the origin;

[0045] 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.

[0046] In some embodiments, obstacles may include static obstacles and dynamic obstacles. The type of obstacles is included in the perception data. The perception data is output by a known upstream perception module. The obstacle grid map of the current frame is established in the Frenet coordinate system with the current position of the vehicle as the origin, including:

[0047] Establish the current frame SL grid map corresponding to the static obstacle in the frenet coordinate system with the current position of the vehicle as the origin; and

[0048] 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.

[0049] Specifically, the center line of the structured road is used as the reference line (i.e., the reference line of the vehicle) in the Frenet coordinate system, and the S and L parameters are used to describe the static obstacles to establish the SL grid map, where S is the longitudinal distance of the static obstacle relative to the reference line after projection, and L is the lateral distance of the static obstacle relative to the reference line after projection, as shown in Figure 2A As shown in the SL coordinate system, 1 represents the vehicle, 2 represents the projection point of the static obstacle on the reference line (the longitudinal distance of the projection point from the vehicle is s), 3 represents the static obstacle, and 4 represents the reference line. For dynamic obstacles, an ST grid diagram can be established, where T is the obstacle time based on the current time, and S is the longitudinal distance after projection relative to the reference line at time T, as shown in Figure 3A In the ST coordinate system, 1 represents the vehicle, 2 represents the dynamic obstacle, 3 represents the collision point between the dynamic obstacle and the vehicle, 4 represents the reference line, and 5 represents the predicted trajectory of the dynamic obstacle. This method can combine the perceived obstacles with the high-precision map.

[0050] S12, determining the current frame occupied area of ​​the obstacle in the current frame obstacle grid map according to the current frame perception data of the obstacle;

[0051] Among them, the perception data is output by the known upstream perception module. The perception module needs to obtain a large amount of environmental information through various sensors, including the state of the vehicle, traffic flow information, road conditions, traffic signs, etc. These sensors mainly include: Lidar, Camera, Millimeter Wave Radar, etc.

[0052] Taking static obstacles as an example, Figure 2B It shows the area occupied by the static obstacle in the previous frame of the obstacle grid map in the previous frame determined by the perception data of the previous frame; Figure 2C , which shows the area occupied by the obstacle in the current frame in the obstacle grid map of the current frame determined by the current frame perception data of the static obstacle.

[0053] S13. Calculate and update the current frame occupancy probability of each grid in the current frame occupancy area of ​​each obstacle in the current frame obstacle grid map to obtain the current frame obstacle risk field.

[0054] Exemplarily, the risk fields for static obstacles and dynamic obstacles are represented as SL risk field and ST risk field respectively. The SL risk field mainly describes the relationship between each occupied grid point (s, l) of a static obstacle and the occupancy probability P, denoted as P(s, l). The ST risk field 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). The static obstacle risk field (i.e., SL risk field) can be represented as {s, l, P(s, l)}, and the dynamic obstacle risk field (i.e., ST risk field) can be represented as {s, t, P(s, t)}.

[0055] The embodiment of the present invention establishes a current frame obstacle grid map based on the Frenet coordinate system. After determining the occupied area of ​​the obstacle in the current frame obstacle grid map, the occupancy probability of each grid in the occupied area is calculated. That is, the obstacle risk field established by the technical solution of the present application is rasterized and refined to each grid with a corresponding occupancy probability, making the obstacle risk field more refined and accurate, and enhancing the effectiveness of the obstacle risk field.

[0056] Before describing the present invention in greater detail, it may be helpful to understand the present invention by providing definitions of certain terms used herein.

