Obstacle SLT spatial risk field environment modeling method and its device, and related products

Abstract

The present invention discloses a method, device, and related products for modeling an obstacle SLT risk field environment. The method comprises: establishing a current-frame SLT spatial coordinate system in a Frenet coordinate system with the vehicle's position as the origin, wherein the S axis of the SLT spatial coordinate system is based on the current lane of the vehicle's current frame, and the current-frame ST grid maps corresponding to each discrete lane are arranged along the L axis according to a preset arrangement order; and determining the current-frame occupied area of ​​each obstacle in the current-frame ST grid map corresponding to each discrete lane and the current-frame occupancy probability of each grid in the current-frame occupied area based on the predicted trajectory of the current-frame obstacle, thereby obtaining the SLT spatial risk field of the current-frame obstacle. The environmental modeling method of the present invention establishes a three-dimensional SLT spatial risk field for obstacles based on the Frenet coordinate system, which can be used by a decision-making and planning module for joint horizontal and vertical planning evaluation.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving lane environment modeling, and in particular to an obstacle SLT spatial risk field environment modeling method, an obstacle SLT spatial risk field environment modeling device, 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] Autonomous driving environment modeling integrates objective data such as perception, positioning, and high-precision maps. Perception data describes the real-time environment, while positioning and high-precision map data describe the objective road topology. Environmental modeling converts input data into an environmental model based on a lane model and supplemented with semantic information, making it easier to apply to decision-making and planning modules. Currently, most environmental modeling methods are based on a three-lane model or an extended three-lane model.

[0004] The current method for establishing an obstacle risk field describes dynamic and static obstacles separately. The inventors have discovered that this separate description prevents back-end decision-making and planning from establishing lateral and longitudinal correlations when calculating based on the environmental model, leading to issues with the output results. Specifically, obstacle modeling is still categorized by dynamic and static characteristics. The SL map (SL_map) influences the vehicle's lateral behavior, while the ST map (ST_map) influences its longitudinal behavior. The decision-making and planning process for these lateral and longitudinal behaviors lacks a unified evaluation, potentially leading to issues with the generated paths and path speeds, such as poor comfort and abrupt obstacle avoidance routes.

[0005] Therefore, the above-mentioned environmental modeling method needs to be further improved. 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 SLT spatial risk field environment, comprising:

[0008] A current-frame SLT spatial coordinate system is established in the Frenet coordinate system with the vehicle's position as the origin. The S axis of the SLT spatial coordinate system is based on the current lane of the vehicle in the current frame, and the current-frame ST grid images corresponding to the discrete lanes are arranged along the L axis according to a preset arrangement order. The discrete lanes include a target lane and an offset lane corresponding to the target lane. The target lane includes the vehicle's current lane and an adjacent lane to the current lane. The offset lane of the target lane refers to a lane obtained by offsetting the target lane according to a preset offset rule.

[0009] According to the predicted trajectory of the obstacle in the current frame, the current frame occupied area of ​​each obstacle in the current frame ST grid map corresponding to each discrete lane and the current frame occupancy probability of each grid in the current frame occupied area are determined to obtain the SLT spatial risk field of the obstacle in the current frame.

[0010] In a second aspect, an embodiment of the present invention provides an obstacle SLT spatial risk field modeling device, comprising:

[0011] An SLT spatial coordinate system construction module is used to establish a current frame SLT spatial coordinate system in the Frenet coordinate system with the vehicle position as the origin, wherein the S axis in the SLT spatial coordinate system is based on the current lane of the vehicle in the current frame, and the current frame ST grid images corresponding to the discrete lanes are arranged along the L axis according to a preset arrangement order; wherein the discrete lanes include a target lane and an offset lane corresponding to the target lane, the target lane includes the vehicle's current lane and an adjacent lane of the current lane, and the offset lane of the target lane refers to a lane obtained by offsetting the target lane according to a preset offset rule; and

[0012] The obstacle SLT spatial risk field construction module is used to determine the current frame occupied area of ​​each obstacle in the current frame ST grid map corresponding to each discrete lane and the current frame occupancy probability of each grid in the current frame occupied area based on the predicted trajectory of the obstacle in the current frame, so as to obtain the SLT spatial risk field of the obstacle in the current frame.

[0013] In a third aspect, an embodiment of the present invention provides a storage medium, in which one or more programs including execution instructions are stored, and the execution instructions can be read and executed by an electronic device (including but not limited to a computer, a server, or a network device, etc.) to execute the steps of the method provided according to the first aspect of the present invention.

[0014] In a fourth aspect, an embodiment of the present invention provides a computer program product comprising instructions, which, when run on a computer, enables the computer to execute the steps of the method provided in the first aspect.

[0015] 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.

[0016] In a sixth aspect, an embodiment of the present invention provides a mobile tool, which includes the electronic device shown in the fifth aspect.

[0017] An embodiment of the present invention proposes a method for modeling the SLT spatial risk field environment of obstacles for structured roads, which uniformly describes the static obstacle SL probability grid map and the dynamic obstacle ST probability grid map, and establishes a three-dimensional SLT spatial risk field of obstacles based on the Frenet coordinate system, which can be used for the decision-making planning module to perform horizontal and vertical joint planning evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] 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.

[0019] Figure 1 1 is a flow chart of a method for modeling an obstacle SLT spatial risk field environment provided by one embodiment of the present invention;

[0020] Figure 2 A schematic diagram of the SLT spatial coordinate system provided in an embodiment of the present invention;

[0021] Figure 3 1 is a flow chart of a method for modeling an obstacle SLT spatial risk field environment provided by another embodiment of the present invention;

[0022] Figure 4 ROI area diagram provided by an embodiment of the present invention;

[0023] Figure 5 is a schematic diagram of the SL_ROI region provided by an embodiment of the present invention;

[0024] Figure 6 Yes Figure 3 The sub-process diagram of step S121 is shown in FIG.

[0025] Figure 7 Yes Figure 3 The sub-process diagram of step S122 is shown in FIG.

