Data processing method and path planning method for moving object

Abstract

The embodiment of the invention provides a data processing method and a path planning method of a moving object, and in the data processing method, end-state lateral displacement data and moving state adjustment safety duration data of the moving object can be calculated in each frame. And further, obtaining transverse offset step length data of the current frame, and further obtaining target transverse offset data corresponding to the current frame in combination with the target transverse offset data of the previous frame. The target lateral offset data corresponding to the current frame is lateral displacement data of the end point of the travel path planned for the moving object in the current frame relative to the initial moving reference line. Therefore, according to the method, the target transverse offset data of the end point of the travel path planned by the current frame for the moving object is adjusted frame by frame, so that the numerical value change process between the target transverse offset data corresponding to the two adjacent frames is relatively gentle, and the moving object can stably bypass the wound object.

Description

Technical Field

[0001] The present application relates to the field of computer data processing, and in particular to a data processing method, device, electronic device, and storage medium. The present application also relates to another data processing method, device, electronic device, and storage medium. The present application also relates to another data processing method, device, electronic device, and storage medium. The present application also relates to a path planning method, device, electronic device, and storage medium for a mobile object. The present application also relates to another path planning method, device, electronic device, and storage medium for a mobile object. Background Art

[0002] During the moving process, a mobile object may encounter an object to be bypassed, and it is necessary to bypass the object. In the prior art, the mobile object moves through the boundary of the object to bypass the object, which easily leads to the mobile object being unable to bypass the object smoothly during the process. Therefore, how to make the mobile object bypass the object relatively smoothly is a problem that needs to be solved. Summary of the invention

[0003] The embodiment of the present application provides a data processing method to enable a mobile object to bypass an object relatively smoothly. The present application also relates to another data processing method, device, electronic device, and storage medium. The present application also relates to another data processing method, device, electronic device, and storage medium. The present application also relates to a path planning method, device, electronic device, and storage medium for a mobile object. The present application also relates to another path planning method, device, electronic device, and storage medium for a mobile object.

[0004] The embodiment of the present application provides a data processing method, comprising: obtaining final state lateral displacement data determined for a mobile object in a current frame, the final state lateral displacement data being lateral displacement data of the mobile object relative to an initial movement reference line used by the mobile object at an initial position point at an expected final state, the expected final state being an expected state with a lateral safety space between the mobile object and the bypassed object required to be reached by the mobile object in order to bypass the bypassed object, and the initial position point being an initial position point of a path required to be traveled by the mobile object in order to bypass the bypassed object; obtaining moving state adjustment safety duration data determined for the mobile object in the current frame, the moving state adjustment safety duration data being expected available duration data for adjusting the moving state starting from the current frame for the mobile object to avoid collision with the bypassed object; and calculating the moving state adjustment safety duration data based on the final state lateral displacement data and the moving state adjustment safety duration data. , obtain the lateral offset step data corresponding to the current frame; according to the lateral offset step data corresponding to the current frame, adjust the target lateral offset data corresponding to the previous frame, obtain the target lateral offset data determined for the mobile object in the current frame, as the target lateral offset data corresponding to the current frame, the target lateral offset data corresponding to the previous frame is the lateral displacement data between the previous frame target position point determined for the mobile object in the previous frame and the initial movement reference line used by the mobile object at the initial position point, the previous frame target position point is the end point of the travel path planned for the mobile object in the previous frame, the target lateral offset data corresponding to the current frame is the lateral displacement data between the current frame target position point and the initial movement reference line used by the mobile object at the initial position point, the current frame target position point is the end point of the travel path planned for the mobile object in the current frame.

[0005] The embodiment of the present application also provides another data processing method, including: obtaining the moving speed data of the moving object in the current frame; obtaining the lateral offset step data corresponding to the current frame according to the moving speed data of the moving object in the current frame, and the correspondence between the moving speed data range and the single-frame lateral offset step data range; adjusting the target lateral offset data corresponding to the previous frame according to the lateral offset step data corresponding to the current frame, and obtaining the target lateral offset data determined for the moving object in the current frame as the target lateral offset data corresponding to the current frame; wherein the target lateral offset data corresponding to the previous frame is the lateral displacement data between the previous frame target position point determined for the moving object in the previous frame and the initial moving reference line used by the moving object at the initial position point, the previous frame target position point is the end point of the travel path planned for the moving object in the previous frame, and the target lateral offset data corresponding to the current frame is the lateral displacement data between the current frame target position point and the initial moving reference line used by the moving object at the initial position point, and the current frame target position point is the end point of the travel path planned for the moving object in the current frame.

[0006] The embodiment of the present application also provides another data processing method, including: obtaining final state lateral displacement data determined for a moving object in a current frame, the final state lateral displacement data being lateral displacement data of the moving object relative to an initial moving reference line used by the moving object at an initial position point in an expected final state, the expected final state being an expected state with a lateral safety space between the moving object and the object to be bypassed, and the initial position point being an initial position point of a path that the moving object needs to travel through in order to bypass the object to be bypassed; obtaining expected final state moving reference line offset data of the moving object based on the final state lateral displacement data, wherein the expected final state moving reference line offset data is lateral displacement data between an expected final state moving reference line and the initial moving reference line, and the expected final state moving reference line is a moving reference line used in the expected final state determined for the moving object in the current frame.

[0007] An embodiment of the present application also provides a path planning method for a mobile object, including: obtaining target lateral offset data determined for the mobile object in the current frame as the target lateral offset data corresponding to the current frame, the target lateral offset data corresponding to the current frame being the lateral displacement data between the target position point of the current frame and the initial moving reference line used by the mobile object at the initial position point, the current frame target position point being the end point of the travel path planned for the mobile object in the current frame, and the initial position point being the initial position point of the travel path that the mobile object needs to travel to bypass the bypassed object; according to the target lateral offset data corresponding to the current frame, obtaining the travel path data planned for the mobile object in the current frame.

[0008] An embodiment of the present application also provides another path planning method for a mobile object, comprising: obtaining final state lateral displacement data determined for the mobile object in a current frame, the final state lateral displacement data being lateral displacement data of the mobile object relative to an initial moving reference line used by the mobile object at an initial position point at an expected final state, the expected final state being an expected state with a lateral safety space between the mobile object and the object to be bypassed, and the initial position point being an initial position point of a path that the mobile object needs to travel through to bypass the object to be bypassed; obtaining expected final state moving reference line offset data of the mobile object based on the final state lateral displacement data, the expected final state moving reference line offset data being lateral displacement data between the expected final state moving reference line and the initial moving reference line, the expected final state moving reference line being a moving reference line used in the expected final state determined for the mobile object in the current frame; obtaining path data planned for the mobile object in the current frame based on the expected final state moving reference line offset data.

[0009] Compared with the prior art, the embodiments of the present application have the following advantages: In the data processing method provided by the embodiment of the present application, in the process of the mobile object bypassing the bypassed object, the final state lateral displacement data of the mobile object and the mobile state adjustment safety time data are calculated in each frame. The final state lateral displacement data is the lateral displacement data of the mobile object relative to the initial moving reference line in the expected final state, that is, the lateral displacement data between the position point of the mobile object and the initial moving reference line when the mobile object reaches the expected state of bypassing the bypassed object. The mobile state adjustment safety time data is the time data for the mobile object to perform the moving operation when it moves from the position of the current frame to the position of the expected final state. Therefore, the method can timely adjust the changes in the final state lateral displacement data of each frame and the changes in the mobile state adjustment safety time data. Further, according to the final state lateral displacement data corresponding to the current frame and the mobile state adjustment safety time data, the lateral offset step data of the current frame is obtained, and then combined with the target lateral offset data of the previous frame, the target lateral offset data corresponding to the current frame is obtained. Based on the change in the lateral offset step data of the current frame, the target lateral offset data corresponding to the current frame will also change. The target lateral displacement data corresponding to the current frame is the lateral displacement data between the target position point of the current frame and the initial moving reference line used by the mobile object at the initial position point, and the target position point of the current frame is the end point of the path planned for the mobile object in the current frame. In other words, the lateral displacement data of the end point of the path planned for the mobile object relative to the initial moving reference line will be calculated in each frame, that is, the displacement data of the end point of the path relative to the initial moving reference line in the lateral direction. Therefore, this method adjusts the target lateral displacement data of the end point position of the path planned for the mobile object frame by frame, so that the numerical change process between the target lateral displacement data corresponding to two adjacent frames is relatively smooth, thereby enabling the mobile object to smoothly bypass the bypassed object.

[0010] Another data processing method provided by the embodiment of the present application is that in the process of a mobile object bypassing an object, each frame combines the moving speed data of the mobile object in the current frame, and the corresponding relationship between the moving speed data range and the single frame offset step data range, to obtain the lateral offset step data corresponding to the current frame. In other words, if the current moving speed data of the mobile object in the current frame changes, the lateral offset step data corresponding to the current frame also changes, and the change value of the lateral offset step data can be adjusted in time according to the change of the moving speed data of the mobile object in each frame. Then, the lateral offset step data corresponding to the current frame is combined with the target lateral offset data corresponding to the previous frame to obtain the target lateral offset step data corresponding to the current frame. That is, in each frame, the target lateral offset data corresponding to the current frame will be adjusted in time according to the corresponding lateral offset step data, that is, the target lateral offset data of the terminal position of the path planned by the mobile object in the current frame is adjusted frame by frame, so that the numerical change process between the target lateral offset data corresponding to two adjacent frames is relatively smooth, so that the mobile object can smoothly bypass the bypassed object.

[0011] Another data processing method provided by the embodiment of the present application determines the final state lateral displacement data for the mobile object in each frame. The final state lateral displacement data is the lateral displacement data of the mobile object relative to the initial moving reference line used by the mobile object at the initial position point when the mobile object is in the expected final state. The expected final state is the expected state with a lateral safety space between the mobile object and the object to be bypassed, and the initial position point is the initial position point of the path that the mobile object needs to travel to bypass the object to be bypassed. According to the final state lateral displacement data, the expected final state moving reference line offset data of the mobile object is obtained, that is, the lateral displacement data between the expected final state moving reference line determined for the mobile object in the current frame and the initial moving reference line. In this process, the initial moving reference line is the moving reference line used by the mobile object before it starts to bypass the object to be bypassed, and the object to be bypassed is located at a position close to the initial moving reference line. Therefore, the final state lateral displacement data changes due to the change in the position of the object to be bypassed. If the object to be bypassed is offset in the lateral direction, the final state displacement data changes accordingly. Accordingly, the expected final state moving reference line offset data changes due to the change in the position of the object being circumvented. In addition, in the method provided in the embodiment of the present application, the moving object needs to move as close as possible to the position of the expected final state moving reference line, so that the moving object is as far away from the object being circumvented as possible and is not easily affected by the boundary jitter of the object being circumvented.

[0012] The path planning method for a mobile object provided by an embodiment of the present application calculates the target lateral offset data corresponding to the current frame in each frame, and obtains the path data planned for the mobile object in the current frame according to the target lateral offset data corresponding to the current frame. The target lateral offset data corresponding to the current frame is the lateral displacement data between the target position point of the current frame and the initial moving reference line used by the mobile object at the initial position point, and the target position point of the current frame is the end point of the path planned for the mobile object in the current frame, and the initial position point is the initial position point of the path that the mobile object needs to travel around the object being bypassed. In other words, each frame is a path that the mobile object needs to travel around the object being bypassed, and the target lateral offset data corresponding to the current frame is used as the lateral displacement data between the end point of the path and the initial moving reference line. By adjusting the target lateral offset data of the end point position of the path planned for the mobile object in the current frame frame by frame, the numerical change process between the target lateral offset data corresponding to the two adjacent frames is relatively gentle, so that the change trend between the paths planned for the mobile object frame by frame is also stable, so that the mobile object can smoothly bypass the object being bypassed.

[0013] The path planning method for a mobile object provided in an embodiment of the present application obtains the final state lateral displacement data determined for the mobile object in the current frame, obtains the expected final state movement reference line offset data of the mobile object based on the final state lateral displacement data, and obtains the path data planned for the mobile object in the current frame based on the expected final state movement reference line offset data. In this process, the final state lateral displacement data changes due to the change in the position of the object being bypassed. If the object being bypassed shifts in the lateral direction, the final state displacement data changes accordingly. Correspondingly, the expected final state movement reference line offset data changes due to the change in the position of the object being bypassed. Therefore, the final state lateral displacement data and the expected final state movement reference line offset data determined in each frame can be adjusted in time according to the change in the position of the object being bypassed. In addition, in the method provided in an embodiment of the present application, the mobile object needs to move as close to the expected final state movement reference line as possible, so that the mobile object is as far away from the object being bypassed as possible and is not easily affected by the boundary jitter of the object being bypassed. Therefore, the distance between the path data planned for the mobile object and the location of the bypassed object is greater than the preset distance threshold. In the process of the mobile object bypassing the bypassed object according to the path generated by the path data, the mobile object will also have a certain distance from the boundary of the bypassed object and will not be affected by the jitter of the boundary of the bypassed object or the change of position, thereby achieving the goal of the mobile object being able to smoothly bypass the bypassed object. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1AA schematic diagram of the path planning for an autonomous driving vehicle to bypass an obstacle on the left side of the obstacle provided in an embodiment of the present application.

[0015] Figure 1B A schematic diagram of the path planning for an autonomous driving vehicle to bypass an obstacle on the right side of the obstacle provided in an embodiment of the present application.

[0016] Figure 2A A schematic diagram of a travel path planned for a mobile object in a starting frame (frame 0) provided in an embodiment of the present application.

[0017] Figure 2B A schematic diagram of a travel path planned for a mobile object in a first frame provided in an embodiment of the present application.

[0018] Figure 2C A schematic diagram of a travel path planned for a mobile object in the n-2th frame provided in an embodiment of the present application.

[0019] Figure 2D A schematic diagram of a travel path planned for a mobile object in the n-1th frame provided in an embodiment of the present application.

[0020] Figure 3 A flowchart of a path planning method provided in an embodiment of the present application.

[0021] Figure 4 A flowchart of a data processing method provided for the first embodiment of the present application.

[0022] Figure 5 This is a schematic diagram of a data processing device provided in the second embodiment of the present application.

[0023] Figure 6 This is a flowchart of another data processing method provided in the third embodiment of the present application.

[0024] Figure 7 This is a schematic diagram of another data processing device provided in the fourth embodiment of the present application.

[0025] Figure 8 This is a flowchart of another data processing method provided in the fifth embodiment of the present application.

[0026] Fig. 9 This is a schematic diagram of another data processing device provided in the sixth embodiment of the present application.

[0027] Fig.10 This is a flowchart of a path planning method for a mobile object provided in the seventh embodiment of the present application.