[0057] In the grid map, the occupied area of ​​a static obstacle in the SL grid map contains multiple grid points. The uppercase SL is defined to represent the collection of grid points contained in the occupied area of ​​a single static obstacle. The leading subscript represents the ID of the static obstacle. For example, the occupied area of ​​a static obstacle with an ID of 5 in the SL grid map can be represented as (5S,5L); the lowercase sl is defined to represent a grid point in SL. The leading superscript represents the grid point number. For example, the 0th occupied grid point in the occupied area of ​​a static obstacle with an ID of 5 can be represented as ( 0 5s, 0 5l). The occupied area of ​​a dynamic obstacle in the ST grid map contains multiple grid points. The uppercase ST is defined to represent the collection of grid points contained in the occupied area of ​​a single dynamic obstacle, and the leading subscript represents the ID of the dynamic obstacle. For example, the occupied area of ​​a dynamic obstacle with an ID of 5 in the ST grid map can be represented as (5S,5T); the lowercase st is defined to represent a grid point in ST, and the leading superscript represents the grid point number. For example, the 0th occupied grid point in the occupied area of ​​a dynamic obstacle with an ID of 5 can be represented as ( 0 5s, 0 5t).

[0058] Since SL and ST are both described based on the frenet coordinate system, the origin is the projection point of the ego vehicle on the reference center line of the structured road. Therefore, the historical risk field ordinate data needs to be continuously updated according to the ego vehicle movement. The subscript m represents the time of establishing the current frame frenet coordinate system, and the superscript n represents the time of establishing the current frame coordinate system of the data. The data refers to the s and t corresponding coordinate values ​​of the dynamic obstacles calculated based on the current frame perception data and the center line of the ego vehicle lane. When the superscript is omitted, it means that the time of the two coordinate systems is unified, for example, s2 1 It represents the projection of the t1 data in the Frenet coordinate system established at time t2, and s2 represents the projection of the t2 data in the Frenet coordinate system established at time t2.

[0059] According to the above definition, the area occupied by the static obstacle with ID 5 in the current frame can be expressed as (5S m,5 L m ), where the 0th grid point can be expressed as ( 0 5s m , 0 5l m ).

[0060] For static obstacles, the maximum value of s is expressed as max_s, the minimum value of s is expressed as min_s, the maximum value of l is expressed as max_l, and the minimum value of l is expressed as min_l.

[0061] For dynamic obstacles, the maximum value of t is expressed as max_t, and the minimum value of t is expressed as min_t. For the same dynamic obstacle, the min_t and max_t have different s ranges (min_s, max_s) in the same time coordinate system, so they are distinguished by superscripts "'", such as {(min_s, max_s), min_t} and {(min_s', max_s'), max_t}.

[0062] Figure 4 and Figure 5 The following schematically illustrates the implementation methods of determining the current occupied area of ​​the obstacle in the obstacle grid map of the current frame when the obstacle is a static obstacle and a dynamic obstacle.

[0063] like Figure 4 As shown, the method for determining the current occupied area of ​​the static obstacle in the obstacle grid map of the current frame includes:

[0064] S1211: Determine the coverage area of ​​the static obstacle according to the position point and obstacle size in the current frame perception data of the static obstacle;

[0065] S1212: Project the coverage area onto the current frame SL grid map to obtain the current frame occupied area of ​​the static obstacle in the current frame SL grid.

[0066] The area occupied by a static obstacle in the current frame SL grid map can be expressed as (min_s, max_s) and (min_l, max_l). The grid occupied by the obstacle ID is marked in the SL grid map. Since the actual positions of obstacles do not overlap, the ID of each grid occupied by the obstacle is theoretically unique.

[0067] It should be noted that for static obstacles, their changes in structured roads are not significant, so historical frame information can be ignored. Figures 2B to 2C As shown, Figure 2B is the area occupied by the static obstacle in the previous frame of the SL grid map. Figure 2C The area occupied by the static obstacle in the current frame SL grid map.

[0068] like Figure 5 As shown, the method for determining the current occupied area of ​​the dynamic obstacle in the obstacle grid map of the current frame includes:

[0069] S1221: Determine the ST region where the dynamic obstacle and the ego vehicle collide according to the current frame predicted trajectory of the dynamic obstacle and the ego vehicle reference line;

[0070] According to the current frame predicted trajectory of the dynamic obstacle and the reference line of the vehicle, the ST area where the dynamic obstacle and the vehicle collide can be expressed as {(min_s, max_s), min_t} and {(min_s', max_s'), max_t}. The grid occupied by the obstacle Id[] is marked in the ST grid map. Since the predicted trajectories of the obstacles may overlap, the ID of each grid occupied by the obstacle is not unique in theory, so it is recorded as Id[].