[0026] Figure 8 1 is a flow chart of a method for modeling an obstacle SLT spatial risk field environment provided by another embodiment of the present invention;

[0027] Figures 9A to 9C 1 is a schematic diagram of calculating the SLT of an obstacle of interest provided by an embodiment of the present invention;

[0028] Figure 10 1 is a flow chart of a method for modeling an obstacle SLT spatial risk field environment provided by another embodiment of the present invention;

[0029] Figure 11A to Figure 11D A schematic diagram of the lateral dispersion focusing on the obstacle prediction line is shown;

[0030] FIG. 12A to FIG. 12B A longitudinal discrete diagram focusing on the obstacle prediction line is schematically shown;

[0031] Figure 13 It is a flowchart of a method for determining the current frame trajectory probability of each grid in the current frame ST grid map corresponding to each discrete lane for focusing on obstacles;

[0032] Figure 14 This is a principle block diagram of an obstacle SLT spatial risk field environment modeling device provided by one embodiment of the present invention;

[0033] Figure 15 This is a principle block diagram of an obstacle SLT spatial risk field environment modeling device provided by another embodiment of the present invention;

[0034] Figure 16 yes Figure 15 FIG. 2 is a block diagram showing a principle of an obstacle determination unit 201; FIG.

[0035] Figure 17 yes Figure 15 Schematic diagram of the current frame occupied area determination unit 203;

[0036] Figure 18 yes Figure 15 Schematic diagram of the current frame occupancy probability determination unit 204;

[0037] Figure 19 yes Figure 17 Principle block diagram of the first estimated occupied area determination subunit 2031;

[0038] Figure 20 yes Figure 18 Principle block diagram of the current frame predicted trajectory probability determination subunit 2041;

[0039] Figure 21 FIG. 1 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] 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.

[0041] It should be noted that, unless there is any conflict, the embodiments and features in the embodiments of this application can be combined with each other.

[0042] 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.

[0043] 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.

[0044] 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.

[0045] 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. The technical solution of the present invention mainly realizes the environmental modeling of the obstacle risk field under structured roads, which is mainly represented by the grid method based on the Frenet coordinate system (also known as the road coordinate system). That is, the spatial position of the obstacle in the grid map, that is, the spatial position in the structured road, is represented by a grid. The risk field defined in the present invention can be represented by the spatial position of each grid point in the grid map and the relationship between its corresponding occupancy probability.

[0046] The road sequence in this paper refers to the form in which lanes are stored in the map, which can be distinguished according to the attributes of the lanes.

[0047] Refer to the following Figure 1 , specifically describe an obstacle SLT spatial risk field environment modeling method provided by an embodiment of the present invention, which can be applied to any mobile tool that can achieve 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 1 As shown, the method includes:

[0048] Step S11: Establish a current-frame SLT spatial coordinate system in the Frenet coordinate system with the vehicle's position as the origin. The S axis in the SLT spatial coordinate system is based on the current lane of the vehicle in the current frame, and the current-frame ST grid images corresponding to the discrete lanes are arranged along the L axis according to a preset arrangement order. The discrete lanes include a target lane and an offset lane corresponding to the target lane. The target lane includes the vehicle's current lane and an adjacent lane to the current lane. The offset lane of the target lane refers to a lane obtained by offsetting the target lane according to a preset offset rule.

[0049] The preset offset rule may be set as needed, for example, the target lane may be offset by a gradually increasing or decreasing distance, or the target lane may be offset by a fixed distance.

[0050] Figure 2 The schematic diagram of the established SLT space coordinate system is shown, in which a three-lane model is used. Figure 2 The left figure shows the target lane, which includes the current lane where the vehicle is located, as well as the left and right lanes adjacent to the current lane. The centerline of the vehicle lane is discretized into a fixed distance △l around l0 to obtain its offset lanes l1 and l2. Similarly, the centerline of the left lane and the centerline of the right lane can also be discretized into corresponding offset lanes. The target lane and the offset lane corresponding to the target lane are collectively referred to as discrete lanes. It can be understood that for the purpose of simplicity, Figure 2 The left image of the ego vehicle omits the left and right lane offsets. Similarly, if the ego vehicle's current lane has no adjacent left lane and only an adjacent right lane, the target lane includes the current lane and its adjacent right lane. If the ego vehicle's current lane has no adjacent right lane and only a left lane, the target lane includes the current lane and its adjacent left lane. If the ego vehicle's current lane has no adjacent left or right lanes, the target lane includes the current lane.

[0051] From this we can get the series l of offset △l k The corresponding ST grid map of the current frame (which can be represented by ST_map) is as follows (i.e., multiple discrete lanes obtained after offset, k>0): Figure 2 As shown in the figure on the right.

[0052] Step S12: Determine the current frame occupied area of ​​each obstacle in the current frame ST grid map corresponding to each discrete lane and the current frame occupancy probability of each grid in the current frame occupied area based on the predicted trajectory of the obstacle in the current frame, so as to obtain the SLT spatial risk field of the obstacle in the current frame.

[0053] The predicted trajectory of the obstacle in the current frame is output by a known upstream prediction module. The prediction module is an upstream module of the environment modeling of the present invention, and the present invention does not limit this.

[0054] The environmental modeling method of the obstacle SLT spatial risk field provided by the embodiment of the present invention arranges the current frame ST grid images corresponding to the discrete lanes consisting of the target lane and the offset lanes obtained by offsetting the target lane according to the preset offset rule along the L axis in a preset arrangement order to establish an SLT three-dimensional spatial coordinate system, and establishes the SLT spatial risk field on this basis, which helps the decision-making planning module to jointly evaluate the horizontal and vertical indicators during the search process, thereby facilitating the generation of the riding comfort of the path.

[0055] Further references Figure 3 , which shows a flowchart of another embodiment of the obstacle SLT space risk field environment modeling method provided by the present invention. The flowchart is mainly in Figure 1 The flowchart of the steps further defined on the basis of step S12.

[0056] like Figure 3 As shown, for each discrete lane, the following steps are performed:

[0057] Step S121: determining an obstacle of interest from among the obstacles based on the predicted trajectory of each obstacle and the center line of the discrete lane;

[0058] Step S122: For each obstacle of interest, determining an ST region where the obstacle of interest conflicts with the discrete lane based on the predicted trajectory of the obstacle of interest and the centerline of the discrete lane; and determining an area occupied by the obstacle of interest in the current frame of the ST grid map of the discrete lane in the current frame based on the ST region.

[0059] Step S123: Calculate the current frame occupancy probability of each occupied grid in the current frame occupied area in the current frame ST grid map corresponding to the discrete lane.

