Trajectory optimization method, electronic device, and storage medium

Abstract

The application discloses a trajectory optimization method, an electronic device and a storage medium. The trajectory optimization method comprises the following steps: obtaining a first to-be-optimized forward trajectory and a second to-be-optimized forward trajectory; determining a to-be-optimized point set from the first to-be-optimized forward trajectory and the second to-be-optimized forward trajectory, performing trajectory smoothing processing on the to-be-optimized point set to obtain at least one first target forward trajectory point; and generating a target trajectory based on second target forward trajectory points not belonging to the to-be-optimized point set in the first to-be-optimized forward trajectory and the second to-be-optimized forward trajectory and the first target forward trajectory points. The application solves the technical problem of poor effect of the prior art memory automatic parking.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a trajectory optimization method, electronic device, and storage medium. Background Technology

[0002] Currently, major automakers are fiercely competing in the autonomous driving field, indicating that intelligent driving is becoming increasingly important in people's lives and is also a popular research topic for researchers. Autonomous driving is divided into five levels, from L1 to L5, with most currently at the L3 stage. One of the more important functions of L3 is Home-zone Parking Assist (HPA). HPA refers to the ability of an autonomous driving system to automatically complete all commands for parking and exiting a specific parking space in a specific area without human intervention. This falls under the category of L3 intelligent driving.

[0003] HPA (Autonomous Driving Automation) can be divided into two parts: mapping (learning) and cruise. Mapping involves learning a new scene through manual methods to generate map information and reference trajectories for autonomous driving. Cruise is the process of autonomously driving along the reference trajectory using existing map information. The reference trajectory required for cruise has high requirements for the curvature of the trajectory and the positional information of each trajectory point. If the local trajectory is not smooth, it will cause the autonomous driving process to jitter, affecting the user experience. If the positional error of some points in the trajectory is too large, it will lead to the risk of collision during cruise. However, during the mapping process, due to the complexity of parking lot scenarios and the lack of experience of drivers, reversing operations are inevitable. These reference trajectories containing reversing trajectories will affect the control of the vehicle during cruise and also affect the user experience. Summary of the Invention

[0004] The main purpose of this application is to provide a trajectory optimization method, electronic device and storage medium, which aims to solve the technical problem that the existing automatic parking system with memory has poor performance.

[0005] To achieve the above objectives, this application provides a trajectory optimization method, which includes the following steps:

[0006] Obtain the first and second forward trajectories to be optimized;

[0007] Determine a set of points to be optimized from the first and second forward trajectories to be optimized, and perform trajectory smoothing on the set of points to be optimized to obtain at least one first target forward trajectory point;

[0008] A target trajectory is generated based on the second target trajectory points that do not belong to the set of points to be optimized in the first and second target trajectories to be optimized, as well as each of the first target trajectory points.

[0009] This application also provides a trajectory optimization device, the trajectory optimization device comprising:

[0010] The acquisition module is used to acquire the first forward trajectory to be optimized and the second forward trajectory to be optimized.

[0011] The trajectory smoothing module is used to determine a set of points to be optimized from the first forward trajectory to be optimized and the second forward trajectory to be optimized, and to perform trajectory smoothing on the set of points to be optimized to obtain at least one first target forward trajectory point.

[0012] The generation module is used to generate a target trajectory based on the second target trajectory points that do not belong to the set of points to be optimized in the first and second target trajectories to be optimized, as well as each of the first target trajectories to be optimized.

[0013] This application also provides an electronic device, which is a physical device, comprising: a memory, a processor, and a program of the trajectory optimization method stored in the memory and executable on the processor. When the program of the trajectory optimization method is executed by the processor, it can implement the trajectory optimization method or the steps of the trajectory optimization method as described above.

[0014] This application also provides a storage medium, which is a computer-readable storage medium, on which a program implementing a trajectory optimization method is stored. When the program implementing the trajectory optimization method is executed by a processor, it implements the trajectory optimization method or the steps of the trajectory optimization method as described above.

[0015] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the trajectory optimization method or the steps of the trajectory optimization method as described above.