[0071] The predicted trajectory of the current frame of the dynamic obstacle is output by a known upstream prediction module. Like the perception module, the prediction module is also an upstream module of the environment modeling of the present invention, and the present invention does not limit this.

[0072] It should be noted that the ST region where the dynamic obstacle collides with the ego vehicle can be determined according to methods known in the prior art, and the present invention does not limit this. Exemplarily, the time t (i.e., min_t) and s (i.e., min_s) of the collision start point during the collision between the ego vehicle and the dynamic obstacle, as well as the time t (i.e., max_t) and s (i.e., max_s) of the collision end point are determined according to the current frame predicted trajectory of the dynamic obstacle, the size of the dynamic obstacle, the ego vehicle reference line, and the size of the ego vehicle. The ST region where the dynamic obstacle collides with the ego vehicle is determined according to (min_t, min_s) and (max_t, max_s), and the ST region is not rasterized. For example, a first virtual frame for representing a dynamic obstacle is generated according to the size of the dynamic obstacle, and a second virtual frame for representing the ego vehicle is generated according to the size of 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 collision between the two begins and ends and their corresponding s coordinate values ​​are recorded.

[0073] S1222: discretizing the ST region grid into the current frame ST grid map to obtain a first estimated occupied area;

[0074] like Figure 3C As shown, the first estimated occupied area obtained by discretizing the ST area grid into the ST grid map of the current frame is shown.

[0075] S1223: Project the occupied area of ​​the previous frame of the dynamic obstacle into the ST grid map of the current frame to obtain a second estimated occupied area.

[0076] Exemplarily, the area occupied by the dynamic obstacle in the previous frame is displaced according to the time variation and the vehicle position variation between the previous frame and the current frame to obtain the second estimated area of ​​the dynamic obstacle in the ST grid map of the current frame.

[0077] 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:

[0078] s m =s m m-1 -△s; (1)

[0079] t m =t m m-1 -△t; (2)

[0080] Here, s m Indicates the ordinate of the current frame corresponding to the time, sm m-1 is the calculation result of the previous frame, that is, the ordinate of the previous frame at the corresponding time. △s is the moving distance of the vehicle along the path direction between the current frame and the previous frame, which can be obtained by calculating the change in the positioning of the vehicle. △t is the time difference between the two frames, for example 0.1 seconds.

[0081] In some implementations, the time interval (ie, update period) between the current frame and the previous frame may be set as required, for example, based on the hardware frequency of the sensor, etc. The present invention is not limited thereto.

[0082] S1224: Determine 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.

[0083] Exemplarily, 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 map. The filtering method may include, for example, Bayesian filtering, Kalman filtering, etc. The present invention is not limited to this. Figure 3B and 3C As shown, Figure 3B is the area occupied by the dynamic obstacle in the previous frame of the ST grid map. Figure 3C is the area occupied by the dynamic obstacle in the current frame ST grid map. Figure 3B The area occupied by the previous frame in is projected onto Figure 3C In the current frame ST grid map, the projected area is compared with Figure 3C For simplicity, the figure does not show the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map obtained by filtering.

[0084] 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 taken.

[0085] In some implementations, a method for implementing a current frame SL risk field for a static obstacle includes:

[0086] For each grid in the current frame occupancy area of ​​each static obstacle in the current frame SL grid map, the obstacle position credibility in the current frame perception data of the grid is used as the current frame occupancy probability of the grid.

[0087] Therefore, for static obstacles, the occupancy probability of each grid in the current frame can be determined directly based on the perception data.