[0060] In the present invention, the obstacles input by the perception need to be screened multiple times, that is, ROI obstacles and SL_ROI obstacles are screened out in sequence, and then the obstacles of interest are further screened out on this basis. For example, the obstacle screening process is as follows:

[0061] First, a perceptual region of interest (ROI) is defined. The ROI is determined based on the perceptual ability relative to the Cartesian coordinate system. Figure 4 The figure shows a schematic diagram of the ROI region. The dotted box indicates the ROI region. ROI_front_x is the front length of the ROI region, ROI_back_x is the back length of the ROI region, and ROI_y is the half-width of the ROI region. The purpose of defining the ROI region is to filter out obstacles that are far away from the vehicle. The filtered obstacles are defined as ROI obstacles.

[0062] Secondly, the ROI obstacle is transformed and projected into the Frenet coordinate system. By framing the SL_ROI region, obstacles that are far away from the ego vehicle lane or the target lane are further filtered out. The filtered obstacles are defined as SL_ROI obstacles. Figure 5FIG2 shows a schematic diagram of the SL_ROI region, where the dotted box indicates the SL_ROI region, wherein SL_ROI_front_1 is the front length of the SL_ROI region, SL_ROI_back_1 is the back length of the SL_ROI region, and SL_ROI_1 is the half-width of the SL_ROI region.

[0063] Then, according to the road sequence that the dynamic obstacle prediction trajectory passes through and combined with the topological relationship in the high-precision map, the dynamic obstacles in the SL_ROI obstacles are further filtered to further determine the obstacles of interest.

[0064] For example, Figure 6 As shown, the method of determining the obstacle of interest from the obstacles based on the predicted trajectory of each obstacle and the center line of the discrete lane may include:

[0065] Step S1211: Determine a first road sequence corresponding to the discrete lane;

[0066] Step S1212: For each obstacle, determine the second road sequence corresponding to the predicted trajectory of the obstacle, and determine whether there is a potential conflict between the first road sequence and the second road sequence. If so, determine the obstacle as a focus obstacle; otherwise, determine the obstacle as a non-focus obstacle.

[0067] Obstacles of interest, as defined herein, can be identified by determining whether there is a potential conflict between the first road sequence corresponding to the discrete lanes and the second road sequence corresponding to each obstacle's predicted trajectory. If so, the obstacle is considered a potential obstacle of interest; otherwise, it is considered a non-obstacle of interest and eliminated. Specifically, all roads in the first and second road sequences are paired, and each pair is determined to determine whether there is a parallel, merging, diverging, or intersecting relationship. If so, a potential conflict is determined.

[0068] In this embodiment, the ST region where the obstacle of interest conflicts with the discrete lane in step S122 can be determined using methods known in the art, which are not limited in the present invention. For example, the collision start time t (i.e., min_t) and s (i.e., min_s) of the collision between the ego vehicle and the obstacle of interest, as well as the collision end time t (i.e., max_t) and s (i.e., max_s) are determined based on the obstacle's predicted trajectory in the current frame, the dynamic obstacle's dimensions, the centerline of the discrete lane, and the ego vehicle's dimensions. The ST region where the obstacle of interest conflicts with the discrete lane 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 obstacle of interest to represent the obstacle of interest, 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 obstacle of interest is used as the center point of the first virtual frame, and each waypoint on the center line of the discrete lane 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 center line of the discrete lane. The time points when the collision between the two begins and ends and their corresponding s coordinate values ​​are recorded.

[0069] Intuitively, the prediction module outputs a predicted trajectory for each obstacle of interest. According to the above method, a ST region can be uniquely determined by a predicted trajectory. Exemplarily, for a dynamic obstacle, the ST region is a parallelogram.

[0070] It should be noted that static obstacles can be calculated as a rectangular ST region. The calculation process for the SLT spatial risk field for static obstacles is similar to that for dynamic obstacles. Because static obstacles do not involve the horizontal and vertical discretization of the predicted trajectory, this article does not provide a detailed description of their SLT spatial risk field implementation.

[0071] Further references Figure 7 , which shows a flowchart of another embodiment of the obstacle SLT space risk field environment modeling method provided by the present invention. The flowchart is mainly in Figure 3 The flowchart of the steps further defined on the basis of step S122.

[0072] like Figure 7 As shown, determining the current frame occupied area of ​​the obstacle of interest in the current frame ST grid map of the discrete lane according to the ST area can be specifically implemented as follows:

[0073] S1221: Discretize the ST area grid into the current frame ST grid map corresponding to the discrete lane to obtain a first estimated occupied area;

[0074] S1222: Projecting the area occupied by the obstacle in the previous frame of the ST grid map corresponding to the discrete lane into the current frame of the ST grid map corresponding to the discrete lane to obtain a second estimated area of ​​occupation;

[0075] 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 the second estimated area of ​​the obstacle in the ST grid map of the current frame.

[0076] 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:

[0077] s m =s m m-1 -△s; (1)

[0078] t m =t m m-1 -△t; (2)

[0079] 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.

[0080] 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.

[0081] S1223: Determine the current frame occupied area of ​​the obstacle of interest in the current frame ST grid map of the discrete lane according to the first estimated occupied area and the second estimated occupied area.

[0082] For example, the first estimated occupied area and the second estimated occupied area can be filtered to obtain the current frame occupied area of ​​the obstacle of interest in the current frame ST grid image of the discrete lane. Filtering can be performed using a common filtering method to obtain the current frame occupied area of ​​the obstacle of interest in the current frame ST grid image of the discrete lane. Filtering methods may include, for example, Bayesian filtering, Kalman filtering, etc. The present invention is not limited thereto.

[0083] 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.

[0084] Further references Figure 8 , which shows a flowchart of another embodiment of the obstacle SLT space risk field environment modeling method provided by the present invention. The flowchart is mainly in Figure 7 The flowchart of the steps further defining step S123 is shown in FIG. Figure 8 As shown, the current frame occupancy probability of each occupied grid in the current frame occupied area in the current frame ST grid map corresponding to the discrete lane is calculated, which can be implemented as follows:

[0085] Step S1231: For each obstacle of interest, calculate the current frame predicted trajectory probability of each grid in the first estimated occupied area of ​​the obstacle of interest in the current frame ST grid map corresponding to the discrete lane;

[0086] Step S1232: For each occupied grid in the current frame ST grid map corresponding to the discrete lane, perform the following steps: determine at least one target obstacle of interest corresponding to each occupied grid based on the current frame occupied area in the current frame ST grid map corresponding to the discrete lane of each obstacle of interest; determine the current frame estimated occupancy probability of the occupied grid based on the current frame predicted trajectory probability of each target obstacle of interest occupying the occupied grid and the preset maximum occupancy probability; calculate the current frame occupancy probability of the occupied grid based on the current frame estimated occupancy probability of the occupied grid and the previous frame occupancy probability of the occupied grid.