[0016] This application provides a trajectory optimization method, electronic device, and storage medium. First, by acquiring a first forward trajectory to be optimized and a second forward trajectory to be optimized, the forward trajectory is selected. Then, by determining a set of points to be optimized from the first and second forward trajectories to be optimized, the set of points to be optimized is subjected to trajectory smoothing processing to obtain at least one first target forward trajectory point, thus achieving a smooth connection between the first optimized forward trajectory and the second forward trajectory to be optimized. Then, by generating a target trajectory based on the second target forward trajectory points in the first and second forward trajectories to be optimized that do not belong to the set of points to be optimized, as well as each of the first target forward trajectory points, a target trajectory is generated, thus achieving the generation of a target trajectory that does not include a reversing trajectory. In this way, on the one hand, by selecting only the forward trajectory, the reversing trajectory generated during the mapping process due to factors such as the complexity of parking lot scenarios and insufficient driver experience is corrected. This ensures that the final target trajectory does not contain the reversing trajectory, only the forward trajectory, thus avoiding the reversing trajectory from affecting vehicle control during subsequent cruise and generating a more accurate and smooth target trajectory. On the other hand, by smoothly connecting two segments of the forward trajectory to be optimized, the smoothness at the connection point is effectively improved. This avoids the jitter caused by uneven local trajectories during the autonomous driving process, improving the smoothness and stability of subsequent cruise and enhancing the user experience. Therefore, compared to existing technologies that cruise based on the complete trajectory memorized during mapping, this application does not restrict the user's reversing operation during mapping. If a reversing operation occurs during mapping, an accurate and smooth target trajectory can be obtained without re-mapping. Therefore, users can flexibly correct errors or inaccurate driving paths by reversing during mapping, effectively improving the flexibility of the HPA function and enhancing the user experience. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the first embodiment of the trajectory optimization method of this application;

[0020] Figure 2 This is a schematic diagram of the forward and reverse trajectories in the embodiments of this application;

[0021] Figure 3 This is a schematic diagram of a matching point pair scenario in an embodiment of this application;

[0022] Figure 4 This is a schematic diagram of an isolated point scenario in an embodiment of this application;

[0023] Figure 5 This is a schematic diagram of a scenario with reference points in an embodiment of this application;

[0024] Figure 6 This is a schematic diagram of a scene for the point set to be optimized in an embodiment of this application;

[0025] Figure 7 This is a schematic diagram illustrating a scenario with variable weight terms in an embodiment of this application.

[0026] Figure 8 This is a flowchart illustrating the second embodiment of the trajectory optimization method of this application;

[0027] Figure 9 This is a schematic diagram of the trajectory optimization device in the embodiments of this application;

[0028] Figure 10 This is a schematic diagram of the device structure of the hardware operating environment involved in the trajectory optimization method in the embodiments of this application.

[0029] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0030] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Example 1

[0032] This application provides a trajectory optimization method. In the first embodiment of the trajectory optimization method of this application, referring to... Figure 1 The trajectory optimization method includes the following steps:

[0033] Step S10: Obtain the first forward trajectory to be optimized and the second forward trajectory to be optimized;

[0034] The execution subject of the method in this embodiment can be a trajectory optimization device, a trajectory optimization terminal device, or a server. This embodiment takes a trajectory optimization device as an example. The trajectory optimization device can be integrated into terminal devices such as vehicles, in-vehicle terminals, smartphones, tablets, and computers with data processing functions.

[0035] In this embodiment, it should be noted that the trajectory optimization method can be applied to vehicles or servers. After the vehicle triggers the HPA function's mapping process, the vehicle's current location map information and driving trajectory can be collected through sensors, positioning devices, radar, and other devices mounted on the vehicle. The driving trajectory consists of a series of trajectory points containing time information. The vehicle can then automatically execute the steps of the trajectory optimization method to optimize the collected trajectory, generate a target trajectory, and store it for use in subsequent cruise operations. Alternatively, the vehicle can upload the collected map information and its driving trajectory to a server or other terminal device, which will then execute the steps of the trajectory optimization method to optimize the collected trajectory, generate a target trajectory, and send the target trajectory to the vehicle for use in subsequent cruise operations.

[0036] During the mapping process, the vehicle's driving trajectory can include both forward and reverse trajectories. These can be identified by detecting the vehicle's gear position and changes in the external environment during driving. For example, the driving trajectory corresponding to a forward gear is the forward trajectory. Whether the vehicle is reversing can be determined by identifying whether the acquired external environment image is moving backward relative to the vehicle. The specific determination can be made based on the actual situation, and this embodiment does not impose any limitations on this. It is easy to understand that if no reversing operation occurs during the vehicle's driving, the vehicle's driving trajectory will be a complete forward trajectory. In this case, no trajectory optimization is needed, and it can be directly used as the target trajectory for subsequent cruise control. If the vehicle performs at least one reversing operation during its driving, the vehicle's driving trajectory will be divided into multiple segments by the reversing trajectory, referring to... Figure 2 The forward and reverse tracks will be arranged sequentially and alternately in chronological order.