[0088] In some embodiments, a method for determining the ST risk field of a dynamic obstacle in the current frame is as follows: Figure 6 As shown, including:

[0089] S131, calculating and updating the current frame predicted trajectory probability of each grid in the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map;

[0090] S132. Determine the current frame occupancy probability of each occupied grid in the current frame ST grid map according to the current frame predicted trajectory probability of each grid in the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map to obtain the current frame ST risk field.

[0091] For example, further reference is made to Figure 7 , step S132 can be specifically implemented as follows:

[0092] For each occupied grid in the current frame ST grid map, perform the following steps:

[0093] S1321, determining at least one target dynamic obstacle corresponding to each occupied grid according to the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid;

[0094] Because each dynamic obstacle has a corresponding predicted trajectory, there may be intersections or overlaps between the predicted trajectories, resulting in some occupied grids being occupied by multiple dynamic obstacles at the same time in the current frame ST grid map. 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 each dynamic obstacle contains the occupied grid in the current frame occupied area in the current frame ST grid map. If so, it is determined that the dynamic obstacle is a target dynamic obstacle occupying the occupied grid.

[0095] S1322, determining the current frame estimated occupancy probability of the occupancy grid according to the current frame predicted trajectory probability of each target dynamic obstacle occupying the occupancy grid and the preset maximum occupancy probability;

[0096] 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 dynamic obstacle occupying the occupied grid j is expressed as i P m = i k m , the estimated occupancy probability of the current frame of the occupied grid j can be expressed as:

[0097] P m =min(max-p,∑,k m ) (3)

[0098] 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.

[0099] S1323. 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.

[0100] Exemplarily, the current frame occupancy probability of the occupancy grid may be calculated based on a Bayesian filtering formula.

[0101]

[0102]

[0103]

[0104] Among them, L is the intermediate variable. According to the physical meaning of the obstacle st, when t m m-1 <0, clear the obstacle P(S,T) risk field.

[0105] In this way, the risk field P(s,t) of each occupied grid point can be calculated and updated. A high probability value of P(s,t) indicates that the prediction result has high stability in the time dimension and high repeatability in the spatial dimension. This can be used to define the threshold interval and determine the risk level of the obstacle intention and the corresponding predicted trajectory.

[0106] Exemplarily, the Bayesian filter may also be replaced by other filtering algorithms, such as the Kalman filter.

[0107] The obstacle risk field calculated and updated by the method of the embodiment of the present invention includes two parts: SL risk field and ST risk field. It not only provides accurate location information - the position of the grid points, but also provides probability values ​​P(s,l) and P(s,t) for each grid point, thereby realizing the establishment of an obstacle risk field environment model. The autonomous driving system can make decisions and plans based on the obstacle risk fields {s,l,P(s,l)} and {s,t,P(s,t)}.

[0108] In addition, from the perspective of a single frame, a certain error within the sensor accuracy range is allowed, and the original "dangerous or not" result expression is changed to a "0-1" probabilistic risk description, which improves the applicability of the data. At the same time, the superposition of multiple frames of continuous results improves the stability of the data, avoids the impact of single frame jumps, and improves the accuracy of obstacle risk assessment.

[0109] The technical solution of the present invention, when determining the occupied area of ​​a dynamic obstacle in the ST grid map, comprehensively considers the ST area where the dynamic obstacle in the current frame collides with the vehicle and the occupied area of ​​the dynamic obstacle in the previous frame of the ST grid map, to determine the current frame occupied area of ​​the dynamic obstacle in the current frame ST grid; and, when calculating the current frame occupancy probability of each occupied grid in the current frame ST grid, comprehensively considers the predicted trajectory probability of multiple target dynamic obstacles occupying the occupied grid in the current frame of the occupied grid and the previous frame occupancy probability of the occupied grid, to determine the current frame occupancy probability of the occupied grid. That is, the technical solution of the present invention can effectively combine the time dimension and the space dimension of the dynamic obstacle at the same time, improve the stability and accuracy of the dynamic obstacle risk field, and provide accurate environmental model support for more reasonable decision-making and planning.