[0028] Fig.11 This is a schematic diagram of a path planning device for a mobile object provided in the eighth embodiment of the present application.

[0029] Fig.12 This is a flowchart of another path planning method for a mobile object provided in the ninth embodiment of the present application.

[0030] Fig.13 This is a schematic diagram of another path planning device for a mobile object provided in the tenth embodiment of the present application. DETAILED DESCRIPTION

[0031] Many specific details are set forth in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below. The terms used in this application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The description methods used in this application and in the attached claims, such as "a", "first", and "second", etc., are not quantitative limitations or sequential limitations, but are used to distinguish the same type of information from each other. In order to solve the above-mentioned problems raised by the background technology, the present application provides multiple embodiments. The various embodiments provided by the present application are introduced below.

[0032] The following first describes the application scenario of the data processing method provided by the embodiment of the present application. The data processing method provided by the present application can be applied to the field of autonomous driving, where the moving object is an autonomous driving vehicle and the object to be bypassed is an obstacle. In the process where the autonomous driving vehicle needs to bypass an obstacle, the autonomous driving system determines the target lateral offset data corresponding to the current frame for the autonomous driving vehicle in the current frame. The data processing method provided by the present application can also be applied to the field of logistics and distribution, where the moving object is a rider and the object to be bypassed is an obstacle that appears in the rider's delivery path. The rider carries a device containing the data processing function. In the process of delivering goods to the delivery address, if an obstacle appears during the delivery process, in order to enable the rider to smoothly bypass the obstacle, the data processing method is used to determine the target lateral offset data corresponding to the current frame for the rider in each frame. Among them, the above-mentioned moving object can also be a robot with a delivery function, and the robot contains a system with the data processing function. In addition, the obstacle can be a static obstacle, such as a car parked on the roadside, or a dynamic obstacle, such as a car moving slowly in front of the road where the moving object is located, or an area in front of the road where the moving object is located that contains a pool of slowly moving water, or an area formed by a pool of oil formed by a truck with a leaking oil tank. The following specifically describes the application process of the method with the moving object being an autonomous driving vehicle and the object being bypassed being an obstacle.

[0033] The data processing method provided in the embodiment of the present application divides the process of the autonomous driving vehicle bypassing the obstacle into multiple stages for data processing, each stage is called a frame. In each frame, the final state lateral displacement data between the autonomous driving vehicle and the initial moving reference line at the expected final state and the expected available time data of the autonomous driving vehicle bypassing from the current frame position to the expected final state are calculated respectively, and the lateral displacement step data corresponding to the frame is obtained according to the final state lateral displacement data and the expected available time data. According to the lateral displacement step data corresponding to the frame, combined with the target lateral displacement data corresponding to the previous frame of the frame, the target lateral displacement data corresponding to the frame is obtained. According to the final state lateral displacement data, the expected final state lateral displacement parameter offset data between the moving reference line of the expected final state planned for the autonomous driving vehicle in the frame and the initial moving reference line is obtained. The expected final state lateral displacement parameter offset data is used as the value of the moving reference line lateral displacement parameter in the objective function. The target lateral displacement data corresponding to the frame is used as the value of the lateral displacement parameter of the final state position point in the objective function. The lateral movement parameter value of the position point in the path planned for the autonomous driving vehicle in the frame is obtained through the objective function. Based on the above process, the path data of the path planned by the autonomous driving vehicle in the frame is obtained, and then the path is generated. The path is the detour path around the obstacle planned by the autonomous driving vehicle in the frame. The objective function is used to calculate the cost function value corresponding to the path. In this process, the target lateral displacement data corresponding to the frame is the lateral displacement data of the end point of the path planned by the autonomous driving vehicle in the frame relative to the initial moving reference line. By adjusting the lateral displacement data corresponding to the end point of the path planned by the autonomous driving vehicle frame by frame, the numerical change process between the target lateral displacement data corresponding to two adjacent frames is relatively gentle. In addition, by determining the final state lateral displacement data frame by frame, the expected final state moving reference line offset data of each frame is determined as the value of the moving reference line lateral displacement parameter in the objective function. When the cost function value of the path planned by the frame is calculated frame by frame, the lateral displacement data of the end point of the path gradually increases, and the absolute value of the difference between the lateral displacement data of the end point of the path and the value of the moving reference line lateral displacement parameter gradually decreases, reducing the calculation amount of the objective function.

[0034] The following describes the detailed process of the data processing method provided by this application in combination with the application scenario.

[0035] 1. Dynamic final state offset strategy: The target lateral offset step data corresponding to each frame is calculated respectively, so that the lateral displacement data between the end point of the path planned for the moving object and the initial moving reference line is increased frame by frame. Finally, the target lateral offset data obtained in the last frame can reach the final state lateral displacement data of the expected final state. This process is the dynamic final state offset strategy.

[0036] The dynamic final state offset strategy is described in detail below: Step 1: The final state lateral displacement data determined for the moving object in the current frame is determined in the following manner: the obstacle that the autonomous driving vehicle needs to bypass is determined in the current frame, and based on the information of the obstacle, the final state lateral displacement data of the expected final state that the autonomous driving vehicle needs to bypass the obstacle is calculated, thereby determining the position of the autonomous driving vehicle at the expected final state.

[0037] Calculate the final state lateral displacement data corresponding to the expected final state of the autonomous driving vehicle according to the following formula 1 or formula 2: .

[0038] Go around the obstacle on the left: Formula 1 Go around the obstacle on the right: Formula 2 Among them, Formula 1 is the final state lateral displacement data of the autonomous driving vehicle when the autonomous driving vehicle bypasses the obstacle from the left side. . Represents the lateral displacement data between the expected final state of the autonomous driving vehicle and the original lane reference line used by the autonomous driving vehicle at the starting point of the obstacle detour, such as Figure 1A In . represents the lateral displacement data between the left boundary of the obstacle and the original lane reference line, such as Figure 1A In . Represents the half-body distance data of the autonomous vehicle. Represents the adjustable parameter between the right boundary of the autonomous vehicle and the left boundary of the obstacle. Formula 2 is the final state lateral displacement data of the autonomous vehicle when the autonomous vehicle bypasses the obstacle from the right side of the obstacle. . Represents the lateral displacement data between the expected final state of the autonomous driving vehicle and the original lane reference line used by the autonomous driving vehicle at the starting point of the detour around the obstacle. represents the lateral displacement data between the right boundary of the obstacle and the original lane reference line, such as Figure 1B In . Represents the half-body distance data of the autonomous vehicle. An adjustable parameter representing the distance between the left boundary of the autonomous vehicle and the right boundary of the obstacle.

[0039] Please refer to Figure 1A and Figure 1B , Figure 1AA schematic diagram of the path planning for an autonomous driving vehicle to bypass an obstacle on the left side of the obstacle provided in an embodiment of the present application. Figure 1B A schematic diagram of the path planning of an autonomous driving vehicle to bypass an obstacle on the right side of the obstacle provided in an embodiment of the present application. The vehicle body coordinate system is established with the starting position of the autonomous driving vehicle bypassing the obstacle as the origin, the tangent direction of the original lane reference line corresponding to the starting position of the autonomous driving vehicle bypassing the obstacle is the vertical axis, called the S axis, and the normal direction of the original lane reference line is the horizontal axis, called the L axis.

[0040] exist Figure 1A In the example, the autonomous vehicle bypasses the obstacle from the left, so the is a positive value, Add the half-body distance data of the autonomous driving vehicle and the sum of the adjustable parameters to obtain the final state lateral displacement data of the autonomous driving vehicle in the expected final state. Figure 1B In the example, the autonomous vehicle bypasses the obstacle from the right, so the If it is a negative value, The sum of the half-body distance data and the adjustable parameters of the autonomous driving vehicle is subtracted to obtain the final state lateral displacement data of the autonomous driving vehicle in the expected final state.

[0041] The above is the process of determining the final state lateral displacement data of the autonomous driving vehicle in the expected final state in the current frame. The following describes the process of obtaining the target lateral displacement data of the current frame: Step 2: Obtain the target lateral displacement data corresponding to the current frame. Obtain the longitudinal speed of the autonomous driving vehicle, the longitudinal distance between the autonomous driving vehicle and the obstacle, such as Figure 1A The distance difference between the starting position of the autonomous driving vehicle in the vehicle body coordinate system and the obstacle in the S-axis direction is used as the longitudinal distance between the autonomous driving vehicle and the obstacle. By dividing the longitudinal distance by the longitudinal driving speed, the expected available time data for adjusting the state of the autonomous driving vehicle during the process of the autonomous driving vehicle bypassing the obstacle is obtained. According to the final state lateral displacement data and the expected available time data, the lateral offset step data corresponding to the autonomous driving vehicle in the current frame is obtained. According to the lateral offset step data corresponding to the current frame, the target lateral offset data corresponding to the previous frame is adjusted to obtain the target lateral offset data corresponding to the current frame. In addition, based on the target lateral offset data corresponding to the current frame, the travel path data planned for the mobile object in the current frame is obtained. Specifically as follows: Please refer to FIG. 2A to FIG. 2D ,That Figure 2A A schematic diagram of a travel path planned for a mobile object in a starting frame (frame 0) provided in an embodiment of the present application. Figure 2B A schematic diagram of a travel path planned for a mobile object in a first frame provided in an embodiment of the present application. Figure 2CA schematic diagram of a travel path planned for a mobile object in the n-2th frame provided in an embodiment of the present application. Figure 2D A schematic diagram of a path planned for a mobile object in frame n-1 provided by an embodiment of the present application. FIG. 2A to FIG. 2D Describe the planned path of the moving object from the initial position to the expected final state in the process of bypassing the bypassed object and the actual offset process of the moving object. Here, the moving object takes an autonomous driving vehicle as an example, and the bypassed object takes a static obstacle as an example. The entire process of the autonomous driving vehicle bypassing the obstacle is carried out through multiple segmented steps, and each segment is a frame. In each frame, a path for bypassing the obstacle within a preset time is planned for the autonomous driving vehicle. The autonomous driving vehicle moves with reference to the path. When entering the next frame, a new path within the preset time is re-planned for the autonomous driving vehicle. And so on. After multiple frames of planned paths, the autonomous driving vehicle reaches the expected final state. The time of each frame is 0.1s, and the autonomous driving system re-plans a path for the autonomous driving vehicle every 0.1s.

[0042] Figure 2A It represents the path of the autonomous driving vehicle around the obstacle in the starting frame (frame 0). Specifically, the vehicle body coordinate system is established with the initial position of the autonomous driving vehicle around the obstacle as the origin, the longitudinal coordinate axis S is established with the tangent direction of the original lane reference line used by the autonomous driving vehicle, and the lateral coordinate axis L is established with the normal direction of the original lane reference line. The lateral displacement data or lateral offset data involved in the following description process refers to the lateral movement parameter value on the L axis.

[0043] The starting frame is a section of the path planned for the autonomous driving vehicle within a preset time (e.g., 7 seconds). According to the current longitudinal speed of the autonomous driving vehicle and the preset time, the longitudinal distance data of the autonomous driving vehicle in the longitudinal direction within the preset time is obtained. . The vertical distance data of the current frame At preset intervals Set multiple path location points, such as Figure 2A Contains S 0-0 , S 0-1 , S 0-2 , S 0-3 , S 0-4 , S 0-5 , S 0-6 , S 0-7 There are seven locations in total.

[0044] First, based on the target lateral offset data determined for the autonomous vehicle in the starting frame , obtain the lateral displacement data of the position point at the final state in the path planned for the mobile object in the starting frame, that is, the position point S of the mobile object at the n-1th path 0-7 The lateral movement parameter value is ,therefore, equal For example, the lateral offset step length data corresponding to the start frame is 0.07 meters, and the start frame has no previous frame. Therefore, the target lateral offset data corresponding to the start frame is is the lateral offset step data corresponding to the starting frame, 0.07 meters. The moving object is at the n-1th path position point S 0-7 The lateral movement parameter value is 0.07 meters.

[0045] Next, the lateral movement parameter value of the position point in the path is obtained through the objective function. Corresponding lateral movement parameter values ​​are assigned to other path position points, and the lateral movement parameter value includes at least one parameter value of lateral displacement value, lateral speed value, lateral acceleration value, and lateral jerk value.

[0046] For example, the autonomous vehicle clicks the car at the initial position of the starting frame. 起0 is the origin of the vehicle coordinate system, and its corresponding lateral displacement value is , which is 0, in S 0-1 The corresponding lateral displacement value is , in S 0-2 The corresponding lateral displacement value is , in S 0-3 The corresponding lateral displacement value is , in S 0-4 The corresponding lateral displacement value is , in S 0-5 The corresponding lateral displacement value is , in S 0-6 The corresponding lateral displacement value is Then, the first-order derivative (lateral velocity value) and second-order derivative (lateral acceleration value) are solved for the lateral displacement value assigned to each path position point.

[0047] The lateral displacement values, lateral velocity values, and lateral acceleration values ​​corresponding to the 8 path position points planned above are called a set of lateral movement parameter variables, such as X={ , ,……, , , ,……, , , ,……, }, input into the objective function to obtain the cost function value cost. In the above process of assigning lateral movement parameter values ​​to the path position points, the other 7 path position points can be assigned multiple lateral movement parameter values, for example, It can be 0.01, 0.02 or 0.03.

[0048] For each path position point, different lateral movement parameter values ​​are assigned, and multiple sets of lateral movement parameter variables are obtained. According to the objective function, the cost function values ​​corresponding to the multiple sets of lateral movement parameter variables are solved. According to the cost value, the target lateral movement parameter variable is obtained. According to the target lateral movement parameter value corresponding to each path position point, the starting frame is generated as the path data planned for the autonomous driving vehicle, thereby generating the path. The above is the description process of planning the path for the mobile object in the starting frame.

[0049] The autonomous vehicle moves with reference to the path planned in the starting frame (frame 0). When it moves to the next frame, that is, the first frame, the autonomous vehicle is replanned in the first frame. The path planned for the autonomous vehicle in the starting frame (frame 0) is not the path that the autonomous vehicle actually moves in that frame. The path that the autonomous vehicle actually moves on may deviate from the path planned in the starting frame. The autonomous vehicle moves a certain distance from the initial position point in frame 0 and arrives at the first frame. In the first frame, the starting position point of the autonomous vehicle is the vehicle. 起1 The actual position point reached after the vehicle actually moves a certain distance in frame 0. The position point may be a position point on the path planned by the autonomous vehicle in frame 0. If the autonomous vehicle does not move according to the planned path in frame 0, the starting position point of the autonomous vehicle is 起1 The lateral movement parameters can be other values, and are not necessarily on the path planned in the start frame. Figure 2B , which describes a new travel path planned for the autonomous driving vehicle in the first frame. In the first frame, a travel path within a preset time is planned for the autonomous driving vehicle, and the autonomous driving vehicle moves with reference to the travel path.