[0087] For example, the occupancy probability of the occupied grid in the current frame can be calculated based on the estimated occupancy probability of the occupied grid in the current frame and the occupancy probability of the occupied grid in the previous frame using a filtering formula. The filtering formula can be, for example, a Bayesian filtering formula or a Kalman filtering formula.

[0088] For each discrete lane, because each obstacle of interest has a corresponding predicted trajectory, there may be intersections or overlaps between the predicted trajectories, resulting in some occupied grids being occupied by multiple obstacles of interest 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 obstacles of interest in the occupied grid. Therefore, it is first necessary to determine which target obstacles of interest occupy each occupied grid. Exemplarily, for example, it can be determined whether the occupied grid is included in the current frame occupied area of ​​each obstacle of interest in the current frame ST grid map. If so, it is determined that the obstacle of interest is the target obstacle of interest occupying the occupied grid.

[0089] Below is one of the discrete lanes l kFor example, the current frame occupancy probability calculation process of each occupied grid in the current frame ST grid map corresponding to the discrete lane is as follows:

[0090] For example, the obstacle ID of interest is represented by i, the grid point number is represented by j, and the current frame predicted trajectory probability of a single target obstacle occupying the occupied grid j is expressed as i P m = i k m , the estimated occupancy probability of the occupied grid j in the current frame is:

[0091] P m =min(max_p,∑ i k m ) (3)

[0092] in i k m is the predicted trajectory probability of the target obstacle with value i in Id[] in the occupied grid j, max_p is the preset maximum occupancy probability, and the current frame occupancy probability of the occupied grid can be obtained according to the following Bayesian filtering formula:

[0093]

[0094]

[0095]

[0096] L is an intermediate variable. In formula (3), it is added when the grid point is occupied and subtracted when it is not occupied. At the same time, according to the physical meaning of the obstacle st, when t m m-1 <0, the P(s,t) risk field of the obstacle is cleared.

[0097] The current frame occupancy probability of each occupied grid in the current frame ST grid map corresponding to each discrete lane is updated and calculated in sequence. Thus, the SLT risk field of the target obstacle in the lane model in the current frame is continuously updated based on the historical frame data.

[0098] Exemplarily, the Bayesian filter may also be replaced by other filtering algorithms, such as the Kalman filter.

[0099] The obstacle SLT risk field calculated and updated using the method of this embodiment allows for a certain degree of error within the sensor's accuracy range from a single-frame perspective. It also transforms the original "hazard present / absent" result representation into a probabilistic "0-1" risk description, improving the data's applicability. Furthermore, the superposition of results from multiple consecutive frames enhances data stability, avoids the impact of single-frame jumps, and improves the accuracy of obstacle risk assessment.

[0100] In some embodiments, after obtaining the obstacles of interest, they are further divided into key obstacles of interest and non-key obstacles of interest. Based on the results of the potential conflict analysis described above, the obstacles of interest surrounding the vehicle and their adjacent obstacles are selected as key obstacles of interest. This selection can be made based on actual needs. For example, the obstacles of interest surrounding the vehicle can be selected as key obstacles of interest based on an area that approximates a "gate shape." The "gate shape" selection area here not only indicates that the obstacle is located within the "gate shape" area, but also indicates that the obstacle's predicted trajectory is contained within or passes through the "gate shape" area. Obstacles other than the key obstacles among the obstacles of interest are determined as non-key obstacles of interest.

[0101] For non-key obstacles, the ST area grid is discretized into the current frame ST grid map corresponding to the discrete lane in step S1221 to obtain a first estimated occupied area, which can be specifically implemented as follows:

[0102] According to the obstacle SL calculation result, the relative position relationship with the current lane can be known. The ST area of ​​the relevant discrete lanes is calculated in order from near to far (i.e., from the left lane to the right lane). Then, the ST area grid corresponding to each discrete lane is discretized into the current frame ST grid map corresponding to the corresponding discrete lane to obtain the first estimated occupied area. For example, Figure 9A As shown, the obstacle is on the left side of the current lane. First, calculate the ST area corresponding to the discrete lane l1 to determine whether the predicted line of the obstacle conflicts with l1. If there is no conflict, that is, ST_map(l1) is empty, then the ST_map of this obstacle corresponding to lanes l0 and l2 must be empty. If there is a conflict, that is, ST_map(l1) is not empty, then calculate ST_map(l0) based on tmax in ST_map(l1). If ST_map(l0) is not empty, then calculate ST_map(l2) based on tmax in ST_map(l0). After calculating ST_map(l1), ST_map(l0) and ST_map(l2), correspond to them in a unified SLT space coordinate system. As shown Figure 9B As shown, the ST areas corresponding to the discrete lane l1 and the discrete lane l0 are shown respectively. Figure 9C The ST regions corresponding to the discrete lanes l0 and l1 distributed in the SLT space coordinate system are shown. It is understood that for simplicity, the first estimated occupied regions in the grid of the current frame ST grid map corresponding to each discrete lane are not shown.

[0103] Accordingly, for non-key obstacles, the current frame trajectory probability of each grid in the first estimated occupied area of ​​the current frame ST grid map corresponding to each discrete lane can be directly determined.

[0104] For the key focus obstacles, the specific implementation of discretizing the ST area grid into the current frame ST grid map corresponding to the discrete lane in step S1221 to obtain the first estimated occupied area is similar to the above-mentioned non-key focus obstacles. The difference is that the predicted trajectory output by the prediction module of the key focus obstacles must be discretized horizontally and vertically.

[0105] Further references Figure 10 , which shows a flowchart of another embodiment of the obstacle SLT spatial risk field environment modeling method provided by the present invention. The flowchart mainly focuses on obstacles and is a flowchart of the steps that further define step S1221.