[0110] Figure 8 The present invention is an obstacle risk field environment modeling device provided according to an embodiment of the present invention. Figure 8 As shown: The device includes:

[0111] A mapping module 100 is used to build a current frame obstacle grid map in the Frenet coordinate system with the current position of the vehicle as the origin;

[0112] An occupied area determination module 200, configured to determine a current frame occupied area of ​​the obstacle in the current frame obstacle grid map according to the current frame perception data of the obstacle;

[0113] The obstacle risk field construction module 300 is used to calculate and update the current frame occupancy probability of each grid in the current frame occupancy area of ​​each obstacle in the current frame obstacle grid map to obtain the current frame obstacle risk field.

[0114] In some embodiments, obstacles may include static obstacles and dynamic obstacles. The type of obstacle is included in the perception data, which is output by a known upstream perception module. Fig. 9 , the mapping module 100 further includes:

[0115] SL grid map establishing unit 101: used to establish a current frame SL grid map corresponding to the static obstacle in the frenet coordinate system with the current position of the vehicle as the origin; and

[0116] The ST grid map establishing unit 102 is used to establish 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.

[0117] The embodiment of the present invention establishes a current frame obstacle grid map based on the Frenet coordinate system. After determining the occupied area of ​​the obstacle in the current frame obstacle grid map, the occupancy probability of each grid in the occupied area is calculated. That is, the obstacle risk field established by the technical solution of the present application is rasterized and refined to each grid with a corresponding occupancy probability, making the obstacle risk field more refined and accurate, and enhancing the effectiveness of the obstacle risk field.

[0118] Fig.10 The present invention is an obstacle risk field environment modeling device provided according to another embodiment of the present invention. Fig.10 As shown: In the device, the occupied area determination module 200 specifically includes:

[0119] The static obstacle occupied area determination unit 210 is used to determine the current frame occupied area of ​​the static obstacle in the current frame obstacle grid map according to the current frame perception data of the static obstacle, specifically including:

[0120] Determine the coverage area of ​​the static obstacle according to the location point and obstacle size in the current frame perception data of the static obstacle;

[0121] The coverage area is projected onto the current frame SL grid map to obtain the current frame occupied area of ​​the static obstacle in the current frame SL grid.

[0122] The dynamic obstacle occupied area determination unit 220 is used to determine the current frame occupied area of ​​the dynamic obstacle in the current frame obstacle grid map according to the current frame perception data of the dynamic obstacle, see Fig.11 , the unit specifically includes:

[0123] The ST region determination subunit 2201 is used to determine the ST region where the dynamic obstacle and the ego vehicle collide according to the current frame prediction trajectory of the dynamic obstacle and the ego vehicle reference line;

[0124] The first estimated occupied region determining subunit 2202 is used to discretize the ST region grid into the current frame ST grid map to obtain a first estimated occupied region;

[0125] The second estimated occupied area determination subunit 2203 is used to project the occupied area of ​​the previous frame of the dynamic obstacle into the ST grid map of the current frame to obtain a second estimated occupied area; and

[0126] The current frame occupied area determination subunit 2204 is used to determine 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.

[0127] In this embodiment, exemplarily, the second estimated occupied area determination subunit 2203 is used to displace the occupied area of ​​the previous frame of the dynamic obstacle according to the time change and the vehicle position change between the previous frame and the current frame, and obtain the second estimated area of ​​the dynamic obstacle in the ST grid map of the current frame.

[0128] The time interval (ie, update period) between the current frame and the previous frame can be set as required, for example, according to the hardware frequency of the sensor, etc. The present invention is not limited to this.

[0129] In this embodiment, illustratively, the filtering method used in the current frame occupied area determination subunit 2204 may include, for example, Bayesian filtering, Kalman filtering, etc. The present invention is not limited to this. 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 is taken.