[0050] According to the current longitudinal speed of the autonomous driving vehicle and the preset duration, the longitudinal distance data of the current frame of the autonomous driving vehicle in the longitudinal direction within the preset duration is obtained. , the current frame longitudinal distance data At preset intervals Set multiple path location points, such as Figure 2B Contains S 1-0 , S 1-1 , S 1-2 , S 1-3 , S 1-4, S 1-5 , S 1-6 There are six position points in total. The target lateral offset data corresponding to the first frame is the sum of the lateral offset step data of the first frame and the target lateral offset data corresponding to the 0th frame. , obtain the lateral displacement data of the last state position point in the first frame of the moving object's planned path, that is, the moving object at the n-1th path position point S 1-6 The lateral movement parameter value is .and Figure 2A Similarly, in the first frame, the lateral movement parameter values ​​are assigned to other path position points to obtain a set of lateral state variables X = { , ,……, , , ,……, , , ,……, }, input into the objective function to obtain the cost function value cost. According to the cost value, S 1-0 , S 1-1 , S 1-2 , S 1-3 , S 1-4 , S 1-5 The corresponding lateral movement parameter value is used to generate the first frame of path data planned for the autonomous driving vehicle.

[0051] Accordingly, Figure 2C The path obtained in the n-2 frame is the path planned for the moving object in the n-2 frame, and the target lateral offset data determined in the n-2 frame is the lateral movement parameter value of the final state position point of the path. The specific process of generating the path is similar to the steps of the first frame. In the n-2 frame, the starting position point of the autonomous driving vehicle is the vehicle. 起（n-2） is the destination obtained by the autonomous driving vehicle after actually moving in the n-3 frame. Correspondingly, the autonomous driving vehicle reaches the n-1 frame after moving a certain distance in the n-2 frame. Among them, the n-1 frame is the way the moving object plans to travel and Figure 2A and Figure 2B Similarly, no further description is given here. In frame n-1, the autonomous driving vehicle moves for a period of time based on its planned path and reaches the actual end position where it can successfully bypass the object being bypassed, such as Figure 2D Car in 末（n-1） The actual end position may be different from the end position point of the expected end state determined in the frame (the vehicle 末目标 ), or there may be a certain deviation between the two, as long as the conditions for avoiding collision between the autonomous driving vehicle and obstacles can be met.

[0052] pass FIG. 2A to FIG. 2D As can be seen from the description, in the embodiment of the present application, the path for the mobile object to bypass the bypassed object is planned once for the mobile object in each frame. The lateral movement parameter value corresponding to the final state of the mobile object in the current frame is determined by the target lateral offset data corresponding to the current frame. Then, the lateral movement parameter values ​​of the mobile object at other path position points are obtained through the objective function, and the path data planned for the mobile object in this frame are obtained. The lateral movement parameter value of the final state position point of the path planned for the mobile object in each frame increases frame by frame until the last frame, when the lateral movement parameter value of the final state position point planned for the mobile object reaches the final state lateral displacement data of the mobile object at the expected final state. The path of the mobile object is planned frame by frame, taking into account the coordinate value of the starting position point of the mobile object in each frame, and the difference in the longitudinal distance data between the starting position point and the bypassed object, and also making the lateral movement parameter value of the final state position point planned frame by frame increase smoothly, so that the actual path of the mobile object to bypass the bypassed object is smooth, reducing the sudden change of the path, and realizing the smooth bypass of the mobile object by the bypassed object.

[0053] In addition, the present application can also obtain the target lateral offset data of the current frame in the following manner, specifically as follows: determine the driving speed data of the autonomous driving vehicle in the current frame, and the corresponding relationship between the driving speed data range and the single-frame lateral offset step data range; obtain the lateral offset step data determined by the autonomous driving vehicle in the current frame according to the driving speed data of the autonomous driving vehicle, and the corresponding relationship between the driving speed data range and the single-frame lateral offset step data range; obtain the target lateral offset data corresponding to the current frame according to the lateral offset step data corresponding to the current frame, combined with the target lateral offset data of the previous frame. The driving speed data range is 3m / s~8m / s, and the single-frame lateral offset step data range is 0.06m~0.1m.

[0054] Therefore, the corresponding relationship between the driving speed data range and the single-frame lateral offset step data range is a linear relationship, specifically, . in, represents the lateral offset step data determined for the moving object in the current frame; Indicates the speed data of the moving object in the current frame. In the current frame, if the speed data of the autonomous driving vehicle is less than 3m / s, it is calculated as 3m / s. If the speed data of the autonomous driving vehicle is greater than 8m / s, it is calculated as 8m / s. Obtain the target lateral offset data corresponding to the current frame according to the following formula 3: Formula 3 in, represents the target lateral offset data determined for the moving object in the current frame; represents the target lateral offset data determined for the moving object in the previous frame as the target lateral offset data corresponding to the previous frame; represents the lateral offset step data determined for the moving object in the current frame; Represents the final state lateral displacement data determined for the moving object in the current frame.

[0055] It can be seen from Formula 3 that if the sum of the target lateral offset data and the lateral offset step corresponding to the previous frame is less than or equal to the final state lateral offset data, then the target lateral offset data corresponding to the current frame is the target lateral offset data and the lateral offset step data corresponding to the previous frame; if the sum of the target lateral offset data and the lateral offset step data corresponding to the previous frame is greater than the final state lateral displacement data, then the target lateral offset data corresponding to the current frame is the final state lateral displacement data.

[0056] 2. Dynamic reference line strategy: In the embodiment of the present application, the final state lateral displacement data is determined for the autonomous driving vehicle in each frame, and the expected final state moving reference line offset data determined for the autonomous driving vehicle in the current frame is obtained based on the final state lateral displacement data. Based on the expected final state moving reference line offset data, the final state moving reference line of the autonomous driving vehicle in the expected final state is generated. Figure 1A In the example, the autonomous vehicle bypasses the obstacle from the left side and moves from the original lane to the overtaking lane.

[0057] Among them, in the n-1th frame, the final state lateral displacement data of the autonomous driving vehicle in the expected final state is determined as Among them, the final state lateral displacement data is the lateral displacement data of the autonomous driving vehicle at the expected final state relative to the original lane reference line used by the autonomous driving vehicle at the initial position point. Therefore, the expected final state moving reference line offset data corresponding to the autonomous driving vehicle is .

[0058] The final state moving reference line of the autonomous driving vehicle is Figure 1A The final state reference line in . Figure 1A What is described is the final state moving reference line determined for the autonomous driving vehicle in the n-1 frame. The method described in this application realizes that during the process of the autonomous driving vehicle circumventing an obstacle, the final state moving reference line is determined for the autonomous driving vehicle in each frame.

[0059] Compared with the prior art, in which the original lane reference line is used, in the embodiment of the present application, after the dynamic reference line is obtained in each frame, the target lateral offset corresponding to each frame is calculated based on the offset of the moving reference line in the final state. The difference between them is reduced, thereby reducing the amount of calculation of the cost function value of the candidate path for the autonomous driving vehicle to bypass the obstacle. Wherein, Formula 4 is the objective function: Formula 4 Wherein, cost represents the cost function value, and in the embodiment of the present application, the lateral movement parameter value of the position point in the path planned for the moving object in the current frame is obtained through the objective function; Represents the lateral displacement data of each path position point in the path, Represents the first-order derivative of the lateral displacement data, lateral velocity data, Represents the second-order derivative of the lateral displacement data, the lateral acceleration data, jerk represents the rate of change of the lateral acceleration data over time, and the lateral jerk data is obtained by differential calculation of the lateral acceleration data: , is a constant; Represents the weight value of the lateral displacement data; Indicates the weight value of the lateral velocity data; Indicates the weight value of lateral acceleration data; Indicates the weight value of lateral acceleration; Represents the lateral displacement data of the moving reference line in the final state; The weight value represents the difference between the lateral displacement data of each path position point in the path and the lateral displacement data of the final state moving reference line; in order to increase the weight of the lateral displacement data, lateral velocity data, and lateral acceleration data of the n-1th position, Additional penalty factor ,right Additional penalty factor ,right Additional penalty factor .

[0060] (1) Boundary constraints:

[0061]

[0062]

[0063]

[0064] in, Indicates the lower boundary of the lateral displacement data of each path position point, such as Figure 1A The lower boundary of , which represents the minimum value of the lateral displacement data; Indicates the upper boundary of the lateral displacement data of each path position point, such as Figure 1AThe upper boundary in represents the maximum value of the lateral displacement data; in other words, the lateral displacement data of each path position point takes a value within the range of the lower boundary of the lateral displacement data and the upper boundary of the lateral displacement data. The lower boundary of the lateral velocity data of each path position point represents the minimum value of the lateral velocity data; Indicates the upper boundary of the lateral velocity data of each path position point, which indicates the maximum value of the lateral velocity data; that is, the lateral velocity data of each path position point takes a value within the range of the lower boundary of the lateral velocity data and the upper boundary of the lateral velocity data. The lower boundary of the lateral acceleration data of each path position point represents the minimum value of the lateral acceleration data; Indicates the upper boundary of the lateral acceleration data of each path position point, which indicates the maximum value of the lateral acceleration data; in other words, the lateral acceleration data of each path position point takes a value within the range of the lower boundary of the lateral acceleration data and the upper boundary of the lateral acceleration data. Indicates the lower boundary of the acceleration difference data between two adjacent path position points. Indicates the upper boundary of the acceleration difference data between two adjacent path position points; in other words, the acceleration difference between two adjacent path position points is within the range of the lower boundary of the acceleration difference data and the upper boundary of the acceleration difference data.

[0065] (2) Continuity constraints: These include velocity continuity constraints and displacement continuity constraints.

[0066] Formula 5 Wherein, Formula 5 represents the velocity continuity constraint relationship between the lateral velocity data of two adjacent path position points. Represents the lateral velocity data of the i+1th path position point, Represents the lateral velocity data of the i-th path position point; Represents the lateral acceleration data of the i-th path position point, Represents the lateral acceleration data of the i+1th path position point; Indicates that autonomous vehicles arrive The time interval is short (for example, 0.1s). Therefore, Formula 5 represents The difference in lateral speed data change between two adjacent path position points within the time interval is less than a preset speed difference threshold.

[0067] Formula 6 Wherein, Formula 6 represents the displacement continuity constraint relationship between the lateral displacement data of two adjacent path position points. Represents the lateral displacement data of the i+1th path position point, Represents the lateral displacement data of the i-th path position point; represents the lateral velocity data of the i-th path position point, Represents the lateral acceleration data of the i-th path position point, Represents the lateral acceleration data of the i+1th path position point; Indicates that autonomous vehicles arrive The time interval is short (for example, 0.1s). Therefore, Formula 6 represents The difference in lateral displacement data change between two adjacent path position points within the time interval is less than a preset displacement difference threshold.

[0068] In summary, Formula 5 and Formula 6 indicate that the difference in lateral speed data change between two adjacent path position points in the target path is less than the preset speed difference threshold, and the difference in lateral displacement data change is less than the preset displacement difference threshold. Therefore, the path curve of the target path is a smooth curve, and jumps or mutations are avoided between two adjacent path position points.

[0069] (3) Initial condition constraints: ; ; In the initial condition constraints here, Represents the lateral displacement data between the position of the autonomous driving vehicle in the current frame and the original lane reference line. Correspondingly, Indicates that the current speed data of the autonomous driving vehicle in the current frame is the initial set speed data, Indicates that the current acceleration data of the autonomous driving vehicle in the current frame is the initial set acceleration data.

[0070] The above is the process of determining the expected final state moving reference line offset data for the autonomous driving vehicle in the current frame, and generating the moving reference line of the expected final state of the moving object determined in the current frame according to the expected final state moving reference line offset data. Then, the process of obtaining the path data planned for the moving object in the current frame is carried out through the objective function.

[0071] Please refer to Figure 3 , which is a flow chart of the path planning method provided in an embodiment of the present application. S301: Define path planning state variables, objective functions, and constraints. The path planning state variable obtained in this step is the above-mentioned lateral state variable X, and its acquisition method is as described above, which will not be repeated here. The objective function is the cost function value described in formula 4, and the constraints include boundary constraints, continuity constraints, and initial condition constraints.

[0072] S302: Dynamic final state offset. The autonomous driving vehicle predetermines the target lateral offset data in each frame. , the target lateral offset data of the current frame is used as the lateral movement parameter value of the final state in the path of the objective function .

[0073] The autonomous driving vehicle refers to the path planned for each frame, moves a certain distance, enters the next frame, and then repeats the above operation. The lateral movement parameter value of the autonomous driving vehicle in the final state is increased frame by frame, and finally reaches the expected final state position to complete the process of bypassing the obstacle. In this process, the final state lateral displacement data, expected available time data, and lateral offset step data are calculated for each frame, and finally the target lateral offset data corresponding to the frame is obtained. The target lateral offset data corresponding to the frame is increased frame by frame, so that each frame obtains a smooth path, thereby allowing the autonomous driving vehicle to bypass obstacles smoothly.

[0074] S303: Generate a dynamic target reference line. This step is to obtain the expected final state moving reference line offset data determined for the autonomous driving vehicle in the current frame according to the final state lateral displacement data. According to the expected final state moving reference line offset data, obtain the moving reference line used in the expected final state determined for the autonomous driving vehicle in the current frame, that is, the dynamic target reference line, such as Figure 1A The final state reference line in the process. In each frame, the expected final state moving reference line offset data corresponding to the frame is calculated to obtain the moving reference line used in the expected final state corresponding to the frame. As a result, when the autonomous driving vehicle is bypassing obstacles, its moving direction gradually approaches the final state reference line and moves away from the original lane reference line, avoiding collision between the autonomous driving vehicle and the bypassed object. After the autonomous driving vehicle reaches the expected final state position from the starting position, the process of bypassing the obstacle is completed. Figure 1A Self-driving vehicles in China 起 Location Arrival Car 末 In addition, setting a dynamic target reference line can also have the following effects: the autonomous driving vehicle 末 When the position returns to the original lane reference line, the method provided by the present application can also be used to return to the original lane reference line by adjusting the target lateral offset data frame by frame. In the process of returning to the original vehicle to the reference line, the expected final state moving reference line offset data is 0, and the expected final state moving reference line is the original lane reference line. Moreover, before returning to the original vehicle to the reference line, the automatic driving vehicle will also adjust the target lateral offset data frame by frame. 末 Whether there are other obstacles on the lane where the position is located determines when to return to the original lane reference line.