[0106] Step S1221A: Laterally discretizing the ST region to obtain a lateral discrete region, and grid-discretizing the lateral discrete region into the current frame ST grid map corresponding to the discrete lane to obtain an estimated lateral occupied area;

[0107] The horizontal line dispersion reflects the degree of deviation of the obstacle from the center line of the structured road.

[0108] Through lateral discretization, the prediction line (i.e., prediction trajectory) area is discretized from the normal prediction line area, i.e., a uniform width band area, to a variable width band area. The normal prediction line area can be understood as a uniform width band area with the prediction line as the axis and symmetrically distributed along the obstacle width. After discretization, it becomes a variable width band area with an initial width equal to the obstacle width, an end width equal to the end lane width, and containing the prediction line.

[0109] For example, the original uniform width band region can be gradually transformed into the variable width band region through linear gradient. Alternatively, the uniform width band region can be discretized into the variable width band regions with larger ranges according to a certain function operation.

[0110] Figure 11A to Figure 11D The schematic diagram of the horizontal discreteness of the obstacle prediction line is shown. Figure 11A shows the obstacle prediction line and the schematic diagram of the equal-width strip area, and Figure 11B shows the schematic diagram of the prediction line and the variable-width strip area after horizontal discreteness. Figure 11C The ST region of equal width corresponding to FIG11A is shown. Figure 11D FIG11B shows a variable width band-shaped ST region, i.e., a lateral discrete region. It can be understood that Figure 11C and Figure 11D Only the lateral discrete area with the center line of the current lane as the discrete lane is drawn, and the lateral discrete area is discretized into the current frame ST grid map corresponding to the discrete lane to obtain the estimated lateral occupied area.

[0111] Step S1221B: performing longitudinal discretization on the ST region to obtain at least one longitudinal discrete region, and grid-discretizing each longitudinal discrete region into the current frame ST grid map corresponding to the discrete lane to obtain a corresponding estimated longitudinal occupied area;

[0112] The longitudinal discretization reflects the different speeds of the obstacle under acceleration and deceleration conditions. For example, this step can be implemented as follows:

[0113] 1): Based on the speed of the obstacle corresponding to the obstacle prediction line, the discrete interval of the speed is set longitudinally and a speed probability density model is established;

[0114] The discrete interval of the set speed can be determined according to constraints such as road speed limit, vehicle comfort, etc. Exemplarily, the speed probability density model is a skewed probability density distribution model centered on the speed of the obstacle corresponding to the prediction line.

[0115] 2): Based on the prediction line and the speed probability density model, determine the longitudinal discrete areas of the key obstacle in the current frame ST grid map corresponding to the discrete lane at each discrete speed within the discrete interval.

[0116] For example, the distance (time) curve can be obtained by integrating the speed (time) curve corresponding to the speed probability density model. Since the position of the prediction line corresponding to the longitudinal discrete process remains unchanged, the S area range in the ST grid map remains unchanged. According to the distance (time) curve, the corresponding T range at different speeds can be obtained by inverse calculation, thereby obtaining each occupied area in the ST grid map, that is, each longitudinal discrete area.

[0117] 3): Discretize each longitudinal discrete area into the current frame ST grid map corresponding to the discrete lane to obtain the corresponding estimated longitudinal occupied area.

[0118] Figures 12A-12B The diagram schematically shows the longitudinal discretization of the obstacle prediction line. Figure 12A As shown in the figure, the speed V0 of the obstacle corresponding to the obstacle prediction line is changed from a single value to a discrete interval (V min ,V max ), and make a probability density skewed distribution with V0 as the center. The S(T) curve can be obtained by integrating different V(T) curves. Since the position of the prediction line corresponding to the longitudinal discrete process remains unchanged, the S area range in the ST area remains unchanged. According to the S(T) curve, the T range corresponding to different speeds V can be inversely calculated to obtain the longitudinal discrete ST area, such as Figure 12B As shown, the discrete interval (V min ,V max ) within V0, Vmin and V max The corresponding ST regions are the longitudinal discrete regions. Figures 12A-12B For simplicity, the corresponding ST grid diagram is not shown, that is, the grid discretization of each longitudinal discrete region into the current frame ST grid diagram corresponding to the discrete lane to obtain the corresponding estimated longitudinal occupied area is not shown.

[0119] Step S1221C: Determine a first estimated occupied area of ​​the obstacle of interest in the current frame ST grid map corresponding to the discrete lane based on the estimated lateral occupied area and each estimated longitudinal occupied area.

[0120] For example, the estimated lateral occupied area and each estimated longitudinal occupied area are filtered to obtain a first estimated occupied area of ​​the obstacle of interest in the current frame ST grid image corresponding to the discrete lane; filtering methods may include, for example, Bayesian filtering, Kalman filtering, etc. The present invention is not limited to this. Alternatively, for example, the estimated lateral occupied area and each estimated longitudinal occupied area can be directly superimposed (or unioned) to obtain the first estimated occupied area.

[0121] Furthermore, for each focused obstacle, the current frame trajectory probability of each grid in the first estimated occupied area in the current frame ST grid map corresponding to each discrete lane is as follows: Figure 13 As shown, it can be determined by the following process:

[0122] Step S1231A: Calculating the lateral occupancy probability of each grid in the estimated lateral occupancy area of ​​the obstacle of interest;

[0123] Step S1231B: Calculate the longitudinal occupancy probability of each grid in each estimated longitudinal occupancy area of ​​the obstacle of interest;

[0124] Step S1231C: For each grid of the first estimated occupied area of ​​the obstacle of interest in the current frame ST grid map corresponding to the discrete lane, the lateral occupancy probability and the longitudinal occupancy probability of the grid are superimposed to obtain the current frame predicted trajectory probability of the grid.

[0125] For example, for the estimated horizontal occupied area obtained in step S1221A, determining the horizontal occupied probability of each grid in the current frame can be achieved by the following method:

[0126] As described above, the estimated lateral occupancy area is a variable-width strip. By performing collision calculations on each point in the variable-width strip corresponding to the predicted line (predicted trajectory) of the key obstacle and the centerline of each discrete lane, a unique parallelogram region is obtained on the ST grid. Points of different widths form a variable-width strip with a slightly expanded range in the stacked ST grids. This allows the equal-probability ST grid for a single obstacle to be converted into a slightly expanded variable-probability ST grid.