[0130] Fig.12 The present invention is an obstacle risk field environment modeling device provided according to another embodiment of the present invention. Fig.13 As shown: In the device, the obstacle risk field construction module 300 specifically includes:

[0131] The static obstacle risk field construction unit 310 is used to use the obstacle position credibility in the current frame perception data of each static obstacle in each grid in the current frame occupancy area of ​​the current frame SL grid map as the current frame occupancy probability of the grid; and

[0132] The dynamic obstacle risk field construction unit 320 is used to calculate and update the current frame occupancy probability of each grid in the current frame occupancy area of ​​each dynamic obstacle in the current frame obstacle grid map to obtain the current frame dynamic obstacle risk field. Fig.13 The dynamic obstacle risk field construction unit 320 further includes:

[0133] The current frame predicted trajectory probability calculation and updating subunit 3201 is used to calculate and update the current frame predicted trajectory probability of each grid in the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map;

[0134] Current frame ST risk field establishment subunit 3202: used to determine the current frame occupancy probability of each occupied grid in the current frame ST grid map according to the current frame predicted trajectory probability of each grid in the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map, so as to obtain the current frame ST risk field.

[0135] Fig.14 The present invention is an obstacle risk field environment modeling device provided according to another embodiment of the present invention. Fig.11As shown: In the device, the current frame ST risk field establishment subunit 3202 specifically includes:

[0136] The target dynamic obstacle determination submodule 3202A is used to determine at least one target dynamic obstacle corresponding to each occupied grid according to the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid.

[0137] The current frame estimated occupancy probability determination submodule 3202B is used to determine the current frame estimated occupancy probability of the occupancy grid according to the current frame predicted trajectory probability of each target dynamic obstacle occupying the occupancy grid and the preset maximum occupancy probability;

[0138] The current frame occupancy probability determination submodule 3202C is used 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.

[0139] It should be noted that the specific implementation process and implementation principle of each module and unit of the obstacle risk field environment modeling device in the embodiment of the present invention can be specifically referred to the corresponding description of the above corresponding method embodiment, so they are not repeated here. Exemplarily, the obstacle risk field environment modeling device in the embodiment of the present invention can be any electronic device with a processor, and the electronic device includes but is not limited to a smart phone, a smart tablet, a personal PC, a computer, a cloud server, a controller, etc.

[0140] The embodiment of the present invention further provides a non-volatile computer storage medium, the computer storage medium stores computer executable instructions, and the computer executable instructions can execute the obstacle risk field environment modeling method in any of the above embodiments;

[0141] As an implementation mode, the non-volatile computer storage medium of the present invention stores computer executable instructions, and the computer executable instructions are configured as follows:

[0142] The obstacle grid map of the current frame is established in the Frenet coordinate system with the current position of the vehicle as the origin;

[0143] Determine, according to the current frame perception data of the obstacle, the current frame occupied area of ​​the obstacle in the current frame obstacle grid map;

[0144] The current frame occupancy probability of each grid in the current frame occupancy area of ​​each obstacle in the current frame obstacle grid map is calculated and updated to obtain the current frame obstacle risk field.

[0145] As a non-volatile computer-readable storage medium, it can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the method in the embodiment of the present invention. One or more program instructions are stored in the non-volatile computer-readable storage medium, and when executed by the processor, the obstacle risk field environment modeling method in any of the above method embodiments is executed.

[0146] 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, an application required for at least one function; the data storage area may store data created according to the use of the device, etc. In addition, the non-volatile computer-readable storage medium may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the non-volatile computer-readable storage medium may optionally include a memory remotely arranged 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.

[0147] In some embodiments, the embodiments of the present application also provide 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 of the above-mentioned obstacle risk field environment modeling methods.

[0148] 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 for structured roads of any embodiment of the present invention.

[0149] Fig.15 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 application, such as Fig.15 As shown, the device includes:

[0150] One or more processors 1510 and memory 1520, Fig.15 A processor 1510 is taken as an example.

[0151] The device for executing the obstacle risk field environment modeling method may further include: an input device 1030 and an output device 1540 .

[0152] The processor 1510, the memory 1520, the input device 1530 and the output device 1540 may be connected via a bus or other means. Fig.15 The example of connecting through bus is taken in the following.