[0075] S304: Initialize and solve the quadratic programming solver. QP (Quadratic Programming): Quadratic programming is a special type of mathematical optimization problem. It involves the minimization problem of a quadratic function, which is constrained by linear equations and inequalities. This type of problem has a wide range of applications in engineering, economics, and operations research. Quadratic programming can only converge to the optimal solution when the cost function is a convex function, so this requires the P matrix to be a semi-positive matrix, so the solution space of the path planning algorithm is a convex space, so that the quadratic programming can converge to an optimal path. In this step, in the current frame, the target lateral offset data is used as the lateral movement parameter value of the final state position point in the objective function, and the expected final state moving reference line offset data is used as the lateral movement parameter value of the final state moving reference line in the objective function. The lateral movement parameter value of the position point in the path planned for the mobile object in the current frame is obtained through the objective function of formula 4, and then the path data planned for the mobile object in the current frame is generated.

[0076] S305: Obtaining an optimized path; the optimized path in this step is generated based on the path data obtained by calculating the above objective function. Figure 1A As shown in , when the autonomous driving vehicle detours from the original lane to the overtaking lane, it is the path planned for the autonomous driving vehicle to bypass the obstacle. S306: Return the path. This step is after the lateral movement parameters of the position points of the path planned for the current frame are assigned. When the target cost function value is not obtained in the objective function, it means that the path data planned for the mobile object in the current frame is not generated, and the path of the current frame is not obtained.

[0077] First embodiment Figure 4 This is a flow chart of a data processing method provided in the first embodiment of the present application. Figure 4 The data processing method provided in the first embodiment of the present application is described in detail. Figure 4The data processing method shown includes: step S401 to step S404. Among them, the data processing method provided by the first embodiment corresponds to the above-mentioned scenario embodiment, and the specific details can be referred to the scenario embodiment, which will not be repeated here. The first embodiment of the present application is to determine the target lateral offset data corresponding to the current frame for the mobile object in the current frame. Specifically, the process of the mobile object bypassing the bypassed object is divided into multiple stages and performed step by step, and one stage is called a frame. In each frame, the target lateral offset data corresponding to the current frame is determined for the mobile object, that is, the lateral displacement data of the target position point that can be reached in the process of the mobile object bypassing the bypassed object in the current frame relative to the initial moving reference line. By determining the target lateral offset data corresponding to the frame frame by frame, the target lateral displacement data corresponding to the frame by frame changes relatively slowly, so that the mobile object can smoothly bypass the bypassed object. Among them, the mobile object can be an autonomous driving vehicle, a robot, an intelligent device used by a rider, etc., and the bypassed object can be an obstacle. The obstacle can be divided into a static obstacle and a dynamic obstacle. The dynamic obstacle can be other vehicles moving slowly in front of the autonomous driving vehicle.

[0078] Step S401: Acquire final state lateral displacement data determined for the mobile object in the current frame, wherein the final state lateral displacement data is the lateral displacement data of the mobile object at the expected final state relative to the initial moving reference line used by the mobile object at the initial position point, the expected final state is the expected state that the mobile object needs to reach in order to bypass the bypassed object with a lateral safety space between the mobile object and the bypassed object, and the initial position point is the initial position point of the path that the mobile object needs to travel in order to bypass the bypassed object.

[0079] This step is used to determine the lateral displacement data between the position of the moving object and the object being moved when the moving object reaches the expected final state in the current frame, which is called the final state lateral displacement data. The lateral displacement data refers to the displacement data between the moving object and the initial moving reference line in the lateral direction, with the tangent direction of the initial moving reference line used by the moving object as the longitudinal direction and the normal direction of the initial moving reference line as the lateral direction. The lateral direction and the longitudinal direction are perpendicular to each other. Figure 1A The position of the autonomous vehicle at the expected final state 末The lateral displacement data between the mobile object and the object being bypassed. According to the obtained final state lateral displacement data, combined with the mobile object obtained in step S402, the lateral offset step data corresponding to the mobile object in the current frame is determined, and then the target lateral offset data corresponding to the current frame is obtained. The expected final state indicates that the mobile object has reached a state of successfully bypassing the object being bypassed. The lateral safety space between the mobile object and the object being bypassed specifically means that when the mobile object bypasses to the expected final state, the lateral distance between it and the object being bypassed can avoid the collision between the mobile object and the object being bypassed. For example, when the autonomous driving vehicle bypasses to the expected final state, it is in a state parallel to the obstacle. For example, the obstacle is a stationary object, the body length of the autonomous driving vehicle is greater than the longitudinal length of the obstacle, and the autonomous driving vehicle bypasses the obstacle from the left side of the obstacle. When the bypass is parallel to the obstacle, the autonomous driving vehicle reaches the expected final state, that is, the state of successfully bypassing the obstacle has been reached.

[0080] For another example, the position of the autonomous driving vehicle when it detours to the expected final state is a state where it exceeds half of the longitudinal length of the obstacle. For example, the obstacle is a truck with a body length of 11 meters, and the autonomous driving vehicle is a car with a body length of 4 meters. The car detours the truck from the left side of the truck starting from the center line of the original lane, detours to the center line position of the overtaking lane on the left side of the original lane (the moving reference line of the expected final state), and the front of the car is located at half the length of the longitudinal body length of the truck. At this time, the lateral displacement data between the car and the truck is the lateral displacement data of the final state corresponding to the expected final state, but the car still needs to maintain the lateral distance with the truck and continue to move forward for a distance until the rear of the car is in front of the front of the truck to complete the process of detour the truck. The lateral safety space refers to the lateral distance between the position of the moving object when it reaches the expected final state and the object being detoured, which can avoid collision between the moving object and the object being detoured. Among them, the process of obtaining the final state lateral displacement data in this step can be implemented in the following manner: the obtaining of the final state lateral displacement data determined for the moving object in the current frame includes: obtaining the lateral displacement data of the wound object, the lateral displacement data of the wound object is the lateral displacement data between the wound side boundary of the wound object and the initial moving reference line; obtaining the lateral buffer data of the moving object in the expected final state, the lateral buffer data is used to represent the lateral safety space between the moving object in the expected final state and the wound object; obtaining the final state lateral displacement data according to the lateral displacement data of the wound object and the lateral buffer data.

[0081] The initial moving reference line is the reference line used by the moving object to move in the original lane before it bypasses the bypassed object, such as Figure 1AThe lateral displacement data between the bypassed side boundary of the bypassed object and the initial moving reference line, such as Figure 1A In This step determines the final state lateral displacement data, that is, determines the lateral displacement data between the position of the moving object and the initial moving reference line when the moving object reaches the expected final state. In order to improve the accuracy of the lateral displacement data between the position of the moving object in the expected final state and the initial moving reference line, if the lateral width data of the wound object itself is large, it is necessary to obtain the lateral displacement data between the wound side boundary of the wound object and the initial moving reference line.

[0082] The lateral buffer data is used to represent the lateral safety space between the moving object in the expected final state and the object being circumvented, such as Figure 1A The purpose of setting the horizontal line buffer data is to prevent the moving object from colliding with the object being circumvented when it circumvents the object. Figure 1A As shown in the figure, if the distance between the left boundary of the obstacle and the left boundary of the original lane is less than the first preset distance threshold, and the distance between the right boundary of the autonomous driving vehicle at the expected final state and the right boundary of the overtaking lane (that is, the left boundary of the original lane) is less than the second preset distance threshold, the autonomous driving vehicle may collide with the obstacle at the expected final state. Therefore, it is necessary to set lateral buffer data to ensure that the autonomous driving vehicle can drive safely and bypass the obstacle at the expected final state.

[0083] In addition, if the lateral width data of the moving object itself is too large, the lateral width data of the moving object also needs to be considered when calculating the final state lateral displacement data. Specifically, the lateral width data of the moving object is obtained; the final state lateral displacement data is obtained according to the lateral displacement data of the bypassed object and the lateral buffer data, including: obtaining the final state lateral displacement data according to the lateral displacement data of the bypassed object, the lateral buffer data and the lateral width data. Because the final state lateral displacement data is the lateral displacement data of the position of the moving object relative to the initial moving reference line when the moving object is in the expected state with a lateral safety space between it and the bypassed object. Therefore, the calculation of the final state lateral displacement data refers to the lateral width data of the moving object, which can avoid the lateral distance of the lateral safety space occupied by the lateral width dimension of the moving object. One application of the final state lateral displacement data is to obtain the final state moving reference line, such as Figure 1A As shown in the figure, the moving object is an autonomous driving vehicle, and the autonomous driving vehicle drives to the vehicle 末When the vehicle is in the final state, it drives along the final state reference line, which is located at the center line of the overtaking lane. Therefore, the calculation of the lateral displacement data at the end of the year needs to consider the lateral displacement data corresponding to the left boundary of the obstacle, the lateral buffer data, and the lateral width data of the moving object.

[0084] The step of obtaining the final state lateral displacement data according to the lateral displacement data, the lateral buffer data and the lateral width data of the object being bypassed comprises: obtaining the sum of the lateral displacement data of the object being bypassed, the lateral buffer data and the preset ratio of the lateral width data as the final state lateral displacement data. The preset ratio of the lateral width data of the moving object may be, for example, half the width data of the horizontal line width data of the moving object. The sum of the lateral displacement data, the lateral buffer data and the preset ratio of the lateral width data of the object being bypassed may be obtained by referring to Formula 1. The moving object bypasses the object from the left side of the object being bypassed, and the left side boundary of the object being bypassed is located on the left side of the initial moving reference line. It may be stipulated that the displacement data located on the left side of the initial moving reference line is a positive value, and the lateral buffer data and the lateral width data are actual values ​​without direction. Therefore, the final state lateral displacement data is calculated as the sum of the lateral displacement data of the object being bypassed, the lateral buffer data and the preset ratio of the lateral width data. In addition, it may be stipulated that the displacement data located on the right side of the initial moving reference line is a positive value, the moving object bypasses the object from the right side of the object being bypassed, and the bypassed side boundary of the object being bypassed is the right side boundary located on the right side of the initial moving reference line. The final state lateral displacement data or the lateral displacement data of the object being wound is calculated, and the sum of the lateral buffer data and the preset ratio of the lateral width data is calculated.

[0085] The above process describes a specific method for determining the final state lateral displacement data for the moving object in the current frame. In the embodiment of the present application, the final state lateral displacement data will be determined for each frame. Because the final state lateral displacement data is related to the wound side boundary of the wound object, if the wound side boundary of the wound object moves in the lateral direction between two adjacent frames, the final state lateral displacement data will also be affected. Specifically as follows: determine whether the wound side boundary of the wound object has undergone a lateral shift in the current frame relative to the previous frame; if the wound side boundary of the wound object has undergone a lateral shift in the current frame relative to the previous frame, determine whether the lateral shift data of the wound side boundary of the wound object in the current frame relative to the previous frame exceeds a preset boundary lateral shift data threshold; if the lateral shift data of the wound side boundary of the wound object in the current frame relative to the previous frame does not exceed the preset boundary lateral shift data threshold, use the final state lateral displacement data determined for the moving object in the previous frame as the final state lateral displacement data determined for the moving object in the current frame; the obtaining of the lateral displacement data of the wound object includes: if the lateral shift data of the wound side boundary of the wound object in the current frame relative to the previous frame exceeds the preset boundary lateral shift data threshold, then obtain the lateral displacement data of the wound object.

[0086] Continue with Figure 1A Take an example to illustrate, if the lateral displacement data corresponding to the left boundary of the obstacle in the previous frame is 4, and the lateral displacement data of the left boundary of the obstacle in the current frame is 4.8, that is, the lateral displacement data of the obstacle in the current frame relative to the previous frame in the lateral direction to the left is 0.8. If the preset boundary lateral displacement data threshold is 0.5, and 0.8 is greater than 0.5, it is necessary to obtain the lateral displacement data corresponding to the left boundary of the obstacle in the current frame, that is, to obtain the lateral displacement data of the object being bypassed. In this way, the accuracy of the final state lateral displacement data determined for the mobile object in the current frame is improved. For another example, the obstacle can be an object whose boundary changes at any time. For example, there is an obstacle in front of the original road where the mobile object is located, which is an area formed by a pool of slowly flowing water, or the obstacle is an area formed by a pool of oil formed by a truck with a leaking oil tank. The mobile object intends to bypass this area to travel. In this area, water or oil flows at any time, increasing the area of ​​the area, and the boundary of the area will also change. In this case, it is necessary to calculate the value of the lateral offset of the boundary of the obstacle in each frame. If the lateral offset data exceeds the preset boundary lateral offset data threshold, it is necessary to recalculate the lateral displacement data of the bypassed side boundary of the obstacle in the current frame.

[0087] Step S402: Obtain the mobile state adjustment safety duration data determined for the mobile object in the current frame, wherein the mobile state adjustment safety duration data is the expected available duration data for the mobile object to adjust its mobile state to avoid collision with the object being surrounded, starting from the current frame. This step is used to obtain the mobile state adjustment safety duration data, that is, the duration data for the mobile object to adjust its mobile state from the position of the current frame before the mobile object collides with the object being surrounded. The mobile object adjusting its mobile state includes the process of the mobile object moving in the longitudinal direction and the transverse direction respectively.

[0088] Wherein, the method for obtaining the mobile state adjustment safety duration data includes: the obtaining of the mobile state adjustment safety duration data determined for the mobile object in the current frame includes: obtaining the longitudinal movement speed data of the mobile object in the current frame; obtaining the longitudinal distance data between the mobile object in the current frame and the object being surrounded; obtaining the mobile state adjustment safety duration data determined for the mobile object in the current frame according to the longitudinal distance data and the longitudinal movement speed data. The longitudinal movement speed of the mobile object in the current frame may vary according to the different motion states of the object being surrounded. If the object being surrounded is a static obstacle, the longitudinal movement speed of the mobile object in the current frame is the longitudinal speed data in the driving speed data of the mobile object in the current frame. If the object being surrounded is a dynamic obstacle, the longitudinal movement speed data of the mobile object in the current frame is the difference between the longitudinal driving speed data of the mobile object and the current movement speed data of the obstacle.

[0089] The longitudinal distance data between the moving object in the current frame and the object being orbited, specifically, the longitudinal distance data between the position of the moving object in the current frame and the position of the object being orbited. Here, a global coordinate system can be established with the position of the moving object in the current frame as the origin, the coordinate point of the object being orbited in the global coordinate system can be obtained, and the distance data between the moving object and the object being orbited can be divided into horizontal distance data and longitudinal distance data. What is needed here is the longitudinal distance data between the moving object and the object being orbited. Figure 1A In the current frame, the position of the autonomous driving vehicle is at the origin of the vehicle coordinate system, and the obstacle is S in the S-axis direction. 障 Position, calculate the data difference between the two positions on the S axis, that is, the longitudinal distance data between the moving object and the object being circumvented in the current frame. This step is used to obtain the target lateral offset step length corresponding to the current frame in combination with step S401.