[0037] As an example, step S10 includes: during the vehicle mapping process, after the vehicle has collected at least two forward trajectories, for example, after the vehicle has finished driving or after the vehicle has detected a reversing operation for the second time, acquiring two adjacent complete forward trajectories as the first forward trajectory to be optimized and the second forward trajectory to be optimized, respectively. It is easy to understand that the two forward trajectories are separated by the reversing trajectory, so adjacent does not mean that the two forward trajectories are continuous in time, but rather that they are arranged in time and are consecutive in order.

[0038] In one feasible implementation, optimization can be performed while the vehicle is driving during the mapping process. This allows for sequential optimization of the already determined complete forward trajectories in chronological order. For example, after the vehicle first detects a reversing operation, a first complete forward trajectory is obtained and designated as the first forward trajectory to be optimized. After the second reversing operation is detected, a second complete forward trajectory is obtained and designated as the second forward trajectory to be optimized. Then, the first and second forward trajectories are optimized. Similarly, if the vehicle continues forward after the second reversing operation, the second reversing trajectory is designated as the first forward trajectory to be optimized. After the third reversing operation is detected, a third complete forward trajectory is obtained and designated as the second forward trajectory to be optimized. This process continues until the vehicle finishes driving and the mapping process is complete.

[0039] In another feasible implementation, optimization is performed uniformly after the vehicle has finished driving. This allows the complete driving trajectory of the vehicle to be obtained after the vehicle has finished driving. The complete driving trajectory is then divided into forward trajectory and reverse trajectory. The forward trajectory segments are arranged in chronological order from morning to night. The first and second segments are obtained first, then the second and third segments are obtained, then the third and fourth segments are obtained, and so on, until all forward trajectories have completed trajectory optimization.

[0040] Step S20: Determine the set of points to be optimized from the first forward trajectory to be optimized and the second forward trajectory to be optimized, and perform trajectory smoothing processing on the set of points to be optimized to obtain at least one first target forward trajectory point;

[0041] In this embodiment, it should be noted that since the first and second forward trajectories to be optimized were originally connected by a reversing trajectory, directly connecting them would essentially convert the reversing trajectory into a forward trajectory. The curvature at the connection point between the forward and reversing trajectories is typically large, which could cause the vehicle to make sharp turns or stationary maneuvers at this point, potentially leading to vibrations or abnormal situations during autonomous driving. In this case, the two forward trajectories usually have at least some overlapping sections; therefore, these overlapping sections can be merged and optimized.

[0042] As an example, step S20 includes: determining the range to be optimized from the first forward trajectory to be optimized and the second forward trajectory to be optimized by means of detecting user operation or automatic recognition, determining the forward trajectory points in the range to be optimized as the set of points to be optimized, and then solving for at least one first target forward trajectory point based on the points to be optimized in the set of points to be optimized by a trajectory smoothing algorithm, so that the first forward trajectory to be optimized and the second forward trajectory to be optimized can be smoothly connected.

[0043] Optionally, the step of determining the set of points to be optimized from the first and second paths to be optimized includes:

[0044] Step A10: Calculate the trajectory point distance between each first forward trajectory point in the first forward trajectory to be optimized and each second forward trajectory point in the second forward trajectory to be optimized;

[0045] Step A20: Form a matching point pair between the first forward trajectory point and the second forward trajectory point of the target that have the smallest distances to each other's trajectory points;

[0046] Step A30: Determine the set of points to be optimized based on the matched point pairs and the preset distance threshold.

[0047] In this embodiment, it should be noted that not all parts of the forward trajectory to be optimized need to be processed and can be processed. Excessive processing may cause the vehicle to collide. Therefore, it is necessary to accurately identify the parts that cause turning back and repeated driving due to reversing and to merge and optimize them.

[0048] As an example, steps A10-A30 include: based on the position information of each first forward trajectory point in the first forward trajectory to be optimized, and the position information of each second forward trajectory point in the second forward trajectory to be optimized, calculating the distance from each second forward trajectory point to each first forward trajectory point, and finding the second forward trajectory point closest to each first forward trajectory point; furthermore, the distance from each first forward trajectory point to each second forward trajectory point can be calculated to find the first forward trajectory point closest to each second forward trajectory point, or, for the second forward trajectory point closest to each first forward trajectory point, calculating whether the first forward trajectory point is its minimum distance trajectory point relative to the second forward trajectory point; if there exist target first forward trajectory points and target second forward trajectory points that are each other's minimum distance trajectory points, then the target first forward trajectory point and the target second forward trajectory point are paired, for example, referring to... Figure 3The forward trajectory Fi includes first forward trajectory points A1 and A2. The minimum distance trajectory point corresponding to A1 and A2 is B1, but the minimum distance trajectory point corresponding to B1 is A2. Therefore, it can be determined that A2 and B1 are a pair of matching points, while A1 does not form a pair of matching points with any trajectory point in the forward trajectory Fi+1. In this way, the matching process can be realized quickly and accurately, effectively avoiding the situation where the minimum distance trajectory point of multiple points is the same point. Furthermore, the matching point pairs and the forward trajectory points whose distance to the matching point pairs is less than a preset distance threshold are determined as the set of points to be optimized.