[0153] The memory 1520, 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 program instructions / modules corresponding to the obstacle risk field environment modeling method in the embodiment of the present application. The processor 1510 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 1520, that is, the obstacle risk field environment modeling method of the above method embodiment is implemented.

[0154] The memory 1520 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store data created according to the use of the obstacle risk field environment modeling device, etc. In addition, the memory 1520 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 1520 may optionally include a memory remotely arranged relative to the processor 1510, and these remote memories may be connected to the obstacle risk field environment modeling 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.

[0155] The input device 1530 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 1540 can include display devices such as display screens.

[0156] The one or more modules are stored in the memory 1520, and when executed by the one or more processors 1510, the obstacle risk field environment modeling method in any of the above method embodiments is executed.

[0157] The above-mentioned product can execute the method provided in the embodiment of the present application, 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 in the embodiment of the present application.

[0158] The electronic device of the embodiment of the present application exists in various forms, including but not limited to:

[0159] (1) Mobile communication equipment: This type of equipment is characterized by having mobile communication functions and its main purpose is to provide voice and data communications. This type of terminal includes: smart phones, multimedia phones, functional phones, and low-end phones.

[0160] (2) Ultra-mobile personal computer devices: These devices fall into the category of personal computers, have computing and processing capabilities, and generally also have mobile Internet access features. These terminals include: PDAs, MIDs, and UMPC devices, such as tablet computers.

[0161] (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.

[0162] (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.

[0163] (5) Other onboard electronic devices with data interaction functions, such as on-board devices installed in vehicles.

[0164] The 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 (such as a passenger car, a sweeper, a sanitation vehicle, a bus, a minibus, a truck, a vacuum cleaner, a floor scrubber), a sweeping robot, etc.

[0165] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0166] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for 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 some parts of the embodiments.

[0167] 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 embodiments of the present invention.

Claims

1. A method for modeling an obstacle risk field environment, characterized in that: include: The obstacle grid map of the current frame is established in the Frenet coordinate system with the current position of the vehicle as the origin; Determine, according to the current frame perception data of the obstacle, the current frame occupied area of ​​the obstacle in the current frame obstacle grid map; Calculate and update the current frame occupancy probability of each grid in the current frame occupancy area of ​​each obstacle in the current frame obstacle grid map to obtain the current frame obstacle risk field, where obstacles include static obstacles and dynamic obstacles. The current frame obstacle grid map is established in the Frenet coordinate system with the current position of the vehicle as the origin, specifically including: Establish the current frame SL grid map corresponding to the static obstacle in the frenet coordinate system with the current position of the vehicle as the origin; and 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. The method of determining the current occupied area of ​​the obstacle in the obstacle grid map of the current frame according to the current frame perception data of the obstacle specifically includes: When the obstacle is a dynamic obstacle, determining the ST area where the dynamic obstacle collides with the ego vehicle according to the current frame predicted trajectory of the dynamic obstacle and the ego vehicle reference line; Discretizing the ST region grid into the current frame ST grid map to obtain a first estimated occupied region; Projecting the occupied area of ​​the previous frame of the dynamic obstacle into the ST grid map of the current frame to obtain a second estimated occupied area; The current frame occupied area of ​​the dynamic obstacle in the current frame ST grid map is determined according to the first estimated occupied area and the second estimated occupied area.

2. The method according to claim 1, characterized in that Determining the current occupied area of ​​the obstacle in the obstacle grid map of the current frame according to the current frame perception data of the obstacle, specifically including: When the obstacle is a static obstacle, the coverage area of ​​the static obstacle is determined according to the position point and obstacle size in the current frame perception data of the static obstacle; the coverage area is projected onto the current frame SL grid map to obtain the current frame occupied area of ​​the static obstacle in the current frame SL grid.

3. The method according to claim 1, characterized in that Projecting the area occupied by the dynamic obstacle in the previous frame onto the ST grid map of the current frame to obtain a second estimated area specifically includes: The area occupied by the dynamic obstacle in the previous frame is displaced according to the time variation and the vehicle position variation between the previous frame and the current frame to obtain a second estimated area of ​​the dynamic obstacle in the ST grid map of the current frame.