[0090] Step S403: Obtain the lateral offset step data corresponding to the current frame according to the final state lateral displacement data and the moving state adjustment safety time data. This step is used to obtain the lateral offset step data corresponding to the current frame, which can be specifically divided by the final state lateral displacement data by the moving state adjustment safety time data to obtain the lateral offset step data that the moving object can offset in the current frame. For example, the final state lateral displacement data is 3 meters, and the moving state adjustment safety time data is 5s, where each frame is 0.1s, and the lateral offset step data corresponding to the current frame is 0.06 meters.

[0091] Step S404: According to the lateral offset step data corresponding to the current frame, the target lateral offset data corresponding to the previous frame is adjusted to obtain the target lateral offset data determined for the mobile object in the current frame as the target lateral offset data corresponding to the current frame, the target lateral offset data corresponding to the previous frame is the lateral displacement data between the target position point of the previous frame determined for the mobile object in the previous frame and the initial moving reference line used by the mobile object at the initial position point, the target position point of the previous frame is the end point of the travel path planned for the mobile object in the previous frame, the target lateral offset data corresponding to the current frame is the lateral displacement data between the target position point of the current frame and the initial moving reference line used by the mobile object at the initial position point, the target position point of the current frame is the end point of the travel path planned for the mobile object in the current frame. This step is used to obtain the target lateral offset data determined for the mobile object in the current frame, that is, the lateral offset data of the end point of the travel path planned for the mobile object to bypass the bypassed object in the current frame that can be offset relative to the initial moving reference line. The target lateral offset data corresponding to the current frame is the sum of the lateral offset data corresponding to the previous frame and the lateral offset step data of the current frame. Based on the change in the lateral offset step data of the current frame, the target lateral offset data corresponding to the current frame will also change. In other words, the lateral displacement data of the end point of the path planned for the mobile object relative to the initial moving reference line will be calculated in each frame, that is, the displacement data of the end point of the path relative to the initial moving reference line in the lateral direction. Therefore, this method adjusts the target lateral offset data of the end point position of the path planned for the mobile object frame by frame, so that the numerical change process between the target lateral offset data corresponding to two adjacent frames is relatively smooth, thereby enabling the mobile object to smoothly bypass the bypassed object.

[0092] In addition, if the current frame is the starting frame, the method further includes: if the current frame is the starting frame for the moving object to bypass the bypassed object, the lateral offset step data corresponding to the current frame is used as the target lateral offset data determined for the moving object in the current frame. Figure 2A, the current frame is the starting frame, and there is no previous frame in the starting frame, then the lateral offset step data corresponding to the current frame is the target lateral displacement data corresponding to the current frame. After obtaining the target lateral offset data corresponding to the current frame in step S404, it can also include: according to the target lateral offset data corresponding to the current frame, obtaining the travel path data planned for the mobile object in the current frame. According to the target lateral offset data corresponding to the current frame, obtaining the travel path data planned for the mobile object in the current frame. The travel path planned for the mobile object in the current frame is the travel path for the mobile object to start moving for a preset time at the position of the current frame. Because the target lateral offset data corresponding to the current frame is the lateral displacement data between the end point of the travel path planned for the mobile object in the current frame and the initial movement reference line used by the mobile object. Therefore, based on the change of the target lateral offset data determined for the mobile object in each frame, the lateral displacement data between the end point of the corresponding travel path and the initial movement reference line will also change, and accordingly, the lateral offset data of other path position points of the travel path planned for the mobile object in each frame will also change. By planning the travel path for the mobile object frame by frame and gradually adjusting the lateral displacement data between the end point of the travel path and the initial moving reference line, in the last frame, the lateral displacement data corresponding to the end point of the travel path planned for the mobile object may be equal to the final state lateral displacement data of the mobile object in the expected final state, so as to complete the process of the mobile object bypassing the bypassed object. The above method plans the travel path frame by frame, and adjusts the lateral displacement data between the end point of the travel path and the initial moving reference line frame by frame, that is, increases the lateral displacement data of the mobile object in the final state frame by frame. This adjustment method makes the change value of the lateral displacement data of the final state obtained by the mobile object in each frame relatively small, and the moving state of the mobile object is relatively stable, thereby realizing that the mobile object smoothly bypasses the bypassed object.

[0093] like FIG. 2A to FIG. 2DThe description of the invention states that the autonomous driving vehicle starts from the starting frame (the 0th frame) to the N-1th frame, and each frame plans a path for the autonomous driving vehicle. The lateral offset data corresponding to the end point of the path between two adjacent frames gradually increases, and finally the lateral offset data corresponding to the end point of the path of the N-1th frame is equal to the final state lateral displacement data of the moving object in the expected final state, thereby realizing that the autonomous driving vehicle smoothly bypasses the obstacle. Among them, according to the target lateral offset data corresponding to the current frame, the specific steps of obtaining the path data planned for the moving object in the current frame are as follows: according to the target lateral offset data corresponding to the current frame, obtaining the path data planned for the moving object in the current frame includes: using the target lateral offset data corresponding to the current frame as the value of the final state lateral displacement parameter in the objective function, and obtaining the lateral movement parameter value of the position point in the path planned for the moving object in the current frame through the objective function, the final state lateral displacement parameter is used to represent the lateral displacement between the final state position point in the path and the initial moving reference line, and the objective function is used to calculate the cost function value corresponding to the path. The target lateral offset data corresponding to the current frame is used as the value of the final state lateral displacement parameter in the objective function.

[0094] According to the above formula 3, the target lateral offset data corresponding to the current frame is , which is used as the reference value of the final state lateral displacement in the objective function, such as the cost function in formula 4 .

[0095] After determining the value of the final state lateral displacement parameter in the objective function, the lateral movement parameter values ​​of other position points in the travel path are obtained through the objective function. Specifically, the lateral movement parameters of other position points are assigned, and then the cost function value is obtained. According to the cost function value, it is determined whether the assigned data of the lateral movement parameters of other position points is appropriate data. If appropriate, the data is used as the lateral movement parameter value of other position points, thereby obtaining the travel path data. Among them, after assigning the horizontal line movement parameters of other position points, the calculated cost function value is located at the target cost function value, which means that the assigned data is appropriate, the target cost function value is less than other cost function values, and the lateral movement parameter values ​​of other position points meet the constraints, and the constraints are the constraints mentioned in the scenario embodiment. The calculated cost function value obtained by the objective function is used to obtain the lateral movement parameter values ​​at the position points in the travel path, thereby obtaining the travel path data, and then generating the travel path according to the travel path data. This process increases the value of the final state lateral displacement parameter of the moving object in the current frame frame by frame, so that the value of the final state lateral displacement parameter increases gradually and smoothly with each frame, and finally makes the generated path a smooth path, so that the moving object can smoothly bypass the bypassed object. In addition, the moving reference line in the above objective function can be an initial moving reference line or an expected final state moving reference line, which is not limited here.

[0096] In addition, in order to increase the accuracy of determining the travel path data based on the cost function value, an embodiment of the present application also includes adding a weight value of the lateral movement parameter value corresponding to the final state position point of the travel path, specifically including: obtaining the weight value of the final state lateral displacement parameter; obtaining the lateral movement parameter value of the position point in the travel path planned for the mobile object in the current frame through the objective function, including: using the weight value of the final state lateral displacement parameter as the weight value of the lateral movement parameter value of the final state lateral displacement parameter in the objective function, and obtaining the lateral movement parameter value of the position point in the travel path planned for the mobile object in the current frame, wherein the weight value of the final state lateral displacement parameter is different from the weight value of the lateral movement parameter of the non-final state position point, and the non-final state position point is other position points other than the end point in the travel path planned for the mobile object in the current frame.

[0097] The obtaining of the weight value of the final state lateral displacement parameter includes: obtaining the weight value of the lateral displacement parameter value of the non-final state position point in the travel path planned for the mobile object in the current frame, adding the weight value of the lateral displacement parameter value of the non-final state position point by the final state weight value for the final state position point, obtaining the weight value of the lateral displacement parameter value of the final state position point as the weight value of the final state lateral displacement parameter. The weight value of the final state lateral displacement parameter is used as the weight value of the lateral displacement parameter value of the final state lateral displacement parameter in the objective function, that is, the weight value of the final state lateral displacement parameter is used as the weight value of the lateral displacement parameter value of the final state position point in the objective function.

[0098] Combining formula 4, it can be seen that when calculating the cost, the weight values ​​of the lateral movement parameters of the non-final state position points of the path are: The weight value is , The weight value is , The weight value is The weight value is added to the lateral movement parameter value of the final state position point of the path. Therefore, the weight value of the lateral movement parameter value of the final state position point is: The weight value is , The weight value is , The weight value is .

[0099] The above description obtains the target lateral displacement data corresponding to the current frame of the moving object, and uses the target lateral displacement data corresponding to the current frame as the lateral displacement data of the end point of the travel path relative to the initial moving reference line. In this process, the lateral displacement data of the end point of the travel path of the moving object relative to the initial moving reference line is adjusted frame by frame, so that the lateral displacement data corresponding to the end point of the travel path of two adjacent frames has a relatively smooth change trend, thereby realizing that the moving object smoothly bypasses the bypassed object.

[0100] Wherein, after step S401 obtains the final state lateral displacement data, it also includes: obtaining the expected final state moving reference line offset data of the moving object according to the final state lateral displacement data; wherein the expected final state moving reference line offset data is the lateral displacement data between the expected final state moving reference line and the initial moving reference line, and the expected final state moving reference line is the moving reference line determined for the moving object in the current frame and used in the expected final state.

[0101] The expected final state moving reference line offset data is the lateral displacement data between the moving reference line used in the expected final state and the initial moving reference line determined for the moving object in the current frame. In the embodiment of the present application, after calculating the final state lateral displacement data in each frame, the expected final state moving reference line offset data is generated based on the final state lateral displacement data to determine the position of the moving reference line of the moving object in the expected final state. The final state lateral displacement data is related to the lateral displacement data and lateral buffer data of the bypassed object. The lateral buffer data is a constant. If the lateral displacement data of the bypassed object changes, the final state lateral displacement data will also change, and accordingly, the expected final state moving reference line offset data will also change.

[0102] The expected final state moving reference line offset data is used to determine the moving reference line used in the expected final state determined for the moving object in the current frame, which is called the expected final state moving reference line. Figure 1A The final state reference line in the figure. The lateral displacement data between it and the initial moving reference line is the expected final state moving reference line offset data. Therefore, in the process of the moving object circling the circumvented object, the moving object will move as far as possible in the direction of the expected final state moving reference line, away from the initial moving reference line, and the circumvented object is located at a preset distance position from the initial moving reference line. Therefore, in the circumventing process, the moving object can be away from the circumvented side boundary of the circumvented object, so that the moving object can smoothly circumvent the circumvented object.

[0103] After obtaining the expected final state moving reference line offset data, it also includes: obtaining the travel path data planned for the mobile object in the current frame according to the expected final state moving reference line offset data. The obtaining of the travel path data planned for the mobile object in the current frame according to the expected final state moving reference line offset data includes: using the expected final state moving reference line offset data as the value of the expected final state moving reference line lateral displacement parameter in the objective function, and obtaining the lateral movement parameter value of the position point in the travel path planned for the mobile object in the current frame through the objective function, the expected final state moving reference line lateral displacement parameter is used to represent the lateral displacement between the position point of the expected final state moving reference line and the initial moving reference line, and the objective function is used to calculate the cost function value corresponding to the travel path.

[0104] The above steps use the expected final state moving reference line offset data as the value of the expected final state moving reference line lateral displacement parameter in the objective function, that is, .

[0105] In the embodiment of the present application, the expected final state moving reference line offset data, that is, the value of the expected final state moving reference line lateral displacement parameter in the objective function, is calculated for each frame. When calculating the objective function of Formula 4, the lateral movement parameter values ​​of other position points in the path obtained frame by frame gradually increase, and the absolute value of the difference between this value and the value of the lateral displacement parameter of the expected final state moving reference line gradually decreases, which can reduce the amount of calculation of the cost function value. In addition, the lateral movement parameter values ​​of the path points in the path planned for the moving object in this frame include a variety of acquisition methods, which are not limited here.

[0106] Second embodiment Corresponding to the embodiment corresponding to the application scenario of the data processing method provided by the present application and the data processing method provided by the first embodiment, the second embodiment of the present application also provides a data processing device. Since the device embodiment is basically similar to the embodiment corresponding to the application scenario and the first embodiment, the description is relatively simple. For relevant parts, please refer to the embodiment corresponding to the application scenario and the partial description of the first embodiment. The device embodiment described below is only exemplary. Please refer to Figure 5, which is a schematic diagram of a data processing device provided in the second embodiment of the present application. It includes: a first final state lateral displacement data acquisition unit 501, used to acquire the final state lateral displacement data determined for the mobile object in the current frame, the final state lateral displacement data is the lateral displacement data of the mobile object relative to the initial moving reference line used by the mobile object at the initial position point when the mobile object is in the expected final state, the expected final state is the expected state with a lateral safety space between the mobile object and the object to be bypassed, and the initial position point is the initial position point of the path that the mobile object needs to travel to bypass the object to be bypassed; a mobile state adjustment safety duration data acquisition unit 502, used to acquire the mobile state adjustment safety duration data determined for the mobile object in the current frame, the mobile state adjustment safety duration data is the expected available duration data for the mobile object to adjust the mobile state to avoid collision with the object to be bypassed starting from the current frame; a first lateral offset step length data acquisition unit 503, used to obtain the final state lateral displacement data and the final state lateral displacement data according to the final state lateral displacement data and the initial position point. The moving state adjusts the safety time data to obtain the lateral offset step data corresponding to the current frame; the first target lateral offset data obtaining unit 504 is used to adjust the target lateral offset data corresponding to the previous frame according to the lateral offset step data corresponding to the current frame, and obtain the target lateral offset data determined for the mobile object in the current frame as the target lateral offset data corresponding to the current frame, the target lateral offset data corresponding to the previous frame is the lateral displacement data between the previous frame target position point determined for the mobile object in the previous frame and the initial movement reference line used by the mobile object at the initial position point, the previous frame target position point is the end point of the travel path planned for the mobile object in the previous frame, the target lateral offset data corresponding to the current frame is the lateral displacement data between the current frame target position point and the initial movement reference line used by the mobile object at the initial position point, and the current frame target position point is the end point of the travel path planned for the mobile object in the current frame.