[0127] For each estimated longitudinal occupied area obtained in step S1221B, the longitudinal occupied probability of each grid is calculated. For example, the velocity probability density model in step S12212 can be integrated according to the preset intervals according to the discrete intervals of the set velocity to obtain the probability P(v i ).

[0128] For example, the probability P(v i ) performs other processing, such as normalization processing, etc., to finally obtain the longitudinal occupancy probability of each grid point in each estimated longitudinal occupancy area.

[0129] In some embodiments, the horizontal occupancy probability and the vertical occupancy probability of the grid may be superimposed using a filtering algorithm, such as Bayesian filtering, to obtain the current frame predicted trajectory probability of the grid. Alternatively, the two may be directly added.

[0130] The embodiment of the present invention can further play a predictive role by discretizing the predicted trajectory of the key obstacle horizontally and vertically, and can predict the impact of reasonable changes in virtual obstacles on the ego vehicle in advance, thereby reserving sufficient planning time and space for the ego vehicle's decision-making planning.

[0131] Figure 14 The present invention provides an obstacle SLT space risk field environment modeling device according to another embodiment of the present invention. Figure 14 As shown: The device includes:

[0132] SLT space coordinate system construction module 100: used to establish the current frame SLT space coordinate system in the Frenet coordinate system with the vehicle position as the origin,

[0133] The S axis in the SLT spatial coordinate system is based on the current lane where the current frame of the vehicle is located, and the current frame ST grid diagram corresponding to each discrete lane is arranged along the L axis according to a preset arrangement order; wherein the discrete lane includes the target lane and the offset lane corresponding to the target lane, the target lane includes the current lane of the vehicle and the adjacent lane of the current lane, and the offset lane of the target lane refers to the lane obtained by offsetting the target lane according to the preset offset rule.

[0134] Obstacle SLT spatial risk field construction module 200: used to determine the current frame occupied area of ​​each obstacle in the current frame ST grid map corresponding to each discrete lane and the current frame occupancy probability of each grid in the current frame occupied area based on the predicted trajectory of the obstacle in the current frame, so as to obtain the SLT spatial risk field of the obstacle in the current frame.

[0135] The environmental modeling device of the obstacle SLT spatial risk field provided by the embodiment of the present invention arranges the current frame ST grid diagrams corresponding to the discrete lanes consisting of the target lane and the offset lanes obtained by offsetting the target lane according to the preset offset rule along the L axis in a preset arrangement order to establish an SLT three-dimensional spatial coordinate system, and establishes the SLT spatial risk field on this basis, which helps the decision-making planning module to jointly evaluate the horizontal and vertical indicators during the search process, thereby facilitating the generation of the riding comfort of the path.

[0136] Figure 15 The present invention provides an obstacle SLT space risk field environment modeling device according to another embodiment of the present invention. Figure 15 As shown: In this device, for each discrete lane, the obstacle SLT spatial risk field construction module 200 specifically includes:

[0137] The obstacle-of-interest determining unit 201 is configured to determine an obstacle of interest from among the obstacles based on the predicted trajectory of each obstacle and the center line of the discrete lane;

[0138] The ST region determining unit 202 is configured to determine, for each obstacle of interest, an ST region where the obstacle of interest conflicts with the discrete lane based on the predicted trajectory of the obstacle of interest and the center line of the discrete lane;

[0139] The current frame occupied area determining unit 203 is configured to determine, for each obstacle of interest, the current frame occupied area of ​​the obstacle of interest in the current frame ST grid map of the discrete lane according to the ST area;

[0140] The current frame occupancy probability determining unit 204 is configured to calculate the current frame occupancy probability of each occupied grid in the current frame occupancy area in the current frame ST grid map corresponding to the discrete lane.

[0141] In some embodiments, reference Figure 16 As shown, the obstacle determination unit 201 further includes:

[0142] The first road sequence determining subunit 2011 is configured to determine a first road sequence corresponding to the discrete lanes;

[0143] Obstacle of interest determination subunit 2012: for each obstacle, determining the second road sequence corresponding to the predicted trajectory of the obstacle, and judging whether there is a potential conflict between the first road sequence and the second road sequence. If so, the obstacle is determined to be an obstacle of interest; otherwise, the obstacle is determined to be a non-obstacle of interest.

[0144] In some embodiments, as Figure 17 As shown, the current frame occupied area determination unit 203 further includes:

[0145] The first estimated occupied area determining subunit 2031 is configured to discretize the ST area grid into the current frame ST grid map corresponding to the discrete lane to obtain a first estimated occupied area;

[0146] The second estimated occupied area determining subunit 2032 is configured to project the occupied area of ​​the obstacle in the previous frame ST grid map corresponding to the discrete lane into the current frame ST grid map corresponding to the discrete lane to obtain a second estimated occupied area; and

[0147] The current frame occupied area determining subunit 2033 is configured to determine the current frame occupied area of ​​the obstacle of interest in the current frame ST grid map of the discrete lane according to the first estimated occupied area and the second estimated occupied area.

[0148] In some embodiments, as Figure 18 As shown, the current frame occupancy probability determining unit 204 further includes:

[0149] The current frame predicted trajectory probability determination subunit 2041 is configured to calculate, for each obstacle of interest, the current frame predicted trajectory probability of each grid in the first estimated occupied area of ​​the obstacle in the current frame ST grid map corresponding to the discrete lane;

[0150] The current frame occupancy probability determination subunit 2042 is used to perform the following steps for each occupied grid in the current frame ST grid map corresponding to the discrete lane: determine at least one target obstacle of interest corresponding to each occupied grid based on the current frame occupied area in the current frame ST grid map corresponding to the discrete lane of each obstacle of interest; determine the current frame estimated occupancy probability of the occupied grid based on the current frame predicted trajectory probability of each target obstacle of interest occupying the occupied grid and the preset maximum occupancy probability; calculate the current frame occupancy probability of the occupied grid based on the current frame estimated occupancy probability of the occupied grid and the previous frame occupancy probability of the occupied grid.

[0151] In some embodiments, as Figure 19 As shown, for the focused obstacle, the first estimated occupied area determination subunit 2031 further includes:

[0152] The estimated lateral occupied area determination submodule 2031A is configured to perform lateral discretization on the ST area to obtain lateral discrete areas, and to grid-discretize the lateral discrete areas into the current frame ST grid map corresponding to the discrete lanes to obtain the estimated lateral occupied area.