[0049] In one feasible implementation, the step of forming a matching point pair between the target first forward trajectory point and the target second forward trajectory point that have the smallest distance to each other's trajectory points may include: forming an initial point pair by combining each of the first forward trajectory points and each of the second forward trajectory points with their respective corresponding minimum distance trajectory points; finding at least one set of target initial point pairs from each set of initial point pairs that contains two identical forward trajectory points, each set of target initial point pairs including two target initial point pairs; and then selecting any one of the target initial point pairs from each set of target initial point pairs as a matching point pair.

[0050] Optionally, the step of determining the set of points to be optimized based on the matching point pairs and a preset distance threshold includes:

[0051] Step A31: Determine the connection range based on each of the matching point pairs;

[0052] Step A32: Delete the first forward trajectory point and / or the second forward trajectory point that belong to the connection range;

[0053] Step A33: The matching point pair and the forward trajectory points whose distance to the matching point pair is less than a preset distance threshold are determined as the set of points to be optimized.

[0054] In this embodiment, it should be noted that, referring to Figure 4 When determining matching point pairs, there may be unmatched isolated points within the connection range to which the matching point pair belongs. To avoid these isolated points affecting subsequent trajectory smoothing processing, they can be directly deleted. The connection range can be divided according to time, position, order, etc. For example, if the forward trajectory points forming the matching point pair in the first forward trajectory to be optimized are within the time range t1-t2, then t1-t2 can be determined as the connection range. Similarly, if the forward trajectory points forming the matching point pair in the second forward trajectory to be optimized are the n1-n2th forward trajectory points, then n1-n2 can be determined as the connection range. The specific range can be determined according to the actual situation, and this embodiment does not impose any restrictions on it.

[0055] As an example, steps A31-A33 include: determining the connection range to which each matching point pair belongs based on information such as the position, time, or order of each matching point pair; deleting the first forward trajectory point and / or the second forward trajectory point that belong to the connection range but do not form a matching point pair; and then determining the matching point pair and the forward trajectory points whose distance to the matching point pair is less than a preset distance threshold as the set of points to be optimized.

[0056] Optionally, the step of performing trajectory smoothing processing on the set of points to be optimized to obtain at least one first target forward trajectory point includes:

[0057] Step B10: Determine the midpoint of the line connecting each pair of matching points as the reference point, and replace the corresponding pair of matching points with the reference point to obtain a new set of points to be optimized;

[0058] Step B20: Perform trajectory smoothing processing on the set of points to be optimized to obtain at least one first target forward trajectory point.

[0059] In this embodiment, it should be noted that matching point pairs can be merged by taking the midpoint of the matching point pairs to achieve the merging of repeated driving trajectories.

[0060] As an example, steps B10-B20 include: referring to Figure 5 Each pair of matching points can be connected by a line, and the midpoint of the line can be used as a reference point. All matching point pairs in the set of points to be optimized can then be replaced with their respective reference points, resulting in a new set of points to be optimized. The new set of points to be optimized is shown below. Figure 6 As shown, furthermore, a trajectory smoothing algorithm can be used to solve for at least one first target forward trajectory point that can be smoothly connected to the first forward trajectory to be optimized and the second forward trajectory to be optimized, based on the points to be optimized in the set of points to be optimized.

[0061] Optionally, the step of performing trajectory smoothing processing on the set of points to be optimized to obtain at least one first target forward trajectory point includes:

[0062] Based on the preset cost function and the position information of each point to be optimized in the set of points to be optimized, the position information of the first target forward trajectory point that minimizes the cost for each point to be optimized is solved. The preset cost function includes a smooth cost sub-function, a compact cost sub-function, and a geometric similarity cost sub-function.

[0063] In this embodiment, it should be noted that the preset cost function includes a smooth cost subfunction, a compact cost subfunction, and a geometric similarity cost subfunction. The smooth cost subfunction, compact cost subfunction, and geometric similarity cost subfunction are similar to those in the prior art, and will not be described in detail again.

[0064] As an example, the position information of each point to be optimized in the set of points to be optimized can be used as the initial position. The cost can be calculated based on a preset cost function. The position information of each point to be optimized can be iteratively optimized based on the cost function to reduce the cost. By iterating repeatedly, the position information of the first target forward trajectory point that minimizes the cost can be obtained.