[0107] Third embodiment The third embodiment of the present application also provides another data processing method corresponding to the embodiment corresponding to the application scenario of the data processing method provided by the first embodiment. The data processing method provided by the third embodiment of the present application corresponds to the scenario embodiment and the first embodiment. For relevant matters, please refer to the partial description of the scenario embodiment and the first embodiment. The following description of the third embodiment is only illustrative. Please refer to Figure 6 , which is a flowchart of another data processing method provided in the third embodiment of the present application. Figure 6The address information processing method shown includes: Step S601 to Step S603. Step S601: Obtain the moving speed data of the moving object in the current frame. In this step, the moving speed data of the moving object in the current frame refers to the moving speed data of the moving object in the longitudinal direction. For example, Figure 2A In the example, the current frame is the starting frame, the moving object is the autonomous driving vehicle, and the moving speed data of the autonomous driving vehicle in the starting frame is the moving speed data of the autonomous driving vehicle in the vehicle. 起0 The purpose of this step is to obtain the lateral offset step length data corresponding to the current frame in step S602 according to the obtained moving speed data.

[0108] Step S602: According to the moving speed data of the moving object in the current frame, and the correspondence between the moving speed data range and the single-frame lateral offset step data range, the lateral offset step data corresponding to the current frame is obtained. This step is to obtain the lateral offset step data corresponding to the current frame according to the correspondence between the moving speed data range and the single-frame lateral offset step data range.

[0109] The corresponding relationship between the moving speed data range and the single frame lateral offset step length data range is as described in the scenario embodiment. After obtaining the lateral offset step length data corresponding to the current frame in this step, the target lateral offset data corresponding to the current frame is obtained in step S603.

[0110] Step S603: According to the lateral offset step data corresponding to the current frame, the target lateral offset data corresponding to the previous frame is adjusted to obtain the target lateral offset data determined for the mobile object in the current frame as the target lateral offset data corresponding to the current frame; wherein the target lateral offset data corresponding to the previous frame is the lateral displacement data between the target position point of the previous frame determined for the mobile object in the previous frame and the initial moving reference line used by the mobile object at the initial position point, the target position point of the previous frame is the end point of the travel path planned for the mobile object in the previous frame, and the target lateral offset data corresponding to the current frame is the lateral displacement data between the target position point of the current frame and the initial moving reference line used by the mobile object at the initial position point, and the target position point of the current frame is the end point of the travel path planned for the mobile object in the current frame. The purpose of this step is to obtain the target lateral offset data corresponding to the current frame, and this step corresponds to step S404 in the first embodiment. For specific description, refer to the description of S404, which will not be repeated here.

[0111] According to the method of step S601 to step S603, if the current moving speed data of the moving object in the current frame changes, the lateral offset step data corresponding to the current frame also changes, so that the change value of the lateral offset step data can be adjusted in time according to the change of the moving speed data of the moving object in each frame. In each frame, the target lateral offset data corresponding to the current frame will be adjusted in time according to the corresponding lateral offset step data, that is, the target lateral offset data of the terminal position of the path planned for the moving object in the current frame is adjusted frame by frame, so that the numerical change process between the target lateral offset data corresponding to two adjacent frames is relatively smooth, so that the moving object can smoothly bypass the bypassed object.

[0112] Combined with the target lateral offset data of the current frame obtained in step S601 to step S603, the target lateral offset data can be used to calculate the path data planned for the mobile object in the current frame. This is the first path data acquisition method: according to the target lateral offset data corresponding to the current frame, the path data planned for the mobile object in the current frame is obtained; wherein, according to the target lateral offset data corresponding to the current frame, the path data planned for the mobile object in the current frame is obtained, including: using the target lateral offset data corresponding to the current frame as the value of the final state lateral displacement parameter in the objective function, obtaining the lateral movement parameter value of the position point in the path planned for the mobile object in the current frame through the objective function, the final state lateral displacement parameter is used to represent the lateral displacement between the final state position point in the path and the initial movement reference line, and the objective function is used to calculate the cost function value corresponding to the path. Use the target lateral offset data corresponding to the current frame as the value of the final state lateral displacement parameter in the objective function.

[0113] According to the above formula 3, the target lateral offset data corresponding to the current frame is , which is used as the reference value of the final state lateral displacement in the objective function, such as the cost function in formula 4 The specific description process of the first method for acquiring the travel path data can refer to the further detailed explanation of step S404 in the first embodiment.

[0114] In addition, the method may further include obtaining the expected final state moving reference line offset data determined for the moving object in the current frame, and the specific process is as follows: obtaining the final state lateral displacement data determined for the moving object in the current frame, the final state lateral displacement data is the lateral displacement data of the moving object relative to the initial moving reference line used by the moving object at the initial position point when the moving object is in the expected final state, the expected final state is the expected state with a lateral safety space between the moving object and the object to be bypassed, and the initial position point is the initial position point of the path that the moving object needs to travel to bypass the object to be bypassed; according to the final state lateral displacement data, obtaining the expected final state moving reference line offset data of the moving object; wherein the expected final state moving reference line offset data is the lateral displacement data between the expected final state moving reference line and the initial moving reference line, and the expected final state moving reference line is the moving reference line used in the expected final state determined for the moving object in the current frame. The method of obtaining the final state lateral displacement data is similar to step S401 in the first embodiment, and the description of step S401 may be referred to, and will not be repeated here.

[0115] The description process of obtaining the expected final state moving reference line offset data of the moving object according to the final state lateral displacement data is similar to the explanation of the further added technical features of step S401 in the first embodiment, and reference may be made to the explanation of the further added technical features of step S401 here. The expected final state moving reference line offset data is used to determine the moving reference line used in the expected final state determined for the moving object in the current frame, which is called the expected final state moving reference line, such as Figure 1A The final state reference line in the figure. The lateral displacement data between it and the initial moving reference line is the expected final state moving reference line offset data. Therefore, when the moving object is circling the circumvented object, the moving object will move as far as possible in the direction of the expected final state moving reference line, away from the initial moving reference line, and the circumvented object is located at a preset distance position of the initial moving reference line. Therefore, the moving object can be away from the circumvented side boundary of the circumvented object, so that the moving object can smoothly circumvent the circumvented object.

[0116] In addition, after obtaining the expected final state moving reference line offset data, a second method for obtaining the travel path data may also be included: specifically, according to the expected final state moving reference line offset data, the travel path data planned for the mobile object in the current frame is obtained; wherein, according to the expected final state moving reference line offset data, the travel path data planned for the mobile object in the current frame is obtained, including: using the expected final state moving reference line offset data as the value of the moving reference line lateral displacement parameter in the objective function, and obtaining the lateral movement parameter value of the position point in the travel path planned for the mobile object in the current frame through the objective function, the moving reference line lateral displacement parameter is used to represent the lateral displacement between the position point of the moving reference line and the initial moving reference line, and the objective function is used to calculate the cost function value corresponding to the travel path. In the embodiment of the present application, the expected final state moving reference line offset data, that is, the value of the expected final state moving reference line lateral displacement parameter in the objective function, is calculated for each frame. When calculating the objective function of Formula 4, the lateral movement parameter values ​​of other position points in the path obtained frame by frame gradually increase, and the absolute value of the difference between this value and the value of the lateral displacement parameter of the expected final state movement reference line gradually decreases, which can reduce the amount of calculation of the cost function value. In addition, the lateral movement parameter values ​​of the path points in the path planned for the moving object in this frame include multiple acquisition methods, which are not limited here.

[0117] Fourth embodiment Corresponding to the embodiment corresponding to the application scenario of the data processing method provided by the present application and the data processing method provided by the third embodiment, the fourth embodiment of the present application also provides another data processing device. Since the device embodiment is basically similar to the embodiment corresponding to the application scenario and the third embodiment, the description is relatively simple, and the relevant parts can be referred to the embodiment corresponding to the application scenario and the partial description of the third embodiment. The device embodiment described below is only illustrative.

[0118] Please refer to Figure 7, which is a schematic diagram of another data processing device provided in the fourth embodiment of the present application. It includes: a moving speed data acquisition unit 701, used to acquire the moving speed data of the moving object in the current frame; a second lateral offset step data acquisition unit 702, used to acquire the lateral offset step data corresponding to the current frame according to the moving speed data of the moving object in the current frame, and the correspondence between the moving speed data range and the single-frame lateral offset step data range; a second target lateral offset data acquisition unit 703, used to adjust the target lateral offset data corresponding to the previous frame according to the lateral offset step data corresponding to the current frame, and acquire the target lateral offset data determined for the moving object in the current frame as the target lateral offset data for the current frame. The target lateral offset data corresponding to the previous frame is the lateral displacement data between the previous frame target position point determined for the mobile object in the previous frame and the initial movement reference line used by the mobile object at the initial position point, and the previous frame target position point is the end point of the travel path planned for the mobile object in the previous frame, and the target lateral offset data corresponding to the current frame is the lateral displacement data between the current frame target position point and the initial movement reference line used by the mobile object at the initial position point, and the current frame target position point is the end point of the travel path planned for the mobile object in the current frame.

[0119] Fifth embodiment The fifth embodiment of the present application also provides another data processing method corresponding to the embodiment corresponding to the application scenario of the data processing method provided by the first embodiment. The data processing method provided by the fifth embodiment of the present application corresponds to the scenario embodiment and the first embodiment. For relevant matters, please refer to the partial description of the scenario embodiment and the first embodiment. The following description of the fifth embodiment is only illustrative. Please refer to Figure 8 , which is a flowchart of another data processing method provided in the fifth embodiment of the present application. Figure 8The method for generating moving reference line offset data shown includes: step S801 to step S802. Step S801: Acquire the final state lateral displacement data determined for the moving object in the current frame, the final state lateral displacement data is the lateral displacement data of the moving object relative to the initial moving reference line used by the moving object at the initial position point when the moving object is in the expected final state, the expected final state is the expected state with a lateral safety space between the moving object and the object to be bypassed, and the initial position point is the initial position point of the path that the moving object needs to travel to bypass the object to be bypassed. This step is used to determine the final state lateral displacement data for the moving object in the current frame, and the process is similar to step S401 of the first embodiment, and reference may be made to the detailed description of step S401. This step acquires the final state lateral displacement data, which is used to determine the moving reference line offset data of the moving object in the current frame according to the final state lateral displacement data in subsequent steps.

[0120] Step S802: Obtain the expected final state moving reference line offset data of the moving object according to the final state lateral displacement data, wherein the expected final state moving reference line offset data is the lateral displacement data between the expected final state moving reference line and the initial moving reference line, and the expected final state moving reference line is the moving reference line used in the expected final state determined for the moving object in the current frame. This step is used to obtain the expected final state moving reference line offset data of the moving object according to the final state lateral displacement data. The expected final state moving reference line offset data is the lateral displacement data between the moving reference line used in the expected final state determined for the moving object in the current frame and the initial moving reference line. In the embodiment of the present application, after calculating the final state lateral displacement data in each frame, the expected final state moving reference line offset data is generated according to the final state lateral displacement data to determine the position of the moving reference line of the moving object in the expected final state. The final state lateral displacement data is related to the lateral displacement data and the lateral buffer data of the bypassed object, and the lateral buffer data is a constant. If the lateral displacement data of the bypassed object changes, the final state lateral displacement data will also change, and correspondingly, the expected final state moving reference line offset data will also change.

[0121] The expected final state moving reference line offset data is used to determine the moving reference line used in the expected final state determined for the moving object in the current frame, which is called the expected final state moving reference line. Figure 1AThe final state reference line in the figure. The lateral displacement data between it and the initial moving reference line is the expected final state moving reference line offset data. Therefore, in the process of the moving object circling the circumvented object, the moving object will move as far as possible in the direction of the expected final state moving reference line, away from the initial moving reference line, and the circumvented object is located at a preset distance position from the initial moving reference line. Therefore, in the circumventing process, the moving object can be away from the circumvented side boundary of the circumvented object, so that the moving object can smoothly circumvent the circumvented object.

[0122] In addition, after obtaining the expected final state moving reference line offset data, it also includes: according to the expected final state moving reference line offset data, obtaining the travel path data planned for the mobile object in the current frame. According to the expected final state moving reference line offset data, obtaining the travel path data planned for the mobile object in the current frame includes: using the expected final state moving reference line offset data as the value of the moving reference line lateral displacement parameter in the objective function, and obtaining the lateral movement parameter value of the position point in the travel path planned for the mobile object in the current frame through the objective function, the moving reference line lateral displacement parameter is used to represent the lateral displacement between the position point of the moving reference line and the initial moving reference line, and the objective function is used to calculate the cost function value corresponding to the travel path.

[0123] In the embodiment of the present application, the expected final state moving reference line offset data, that is, the value of the expected final state moving reference line lateral displacement parameter in the objective function, is calculated for each frame. When calculating the objective function of Formula 4, the lateral movement parameter values ​​of other position points in the path obtained frame by frame gradually increase, and the absolute value of the difference between this value and the value of the lateral displacement parameter of the expected final state moving reference line gradually decreases, which can reduce the amount of calculation of the cost function value. In addition, the lateral movement parameter values ​​of the path points in the path planned for the moving object in this frame include a variety of acquisition methods, which are not limited here.

[0124] In addition, in addition to the above-mentioned method of obtaining the travel path data planned for the mobile object in the current frame, the travel path data can also be obtained in the following method: obtaining the target lateral offset data corresponding to the current frame, the target lateral offset data corresponding to the current frame is the lateral displacement data between the target position point of the current frame and the initial moving reference line used by the mobile object at the initial position point, and the current frame target position point is the end point of the travel path planned for the mobile object in the current frame; based on the target lateral offset data corresponding to the current frame, obtain the travel path data planned for the mobile object in the current frame.

[0125] The method of obtaining the travel path data planned for the mobile object in the current frame based on the target lateral offset data corresponding to the current frame includes: using the target lateral offset data corresponding to the current frame as the value of the final-state lateral displacement parameter in the objective function, and obtaining the lateral movement parameter value of the position point in the travel path planned for the mobile object in the current frame through the objective function, the final-state lateral displacement parameter is used to represent the lateral displacement between the final-state position point in the travel path and the initial movement reference line, and the objective function is used to calculate the cost function value corresponding to the travel path.

[0126] According to the target lateral offset data corresponding to the current frame, the path data planned for the mobile object in the current frame is obtained. The path planned for the mobile object in the current frame is the path of the mobile object starting to move for a preset time at the position of the current frame. Because the target lateral offset data corresponding to the current frame is the lateral displacement data between the end point of the path planned for the mobile object in the current frame and the initial moving reference line used by the mobile object. Therefore, based on the change of the target lateral offset data determined for the mobile object in each frame, the lateral displacement data between the end point of the corresponding path and the initial moving reference line will also change, and accordingly, the lateral offset data of other path position points of the path planned for the mobile object in each frame will also change. By planning the path for the mobile object frame by frame, and gradually adjusting the lateral displacement data between the end point of the path and the initial moving reference line, in the last frame, the lateral displacement data corresponding to the end point of the path planned for the mobile object may be equal to the final state lateral displacement data of the mobile object in the expected final state, so as to complete the process of the mobile object bypassing the bypassed object. The above method plans the path frame by frame, and conditions the lateral displacement data of the end point of the path and the initial moving reference line frame by frame, that is, increases the lateral displacement data of the moving object in the final state frame by frame. This adjustment method makes the change value of the lateral displacement data of the final state obtained by the moving object in each frame relatively small, and the moving state of the moving object is relatively stable, thus realizing that the moving object can smoothly bypass the bypassed object.