4. The method according to claim 1, characterized in that: Calculate and update the current frame occupancy probability of each grid in the current frame occupancy area of ​​each obstacle in the current frame obstacle grid map to obtain the current frame obstacle risk field, specifically including: For static obstacles, the current frame occupancy probability of each grid in the current frame occupancy area of ​​each static obstacle in the current frame SL grid map is calculated and updated to obtain the current frame SL risk field.

5. The method according to claim 4, characterized in that Calculate and update the current frame occupancy probability of each grid in the current frame occupancy area of ​​each static obstacle in the current frame SL grid map, specifically including: For each grid in the current frame occupancy area of ​​each static obstacle in the current frame SL grid map, the obstacle position credibility in the current frame perception data of the grid is determined as the current frame occupancy probability of the grid.

6. The method according to claim 1, characterized in that Calculate and update the current frame occupancy probability of each grid in the current frame occupancy area of ​​each obstacle in the current frame obstacle grid map to obtain the current frame obstacle risk field, specifically including: For dynamic obstacles, calculate and update the current frame predicted trajectory probability of each grid in the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map; The current frame occupancy probability of each occupied grid in the current frame ST grid map is determined according to the current frame predicted trajectory probability of each grid in the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map to obtain the current frame ST risk field.

7. The method according to claim 6, characterized in that Determining the current frame occupancy probability of each occupied grid in the current frame ST grid map according to the current frame predicted trajectory probability of each grid in the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid map specifically includes: For each occupied grid in the current frame ST grid map, the following steps are performed: according to the current frame occupied area of ​​each dynamic obstacle in the current frame ST grid, at least one target dynamic obstacle corresponding to each occupied grid is determined; according to the current frame predicted trajectory probability of each target dynamic obstacle occupying the occupied grid and the preset maximum occupancy probability, the current frame estimated occupancy probability of the occupied grid is determined; according to the current frame estimated occupancy probability of the occupied grid and the previous frame occupancy probability of the occupied grid, the current frame occupancy probability of the occupied grid is calculated.

8. An obstacle risk field environment modeling device, characterized in that: include: The mapping module is used to build the obstacle grid map of the current frame in the Frenet coordinate system with the current position of the vehicle as the origin; An occupied area determination module, used to determine the current frame occupied area of ​​the obstacle in the current frame obstacle grid map according to the current frame perception data of the obstacle; The obstacle risk field construction module is used to calculate and update the current frame occupancy probability of each grid in the current frame occupancy area of ​​each obstacle in the current frame obstacle grid map to obtain the current frame obstacle risk field. The mapping module further comprises: An SL grid map establishing unit, used to establish a current frame SL grid map corresponding to a static obstacle in a frenet coordinate system with the current position of the vehicle as the origin; and The ST grid map establishing unit is used to establish 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. The occupied area determination module includes a dynamic obstacle occupied area determination unit, which is used to determine the current frame occupied area of ​​the dynamic obstacle in the current frame obstacle grid map according to the current frame perception data of the dynamic obstacle, and the dynamic obstacle occupied area determination unit includes: The ST region determination subunit is used to determine the ST region where the dynamic obstacle and the ego vehicle collide based on the current frame prediction trajectory of the dynamic obstacle and the ego vehicle reference line; A first estimated occupied region determining subunit, configured to discretize the ST region grid into a current frame ST grid map to obtain a first estimated occupied region; A second estimated occupied area determination subunit is configured to project the occupied area of ​​the dynamic obstacle in a previous frame into the ST grid map of the current frame to obtain a second estimated occupied area; and The current frame occupied area determination subunit is used to determine 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.

9. A storage medium storing a computer program, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

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

11. An electronic device, characterized in that: include: 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 described in any one of claims 1 to 7.

12. A mobile tool, characterized in that: The electronic device comprising claim 11.

Images