[0153] The estimated longitudinal occupied area determination submodule 2031B is configured to longitudinally discretize the ST area to obtain at least one longitudinal discrete area, and grid-discretize each longitudinal discrete area into the current frame ST grid map corresponding to the discrete lane to obtain a corresponding estimated longitudinal occupied area; and

[0154] The first estimated occupied area determination submodule 2031C is configured to determine the first estimated occupied area of ​​the obstacle of interest in the current frame ST grid map corresponding to the discrete lane according to the estimated lateral occupied area and each estimated longitudinal occupied area.

[0155] Furthermore, if Figure 20 As shown, for the focused obstacle, the current frame predicted trajectory probability determination subunit 2041 further includes:

[0156] The lateral occupancy probability determination submodule 2041A is used to calculate the lateral occupancy probability of each grid in the estimated lateral occupancy area of ​​the obstacle of interest;

[0157] The longitudinal occupancy probability determination submodule 2041B is used to calculate the longitudinal occupancy probability of each grid in each estimated longitudinal occupancy area of ​​the obstacle of interest; and

[0158] The current frame predicted trajectory probability determination submodule 2041C is used to superimpose the lateral occupancy probability and the longitudinal occupancy probability of each grid in the first estimated occupancy area of ​​the current frame ST grid map corresponding to the discrete lane of the obstacle of interest to obtain the current frame predicted trajectory probability of the grid.

[0159] The embodiment of the present invention can further play a predictive role by discretizing the predicted trajectory of the key obstacle horizontally and vertically, and can predict the impact of reasonable changes in virtual obstacles on the ego vehicle in advance, thereby reserving sufficient planning time and space for the ego vehicle's decision-making planning.

[0160] It should be noted that the specific implementation process and implementation principles of the various modules and units of the obstacle SLT spatial risk field environment modeling device in the embodiment of the present invention can be found in the corresponding description of the corresponding method embodiment above, and therefore will not be repeated here. For example, the obstacle risk field environment modeling device in the embodiment of the present invention can be any electronic device with a processor, such as but not limited to a smartphone, smart tablet, personal PC, computer, cloud server, controller, etc.

[0161] An embodiment of the present invention further provides a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions can execute the obstacle SLT spatial risk field environment modeling method in any of the above embodiments;

[0162] 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:

[0163] A current-frame SLT spatial coordinate system is established in the Frenet coordinate system with the vehicle's position as the origin. The S axis of the SLT spatial coordinate system is based on the current lane of the vehicle in the current frame, and the current-frame ST grid images corresponding to the discrete lanes are arranged along the L axis according to a preset arrangement order. The discrete lanes include a target lane and an offset lane corresponding to the target lane. The target lane includes the vehicle's current lane and an adjacent lane to the current lane. The offset lane of the target lane refers to a lane obtained by offsetting the target lane according to a preset offset rule.

[0164] According to the predicted trajectory of the obstacle in the current frame, the current frame occupied area of ​​each obstacle in the current frame ST grid map corresponding to each discrete lane and the current frame occupancy probability of each grid in the current frame occupied area are determined to obtain the SLT spatial risk field of the obstacle in the current frame.

[0165] 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 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 SLT spatial risk field environment modeling method in any of the above-mentioned method embodiments.

[0166] 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.

[0167] 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 structured road obstacle SLT spatial risk field environment modeling methods.

[0168] 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 to enable the at least one processor to execute the SLT spatial risk field environment modeling method for structured road obstacles of any embodiment of the present invention.

[0169] Figure 21 This is a hardware structure diagram of an electronic device for executing a structured road obstacle SLT spatial risk field environment modeling method provided by another embodiment of the present application, such as Figure 21 As shown, the device includes:

[0170] One or more processors 2110 and memory 2120, Figure 21 A processor 2110 is taken as an example.

[0171] The device for executing the SLT spatial risk field environment modeling method for structured road obstacles may also include: an input device 2130 and an output device 2140.

[0172] The processor 2110, the memory 2120, the input device 2130 and the output device 2140 may be connected via a bus or other means. Figure 21 The bus connection is taken as an example.

[0173] Memory 2120, 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 method for modeling the SLT spatial risk field environment for obstacles on structured roads in the embodiments of the present application. Processor 2110 executes the non-volatile software programs, instructions, and modules stored in memory 2120 to execute various server functional applications and data processing, thereby implementing the method for modeling the SLT spatial risk field environment for obstacles on structured roads in the aforementioned method embodiment.

[0174] The memory 2120 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 structured road obstacle SLT spatial risk field environment modeling device, etc. In addition, the memory 2120 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 2120 may optionally include a memory remotely located relative to the processor 2110, and these remote memories may be connected to the structured road obstacle SLT spatial 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.

[0175] The input device 2130 can receive input digital or character information and generate signals related to user settings and function control of the structured road obstacle SLT spatial risk field environment modeling device. The output device 2140 can include a display device such as a display screen.

[0176] The one or more modules are stored in the memory 2120, and when executed by the one or more processors 2110, perform the SLT spatial risk field environment modeling method for obstacles on structured roads in any of the above method embodiments.

[0177] The above-mentioned product can execute the method provided in the embodiment of this 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 this application.

[0178] The electronic devices of the embodiments of the present application exist in various forms, including but not limited to:

[0179] (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.

[0180] (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.

[0181] (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.

[0182] (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.

[0183] (5) Other onboard electronic devices with data interaction functions, such as on-board devices installed in vehicles.

[0184] 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.

[0185] 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.

[0186] 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.