[0127] Sixth embodiment Corresponding to the embodiment corresponding to the application scenario of the data processing method provided by the present application and another data processing method provided by the fifth embodiment, the sixth embodiment of the present application also provides another data processing device. Since the device embodiment is basically similar to the embodiment corresponding to the application scenario and the fifth embodiment, the description is relatively simple, and the relevant parts refer to the embodiment corresponding to the application scenario and the partial description of the fifth embodiment. The device embodiment described below is only illustrative.

[0128] Please refer to Fig. 9, which is a schematic diagram of another data processing device provided in the sixth embodiment of the present application. It includes: a second final state lateral displacement data acquisition unit 901, used to acquire the final state lateral displacement data determined for the mobile object in the current frame, the final state lateral displacement data is the lateral displacement data of the mobile object relative to the initial moving reference line used by the mobile object at the initial position point when the mobile object is in the expected final state, the expected final state is the expected state with a lateral safety space between the mobile object and the bypassed object required to bypass the bypassed object, and the initial position point is the initial position point of the path that the mobile object needs to travel to bypass the bypassed object; a first moving reference line offset data acquisition unit 902, used to acquire the expected final state moving reference line offset data of the mobile object according to the final state lateral displacement data, wherein the expected final state moving reference line offset data is the lateral displacement data between the expected final state moving reference line and the initial moving reference line, and the expected final state moving reference line is the moving reference line used in the expected final state determined for the mobile object in the current frame.

[0129] Seventh embodiment Corresponding to the embodiments corresponding to the application scenarios of the data processing method provided by the present application, the data processing method provided by the first embodiment, and the data processing method provided by the second embodiment, the seventh embodiment of the present application also provides a path planning method for a mobile object. The path planning method for a mobile object provided by the seventh embodiment of the present application corresponds to the scenario embodiments, the first embodiment, and the second embodiment. For relevant matters, please refer to the partial description of the scenario embodiments, the first embodiment, and the second embodiment. The following description of the seventh embodiment is merely illustrative. Please refer to Fig.10 , which is a flowchart of a path planning method for a mobile object provided in the seventh embodiment of the present application, including: step S1001 to step S1002.

[0130] Step S1001: Obtain the target lateral offset data determined for the mobile object in the current frame as the target lateral offset data corresponding to the current frame. The target lateral offset data corresponding to the current frame is the lateral displacement data between the target position point of the current frame and the initial moving reference line used by the mobile object at the initial position point. The target position point of the current frame is the end point of the travel path planned for the mobile object in the current frame, and the initial position point is the initial position point of the travel path that the mobile object needs to travel in order to bypass the bypassed object. This step is used to obtain the target lateral offset data determined for the mobile object in the current frame, so that the travel path data planned for the mobile object in the current frame can be obtained in step S1002. Among them, there are at least two ways to determine the target lateral offset data, which are described below respectively: The first method for determining target lateral offset data: the target lateral offset data determined for the mobile object in the current frame is obtained, including: obtaining final state lateral displacement data determined for the mobile object in the current frame, the final state lateral displacement data being the lateral displacement data of the mobile object relative to the initial moving reference line used by the mobile object at the initial position point at the expected final state, the expected final state being the expected state with a lateral safety space between the mobile object and the bypassed object required to be achieved in order to bypass the bypassed object; obtaining moving state adjustment safety time length data determined for the mobile object in the current frame, the moving state adjustment safety time length data being the time length from the current frame for the mobile object to adjust the moving state to avoid collision with the bypassed object. The expected available duration data of the whole moving state; according to the final state lateral displacement data and the moving state adjustment safety duration data, the lateral offset step length data corresponding to the current frame is obtained; according to the lateral offset step length data corresponding to the current frame, the target lateral offset data corresponding to the previous frame is adjusted to obtain the target lateral offset data determined for the moving object in the current frame, as the target lateral offset data corresponding to the current frame, the target lateral offset data corresponding to the previous frame is the lateral displacement data between the previous frame target position point determined for the moving object in the previous frame and the initial moving reference line used by the moving object at the initial position point, and the previous frame target position point is the end point of the travel path planned for the moving object in the previous frame. Among them, the first method for determining the target lateral offset data is similar to the description of steps S401 to S404 in the first embodiment, and the detailed explanation of steps S401 to S404 can be referred to.

[0131] The second method for determining the target lateral offset data: the acquisition of the target lateral offset data determined for the mobile object in the current frame includes: acquiring the moving speed data of the mobile object in the current frame; obtaining the lateral offset step data corresponding to the current frame according to the moving speed data of the mobile object in the current frame and the correspondence between the moving speed data range and the single-frame lateral offset step data range; adjusting the target lateral offset data corresponding to the previous frame according to the lateral offset step data corresponding to the current frame, and obtaining the target lateral offset data determined for the mobile object in the current frame as the target lateral offset data corresponding to the current frame; wherein the target lateral offset data corresponding to the previous frame is the lateral displacement data between the previous frame target position point determined for the mobile object in the previous frame and the initial movement reference line used by the mobile object at the initial position point, and the previous frame target position point is the end point of the travel path planned for the mobile object in the previous frame. The second method for determining the target lateral offset data is similar to the description of steps S601 to S603 in the second embodiment, and the detailed explanation of steps S601 to S603 can be referred to.

[0132] Step S1002: According to the target lateral offset data corresponding to the current frame, obtain the path data planned for the mobile object in the current frame. In this step, according to the target lateral offset data corresponding to the current frame, obtain the path data planned for the mobile object in the current frame. According to the target lateral offset data corresponding to the current frame, obtain the path data planned for the mobile object in the current frame. The path planned for the mobile object in the current frame is the path of the mobile object starting to move for a preset time at the position of the current frame. Because the target lateral offset data corresponding to the current frame is the lateral displacement data between the end point of the path planned for the mobile object in the current frame and the initial moving reference line used by the mobile object. Therefore, based on the change of the target lateral offset data determined for the mobile object in each frame, the lateral displacement data between the end point of the corresponding path and the initial moving reference line will also change, and accordingly, the lateral offset data of other path position points of the path planned for the mobile object in each frame will also change. By planning the travel path for the mobile object frame by frame and gradually adjusting the lateral displacement data between the end point of the travel path and the initial moving reference line, in the last frame, the lateral displacement data corresponding to the end point of the travel path planned for the mobile object may be equal to the final state lateral displacement data of the mobile object in the expected final state, so as to complete the process of the mobile object bypassing the bypassed object. The above method plans the travel path frame by frame, and adjusts the lateral displacement data between the end point of the travel path and the initial moving reference line frame by frame, that is, increases the lateral displacement data of the mobile object in the final state frame by frame. This adjustment method makes the change value of the lateral displacement data of the final state obtained by the mobile object in each frame relatively small, and the moving state of the mobile object is relatively stable, thereby realizing that the mobile object smoothly bypasses the bypassed object.

[0133] Wherein, the step of obtaining the path data planned for the mobile object in the current frame according to the target lateral offset data corresponding to the current frame includes: using the target lateral offset data corresponding to the current frame as the value of the final state lateral displacement parameter in the objective function, obtaining the lateral movement parameter value of the position point in the path planned for the mobile object in the current frame through the objective function, the final state lateral displacement parameter is used to represent the lateral displacement between the final state position point in the path and the initial movement reference line, and the objective function is used to calculate the cost function value corresponding to the path. Using the target lateral offset data corresponding to the current frame as the value of the final state lateral displacement parameter in the objective function.

[0134] According to the above formula 3, the target lateral offset data corresponding to the current frame is , which is used as the reference value of the final state lateral displacement in the objective function, such as the cost function in formula 4 The above process may refer to the explanation of further adding technical features in step S404.

[0135] In addition, in order to reduce the amount of calculation of the objective function, it can also include: obtaining the final state lateral displacement data determined for the moving object in the current frame, the final state lateral displacement data is the lateral displacement data of the moving object relative to the initial moving reference line used by the moving object at the initial position point when the moving object is in the expected final state, and the expected final state is the expected state that the moving object needs to achieve in order to bypass the bypassed object and has a lateral safety space between the bypassed object; according to the final state lateral displacement data, obtaining the expected final state moving reference line offset data of the moving object, the expected final state moving reference line offset data is the lateral displacement data between the expected final state moving reference line and the initial moving reference line, and the expected final state moving reference line is the moving reference line used in the expected final state determined for the moving object in the current frame; using the expected final state moving reference line offset data as the value of the moving reference line lateral displacement parameter in the objective function, the moving reference line lateral displacement parameter is used to represent the lateral displacement between the position point of the moving reference line and the initial moving reference line, and the objective function is used to calculate the cost function value corresponding to the travel path.

[0136] In the embodiment of the present application, the expected final state moving reference line offset data, that is, the value of the expected final state moving reference line lateral displacement parameter in the objective function, is calculated for each frame. When calculating the objective function of Formula 4, the lateral movement parameter values ​​of other position points in the path obtained frame by frame gradually increase, and the absolute value of the difference between this value and the value of the lateral displacement parameter of the expected final state moving reference line gradually decreases, which can reduce the amount of calculation of the cost function value. In addition, the lateral movement parameter values ​​of the path points in the path planned for the moving object in this frame include a variety of acquisition methods, which are not limited here.

[0137] Eighth embodiment Corresponding to the embodiment corresponding to the application scenario of the data processing method provided by the present application and the path planning method for a mobile object provided by the seventh embodiment, the eighth embodiment of the present application also provides a path planning device for a mobile object. Since the device embodiment is basically similar to the embodiment corresponding to the application scenario and the seventh embodiment, the description is relatively simple. For relevant parts, please refer to the embodiment corresponding to the application scenario and the partial description of the seventh embodiment. Please refer to Fig.11, which is a schematic diagram of a path planning device for a mobile object provided in the eighth embodiment of the present application. It includes: a third target lateral offset data acquisition unit 1101, used to acquire the target lateral offset data determined for the mobile object in the current frame as the target lateral offset data corresponding to the current frame, the target lateral offset data corresponding to the current frame is the lateral displacement data between the target position point of the current frame and the initial moving reference line used by the mobile object at the initial position point, the current frame target position point is the end point of the travel path planned for the mobile object in the current frame, and the initial position point is the initial position point of the travel path that the mobile object needs to travel to bypass the bypassed object; a first travel path data acquisition unit 1102, used to acquire the travel path data planned for the mobile object in the current frame according to the target lateral offset data corresponding to the current frame.

[0138] Ninth embodiment Corresponding to the embodiments corresponding to the application scenarios of the data processing method provided by the present application, the data processing method provided by the first embodiment, and the data processing method provided by the second embodiment, the ninth embodiment of the present application also provides another path planning method for a mobile object. The path planning method for a mobile object provided by the ninth embodiment of the present application corresponds to the scenario embodiments, the first embodiment, and the second embodiment. For relevant matters, please refer to the partial description of the scenario embodiments, the first embodiment, and the second embodiment. The following description of the ninth embodiment is merely illustrative. Please refer to Fig.12 , which is a flowchart of another path planning method for a mobile object provided in the ninth embodiment of the present application. Fig.12 The path planning method for a mobile object shown includes: steps S1201 to S1203.

[0139] Step S1201: Acquire the final state lateral displacement data determined for the mobile object in the current frame, the final state lateral displacement data is the lateral displacement data of the mobile object relative to the initial moving reference line used by the mobile object at the initial position point when the mobile object is in the expected final state, the expected final state is the expected state with a lateral safety space between the mobile object and the object to be bypassed, and the initial position point is the initial position point of the path that the mobile object needs to travel to bypass the object to be bypassed. This step is used to acquire the final state lateral displacement data determined for the mobile object in the current frame, and the process is similar to step S401 in the first embodiment. For details, please refer to the description of step S401.

[0140] The method of obtaining the final state lateral displacement data determined for the mobile object in the current frame includes: obtaining the lateral displacement data of the wound object, wherein the lateral displacement data of the wound object is the lateral displacement data between the wound side boundary of the wound object and the initial movement reference line; obtaining the lateral buffer data of the mobile object in the expected final state, wherein the lateral buffer data is used to represent the lateral safety space between the mobile object in the expected final state and the wound object; and obtaining the final state lateral displacement data according to the lateral displacement data of the wound object and the lateral buffer data.

[0141] In addition, it also includes: obtaining the lateral width data of the moving object; obtaining the final state lateral displacement data based on the lateral displacement data of the bypassed object and the lateral buffer data, including: obtaining the final state lateral displacement data based on the lateral displacement data of the bypassed object, the lateral buffer data and the lateral width data.

[0142] The step of obtaining the final state lateral displacement data according to the lateral displacement data of the wound object, the lateral buffer data and the lateral width data comprises: obtaining the sum of the lateral displacement data of the wound object, the lateral buffer data and a preset ratio of the lateral width data as the final state lateral displacement data.

[0143] The method further includes: determining whether the wound side boundary of the wound object has been laterally offset in the current frame relative to the previous frame; if the wound side boundary of the wound object has been laterally offset in the current frame relative to the previous frame, determining whether the lateral offset data of the wound side boundary of the wound object in the current frame relative to the previous frame exceeds a preset boundary lateral offset data threshold; if the lateral offset data of the wound side boundary of the wound object in the current frame relative to the previous frame does not exceed the preset boundary lateral offset data threshold, using the final state lateral displacement data determined for the moving object in the previous frame as the final state lateral displacement data determined for the moving object in the current frame; the acquiring of the lateral displacement data of the wound object includes: acquiring the lateral displacement data of the wound object if the lateral offset data of the wound side boundary of the wound object in the current frame relative to the previous frame exceeds the preset boundary lateral offset data threshold. The above-mentioned specific description of the acquisition method of the final state lateral displacement data determined for the moving object in the current frame can refer to the further detailed description of step S401 in the first embodiment.