[0187] 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 the spatial risk field of an obstacle SLT, characterized in that: include: A current-frame SLT spatial coordinate system is established in the Frenet coordinate system with the vehicle's position as the origin. The S axis of the SLT spatial coordinate system is based on the current lane of the vehicle in the current frame, and the current-frame ST grid images corresponding to the discrete lanes are arranged along the L axis according to a preset arrangement order. The discrete lanes include a target lane and an offset lane corresponding to the target lane. The target lane includes the vehicle's current lane and an adjacent lane to the current lane. The offset lane of the target lane refers to a lane obtained by offsetting the target lane according to a preset offset rule. According to the predicted trajectory of the obstacle in the current frame, the current frame occupied area of ​​each obstacle in the current frame ST grid map corresponding to each discrete lane and the current frame occupied probability of each grid in the current frame occupied area are determined to obtain the SLT spatial risk field of the obstacle in the current frame. The current frame occupied area of ​​each obstacle in the current frame ST grid map corresponding to each discrete lane and the current frame occupied probability of each grid in the current frame occupied area are determined according to the predicted trajectory of the obstacle in the current frame, specifically including: For each discrete lane, perform the following steps: Determining an obstacle of interest from among the obstacles based on the predicted trajectory of each obstacle and the center line of the discrete lane; For each obstacle of interest, determining an ST region where the obstacle of interest conflicts with the discrete lane based on the predicted trajectory of the obstacle of interest and the center line of the discrete lane; determining an area occupied by the obstacle of interest in the current frame of the ST grid map of the discrete lane based on the ST region; and Calculate the current frame occupancy probability of each occupied grid in the current frame occupied area in the current frame ST grid map corresponding to the discrete lane, The step of determining the area occupied by the obstacle in the current frame ST grid map of the discrete lane according to the ST area specifically includes: Discretizing the ST area grid into the current frame ST grid map corresponding to the discrete lane to obtain a first estimated occupied area; Projecting the previous frame occupied area of ​​the obstacle of interest in the previous frame ST grid map corresponding to the discrete lane into the current frame ST grid map corresponding to the discrete lane to obtain a second estimated occupied area; and The current frame occupied area of ​​the obstacle of interest in the current frame ST grid map of the discrete lane 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 obstacle of interest from the obstacles based on the predicted trajectory of each obstacle and the center line of the discrete lane, specifically including: Determining a first road sequence corresponding to the discrete lane; For each obstacle, a second road sequence corresponding to the predicted trajectory of the obstacle is determined, and a determination is made as to whether there is a potential conflict between the first road sequence and the second road sequence. If so, the obstacle is determined to be a concern obstacle; otherwise, the obstacle is determined to be a non-concern obstacle.

3. The method according to claim 1, characterized in that Discretizing the ST region grid into the current frame ST grid map corresponding to the discrete lane to obtain a first estimated occupied area specifically includes: Laterally discretizing the ST region to obtain a lateral discrete region, and grid-discretizing the lateral discrete region into the current frame ST grid map corresponding to the discrete lane to obtain an estimated lateral occupied area; Longitudinally discretizing the ST region to obtain at least one longitudinal discrete region, and grid-discretizing each longitudinal discrete region into the current frame ST grid map corresponding to the discrete lane to obtain a corresponding estimated longitudinal occupied area; A first estimated occupied area of ​​the obstacle of interest in the current frame ST grid map corresponding to the discrete lane is determined according to the estimated lateral occupied area and each estimated longitudinal occupied area.

4. The method according to claim 3, characterized in that Calculating the current frame occupancy probability of each grid in the current frame occupancy area in the current frame ST grid map corresponding to the discrete lane specifically includes: For each obstacle of interest, the lateral occupancy probability of each grid in the estimated lateral occupancy area of ​​the obstacle of interest is calculated; and the longitudinal occupancy probability of each grid in the estimated longitudinal occupancy area of ​​the obstacle of interest is calculated. For each grid in the first estimated occupancy area of ​​the obstacle of interest in the current frame ST grid map corresponding to the discrete lane, the lateral occupancy probability and the longitudinal occupancy probability of the grid are superimposed to obtain the current frame predicted trajectory probability of the grid; For each occupied grid in the current frame ST grid map corresponding to the discrete lane, at least one target dynamic obstacle corresponding to each occupied grid is determined based on the current frame occupied area of ​​each obstacle in the current frame ST grid map corresponding to the discrete lane; the current frame estimated occupancy probability of the occupied grid is determined based on the current frame predicted trajectory probability of each target dynamic obstacle occupying the occupied grid and the preset maximum occupancy probability; the current frame occupancy probability of the occupied grid is calculated based on the current frame estimated occupancy probability of the occupied grid and the previous frame occupancy probability of the occupied grid.

5. An obstacle SLT spatial risk field modeling device, characterized in that: include: An SLT spatial coordinate system construction module is used to establish a current-frame SLT spatial coordinate system in the Frenet coordinate system with the vehicle position as the origin, wherein the S axis in the SLT spatial coordinate system is based on the current lane in which the vehicle is currently located in the current frame, and the current-frame ST grid images corresponding to the discrete lanes are arranged along the L axis in a preset arrangement order; wherein the discrete lanes include a target lane and an offset lane corresponding to the target lane, the target lane includes the vehicle's current lane and an adjacent lane to the current lane, and the offset lane of the target lane refers to a lane obtained by offsetting the target lane according to a preset offset rule; The obstacle SLT spatial risk field construction module is used to determine the current frame occupied area of ​​each obstacle in the current frame ST grid map corresponding to each discrete lane and the current frame occupancy probability of each grid in the current frame occupied area based on the predicted trajectory of the obstacle in the current frame, so as to obtain the SLT spatial risk field of the obstacle in the current frame. The current frame occupied area of ​​each obstacle in the current frame ST grid map corresponding to each discrete lane and the current frame occupied probability of each grid in the current frame occupied area are determined according to the predicted trajectory of the obstacle in the current frame, specifically including: For each discrete lane, perform the following steps: Determining an obstacle of interest from among the obstacles based on the predicted trajectory of each obstacle and the center line of the discrete lane; For each obstacle of interest, determining an ST region where the obstacle of interest conflicts with the discrete lane based on the predicted trajectory of the obstacle of interest and the center line of the discrete lane; determining an area occupied by the obstacle of interest in the current frame of the ST grid map of the discrete lane based on the ST region; and Calculate the current frame occupancy probability of each occupied grid in the current frame occupied area in the current frame ST grid map corresponding to the discrete lane, The step of determining the area occupied by the obstacle in the current frame ST grid map of the discrete lane according to the ST area specifically includes: Discretizing the ST area grid into the current frame ST grid map corresponding to the discrete lane to obtain a first estimated occupied area; Projecting the previous frame occupied area of ​​the obstacle of interest in the previous frame ST grid map corresponding to the discrete lane into the current frame ST grid map corresponding to the discrete lane to obtain a second estimated occupied area; and The current frame occupied area of ​​the obstacle of interest in the current frame ST grid map of the discrete lane is determined according to the first estimated occupied area and the second estimated occupied area.

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, 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 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