[0144] Step S1202: According to the final state lateral displacement data, obtain the expected final state moving reference line offset data of the moving object, the expected final state moving reference line offset data is the lateral displacement data between the expected final state moving reference line and the initial moving reference line, and the expected final state moving reference line is the moving reference line used in the expected final state determined for the moving object in the current frame. This step is used to obtain the expected final state moving reference line offset data of the moving object according to the final state lateral displacement data. The expected final state moving reference line offset data is used to determine the moving reference line used in the expected final state determined for the moving object in the current frame, which is called the expected final state moving reference line, such as Figure 1A The lateral displacement data between the moving reference line and the initial moving reference line is the expected moving reference line offset data of the final state. Therefore, when the moving object is circling the circumvented object, the moving object will move as far as possible in the direction of the expected final state moving reference line, away from the initial moving reference line, and the circumvented object is located at a preset distance position from the initial moving reference line. Therefore, during the circumventing process, the moving object can be away from the circumvented side boundary of the circumvented object, so that the moving object can smoothly circumvent the circumvented object.

[0145] Step S1203: According to the expected final state moving reference line offset data, obtain the travel path data planned for the mobile object in the current frame. According to the expected final state moving reference line offset data, obtain the travel path data planned for the mobile object in the current frame, including: using the expected final state moving reference line offset data as the value of the moving reference line lateral displacement parameter in the objective function, obtaining the lateral movement parameter value of the position point in the travel path planned for the mobile object in the current frame through the objective function, the moving reference line lateral displacement parameter is used to represent the lateral displacement between the position point of the moving reference line and the initial moving reference line, and the objective function is used to calculate the cost function value corresponding to the travel path.

[0146] In the embodiment of the present application, the expected final state moving reference line offset data, that is, the value of the expected final state moving reference line lateral displacement parameter in the objective function, is calculated for each frame. When calculating the objective function of Formula 4, the lateral movement parameter values ​​of other position points in the path obtained frame by frame gradually increase, and the absolute value of the difference between this value and the value of the lateral displacement parameter of the expected final state moving reference line gradually decreases, which can reduce the amount of calculation of the cost function value. In addition, the lateral movement parameter values ​​of the path points in the path planned for the moving object in this frame include a variety of acquisition methods, which are not limited here.

[0147] In order to improve the stability of the moving object when it bypasses the bypassed object, and improve the accuracy of the value of the final state lateral displacement parameter for the cost function value of the calculated objective function, the method further includes: obtaining the target lateral displacement data determined for the moving object in the current frame as the target lateral displacement data corresponding to the current frame, the target lateral displacement data corresponding to the current frame is the lateral displacement data between the target position point of the current frame and the initial moving reference line used by the moving object at the initial position point, the target position point of the current frame is the end point of the travel path planned for the moving object in the current frame, and the initial position point is the initial position point of the travel path required for the moving object to bypass the bypassed object; using the target lateral displacement data corresponding to the current frame as the value of the final state lateral displacement parameter in the objective function, the final state lateral displacement parameter is used to represent the lateral displacement between the position point of the final state in the travel path and the initial moving reference line. Using the target lateral displacement data corresponding to the current frame as the value of the final state lateral displacement parameter in the objective function. The lateral displacement parameter value at the position point in the travel path is obtained by calculating the cost function value obtained by the objective function, thereby obtaining the travel path data, and then generating the travel path according to the travel path data. This process increases the value of the final state lateral displacement parameter of the moving object in the current frame frame by frame, so that the value of the final state lateral displacement parameter increases gradually and smoothly with each frame, and finally makes the generated path a smooth path, so that the moving object can smoothly bypass the bypassed object. In addition, the moving reference line in the above objective function can be an initial moving reference line or an expected final state moving reference line, which is not limited here.

[0148] Tenth embodiment Corresponding to the embodiment corresponding to the application scenario of the data processing method provided by the present application and another path planning method for a mobile object provided by the ninth embodiment, the tenth embodiment of the present application also provides another path planning device for a mobile object. Since the device embodiment is basically similar to the embodiment corresponding to the application scenario and the ninth embodiment, the description is relatively simple. For relevant matters, please refer to the embodiment corresponding to the application scenario and the partial description of the ninth embodiment. The device embodiment described below is only exemplary. Please refer to Fig.13, which is a schematic diagram of another path planning device for a mobile object provided in the tenth embodiment of the present application. The path planning device for a mobile object provided in the tenth embodiment of the present application includes: a third final state lateral displacement data acquisition unit 1301, which is used to acquire final state lateral displacement data determined for the mobile object in the current frame, wherein the final state lateral displacement data is the lateral displacement data of the mobile object relative to the initial moving reference line used by the mobile object at the initial position point at the expected final state, wherein the expected final state is the expected state that the mobile object needs to reach in order to bypass the bypassed object and has a lateral safety space between the mobile object and the bypassed object, and the initial position point is the initial position of the path that the mobile object needs to travel in order to bypass the bypassed object. point; a second moving reference line offset data obtaining unit 1302, used to obtain the expected final state moving reference line offset data of the moving object according to the final state lateral displacement data, the expected final state moving reference line offset data is the lateral displacement data between the expected final state moving reference line and the initial moving reference line, the expected final state moving reference line is the moving reference line used in the expected final state determined for the moving object in the current frame; a second traveling path data obtaining unit 1303, used to obtain the traveling path data planned for the moving object in the current frame according to the expected final state moving reference line offset data.

[0149] The eleventh embodiment of the present application further provides an electronic device, comprising: a processor; and a memory for storing a computer program, wherein the electronic device is powered on and the computer program is run by the processor to execute the above method. The twelfth embodiment of the present application further provides a computer storage medium, wherein the storage medium stores a computer program, and the computer program is run by the processor to execute the above method.

[0150] The above-mentioned device embodiments, electronic device embodiments and storage medium embodiments correspond to the above-mentioned method embodiments, and the method embodiments may be specifically referred to. Although the present application is disclosed as above with preferred embodiments, it is not intended to limit the present application. Any person skilled in the art may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be subject to the scope defined by the claims of the present application. In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces and memory. The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium. 1. Computer-readable media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information may be a computer-readable instruction, a data structure, a module of a program or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include non-transitory media such as modulated data signals and carriers. 2. It should be understood by those skilled in the art that the embodiments of the present application can be provided as methods, systems or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. It should be noted that the embodiments of the present application may involve the use of user data. In actual applications, user-specific personal data can be used in the scheme described in this article within the scope permitted by applicable laws and regulations, subject to the requirements of applicable laws and regulations of the country where the user is located (for example, with the user's explicit consent, effective notification to the user, etc.).It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

Claims

1. A data processing method, characterized in that: include: Acquire final state lateral displacement data determined for the mobile object in the current frame, the final state lateral displacement data being lateral displacement data of the mobile object relative to an initial movement reference line used by the mobile object at an initial position point at an expected final state, the expected final state being an expected state with a lateral safety space between the mobile object and the bypassed object required to be reached by the mobile object in order to bypass the bypassed object, and the initial position point being an initial position point of a path that the mobile object needs to travel along in order to bypass the bypassed object; Acquire the moving state adjustment safety duration data determined for the moving object in the current frame, wherein the moving state adjustment safety duration data is the expected available duration data for the moving object to adjust the moving state to avoid collision with the object being surrounded, starting from the current frame; Obtaining lateral offset step length data corresponding to the current frame according to the final state lateral displacement data and the moving state adjustment safety time length data; According to the lateral offset step data corresponding to the current frame, the target lateral offset data corresponding to the previous frame is adjusted to obtain the target lateral offset data determined for the mobile object in the current frame as the target lateral offset data corresponding to the current frame. The target lateral offset data corresponding to the previous frame is the lateral displacement data between the previous frame target position point determined for the mobile object in the previous frame and the initial moving reference line used by the mobile object at the initial position point. The previous frame target position point is the end point of the travel path planned for the mobile object in the previous frame. The target lateral offset data corresponding to the current frame is the lateral displacement data between the current frame target position point and the initial moving reference line used by the mobile object at the initial position point. The current frame target position point is the end point of the travel path planned for the mobile object in the current frame.

2. The method according to claim 1, characterized in that The acquiring of the final state lateral displacement data determined for the moving object in the current frame includes: acquiring the lateral displacement data of the wound object, the lateral displacement data of the wound object being the lateral displacement data between the wound side boundary of the wound object and the initial moving reference line; acquiring the lateral buffer data of the moving object in the expected final state, the lateral buffer data being used to represent the lateral safety space between the moving object in the expected final state and the wound object; acquiring a preset ratio of the lateral width data of the moving object; and obtaining the sum of the lateral displacement data of the wound object, the lateral buffer data and the preset ratio of the lateral width data as the final state lateral displacement data; The method further includes: determining whether a wound side boundary of the wound object has been laterally offset in the current frame relative to the previous frame; if a wound side boundary of the wound object has been laterally offset in the current frame relative to the previous frame, determining whether the lateral offset data of the wound side boundary of the wound object in the current frame relative to the previous frame exceeds a preset boundary lateral offset data threshold; if the lateral offset data of the wound side boundary of the wound object in the current frame relative to the previous frame does not exceed the preset boundary lateral offset data threshold, using the final state lateral displacement data determined for the moving object in the previous frame as the final state lateral displacement data determined for the moving object in the current frame; obtaining the lateral displacement data of the wound object includes: if the lateral offset data of the wound side boundary of the wound object in the current frame relative to the previous frame exceeds the preset boundary lateral offset data threshold, obtaining the lateral displacement data of the wound object.

3. The method according to claim 1, characterized in that: Also includes: According to the target lateral offset data corresponding to the current frame, the travel path data planned for the mobile object in the current frame is obtained; wherein, according to the target lateral offset data corresponding to the current frame, the travel path data planned for the mobile object in the current frame is obtained, including: using the target lateral offset data corresponding to the current frame as the value of the final state lateral displacement parameter in the objective function, and obtaining the lateral movement parameter value of the position point in the travel path planned for the mobile object in the current frame through the objective function, the final state lateral displacement parameter is used to represent the lateral displacement between the final state position point in the travel path and the initial movement reference line, and the objective function is used to calculate the cost function value corresponding to the travel path; The method further includes: obtaining a weight value of the final state lateral displacement parameter; obtaining the lateral displacement parameter value of a position point in the travel path planned for the mobile object in the current frame through the objective function, including: using the weight value of the final state lateral displacement parameter as the weight value of the lateral displacement parameter of the final state in the objective function, and obtaining the lateral displacement parameter value of the position point in the travel path planned for the mobile object in the current frame, wherein the weight value of the final state lateral displacement parameter is different from the weight value of the lateral displacement parameter of a non-final state position point, and the non-final state position point is other position points in the travel path planned for the mobile object in the current frame except the end point.

4. The method according to claim 1, characterized in that Also includes: According to the final state lateral displacement data, the expected final state moving reference line offset data of the moving object is obtained; according to the expected final state moving reference line offset data, the travel path data planned for the moving object in the current frame is obtained; wherein, the expected final state moving reference line offset data is the lateral displacement data between the expected final state moving reference line and the initial moving reference line, and the expected final state moving reference line is the moving reference line used in the expected final state determined for the moving object in the current frame; the obtaining of the travel path data planned for the moving object in the current frame according to the expected final state moving reference line offset data includes: using the expected final state moving reference line offset data as the value of the moving reference line lateral displacement parameter in the objective function, obtaining the lateral movement parameter value of the position point in the travel path planned for the moving object in the current frame through the objective function, the moving reference line lateral displacement parameter is used to represent the lateral displacement between the position point of the moving reference line and the initial moving reference line, and the objective function is used to calculate the cost function value corresponding to the travel path.

5. A data processing method, characterized in that: include: Get the moving speed data of the moving object in the current frame; Obtaining the lateral offset step length data corresponding to the current frame according to the moving speed data of the moving object in the current frame, and the corresponding relationship between the moving speed data range and the single-frame lateral offset step length data range; According to the lateral offset step data corresponding to the current frame, the target lateral offset data corresponding to the previous frame is adjusted to obtain the target lateral offset data determined for the moving object in the current frame as the target lateral offset data corresponding to the current frame; Among them, the target lateral offset data corresponding to the previous frame is the lateral displacement data between the previous frame target position point determined for the mobile object in the previous frame and the initial moving reference line used by the mobile object at the initial position point, and the previous frame target position point is the end point of the travel path planned for the mobile object in the previous frame, and the target lateral offset data corresponding to the current frame is the lateral displacement data between the current frame target position point and the initial moving reference line used by the mobile object at the initial position point, and the current frame target position point is the end point of the travel path planned for the mobile object in the current frame.

6. A data processing method, characterized in that: include: Acquire final state lateral displacement data determined for the mobile object in the current frame, the final state lateral displacement data being lateral displacement data of the mobile object relative to an initial movement reference line used by the mobile object at an initial position point at an expected final state, the expected final state being an expected state with a lateral safety space between the mobile object and the bypassed object required to be reached by the mobile object in order to bypass the bypassed object, and the initial position point being an initial position point of a path that the mobile object needs to travel along in order to bypass the bypassed object; According to the final state lateral displacement data, the expected final state moving reference line offset data of the moving object is obtained, wherein the expected final state moving reference line offset data is the lateral displacement data between the expected final state moving reference line and the initial moving reference line, and the expected final state moving reference line is the moving reference line determined for the moving object in the current frame and used in the expected final state.

7. A path planning method for a mobile object, characterized in that: include: Obtain target lateral offset data determined for the mobile object in the current frame as the target lateral offset data corresponding to the current frame, the target lateral offset data corresponding to the current frame being the lateral displacement data between the target position point of the current frame and the initial moving reference line used by the mobile object at the initial position point, the target position point of the current frame being the end point of the travel path planned for the mobile object in the current frame, and the initial position point being the initial position point of the travel path that the mobile object needs to travel in order to bypass the bypassed object; According to the target lateral offset data corresponding to the current frame, the travel path data planned for the mobile object in the current frame is obtained.

8. A path planning method for a mobile object, characterized in that: include: Acquire final state lateral displacement data determined for the mobile object in the current frame, the final state lateral displacement data being lateral displacement data of the mobile object relative to an initial movement reference line used by the mobile object at an initial position point at an expected final state, the expected final state being an expected state with a lateral safety space between the mobile object and the bypassed object required to be reached by the mobile object in order to bypass the bypassed object, and the initial position point being an initial position point of a path that the mobile object needs to travel along in order to bypass the bypassed object; According to the final state lateral displacement data, obtaining the expected final state moving reference line offset data of the moving object, wherein the expected final state moving reference line offset data is the lateral displacement data between the expected final state moving reference line and the initial moving reference line, and the expected final state moving reference line is the moving reference line used in the expected final state determined for the moving object in the current frame; According to the expected final state moving reference line offset data, the travel path data planned for the moving object in the current frame is obtained.

9. An electronic device, comprising: processor; and a memory for storing a computer program, wherein after the electronic device is powered on and the computer program is run by the processor, any method described in 1-8 is executed.

10. A computer storage medium storing a computer program, wherein the computer program is executed by a processor to perform any one of the methods described in 1-8.

Images