[0065] Optionally, the preset cost function is:

[0066]

[0067] Where n is the number of points to be optimized in the set of points to be optimized, and P k-ref Let P be the coordinates of the k-th point to be optimized. k P represents the coordinates of the first target trajectory point obtained after optimizing the k-th point to be optimized. k-1 P represents the coordinates of the first target trajectory point obtained after optimizing the (k-1)th point to be optimized. k+1 The coordinates of the first target trajectory point obtained after optimizing the (k+1)th point are α, β, and γ. k For weights;

[0068] The reference points include the earliest and latest endpoints, γ. k With P k Point and distance P k The distance between two endpoints that are relatively close to each other is positively correlated.

[0069] In this embodiment, the weight value γ of the compact cost sub-function k For variable weight terms, refer to Figure 7 The points to be optimized, in chronological order, include the points to be optimized in the first trajectory Gi, the reference point, and the points to be optimized in the second trajectory Gi+1. The points to be optimized in the first trajectory Gi originally belong to that trajectory; therefore, the smaller the adjustment, the smoother the optimized trajectory. The reference point is far from both Gi and Gi+1, located between the two trajectories. Therefore, the portion closer to the first trajectory needs a larger adjustment to connect more smoothly with it. Thus, γ is determined... k With P k Point and distance P k The distance between two endpoints that are relatively close to each other is positively correlated; that is, the closer point Pk is to the endpoint, the greater the γ. k The smaller the value of γ, the greater the difference in position before and after optimization, and the greater the optimization magnitude. The farther point Pk is from the endpoint, the greater the γ value. k The larger the value, the smaller the positional difference before and after optimization, and the smaller the optimization magnitude. Figure 7 In the middle, γk The value gradually decreases in the direction of the arrow. This ensures a smoother connection between the optimized trajectory points and the unoptimized trajectory points in the first and second unoptimized forward trajectories.

[0070] Step S30: Generate a target trajectory based on the second target trajectory points that do not belong to the set of points to be optimized in the first and second target trajectories to be optimized, as well as each of the first target trajectories to be optimized.

[0071] As an example, step S30 includes: after determining the first target forward trajectory point, determining the trajectory points in the first and second forward trajectories that do not belong to the set of points to be optimized as the second target forward trajectory points, and connecting each of the second target forward trajectory points and each of the first target forward trajectory points in chronological order to obtain the target trajectory after trajectory optimization.

[0072] In one feasible implementation, during the mapping process, not only during a single reversing operation, but also during any two adjacent forward trajectories, trajectory optimization is required to obtain a complete smooth trajectory. Therefore, after step S30, the process may further include: returning to the step of obtaining the first forward trajectory to be optimized and the second forward trajectory to be optimized, until all forward trajectories generated during the mapping process have completed trajectory optimization processing.

[0073] In this embodiment, firstly, by acquiring a first forward trajectory to be optimized and a second forward trajectory to be optimized, the forward trajectory is selected. Then, by determining a set of points to be optimized from the first and second forward trajectories to be optimized, the set of points to be optimized is subjected to trajectory smoothing processing to obtain at least one first target forward trajectory point, thus achieving a smooth connection between the first optimized forward trajectory and the second forward trajectory to be optimized. Then, by generating a target trajectory based on the second target forward trajectory points in the first and second forward trajectories to be optimized that do not belong to the set of points to be optimized, as well as each of the first target forward trajectory points, a target trajectory that does not include a reversing trajectory is generated. In this way, on the one hand, by selecting only the forward trajectory, the reversing trajectory generated during the mapping process due to factors such as the complexity of parking lot scenarios and insufficient driver experience is corrected. This ensures that the final target trajectory does not contain the reversing trajectory, only the forward trajectory, thus avoiding the reversing trajectory from affecting vehicle control during subsequent cruise and generating a more accurate and smooth target trajectory. On the other hand, by smoothly connecting two segments of the forward trajectory to be optimized, the smoothness at the connection point is effectively improved. This avoids the jitter caused by uneven local trajectories during the autonomous driving process, improving the smoothness and stability of subsequent cruise and enhancing the user experience. Therefore, compared to existing technologies that cruise based on the complete trajectory memorized during mapping, this application does not restrict the user's reversing operation during mapping. If a reversing operation occurs during mapping, an accurate and smooth target trajectory can be obtained without re-mapping. Therefore, users can flexibly correct errors or inaccurate driving paths by reversing during mapping, effectively improving the flexibility of the HPA function and enhancing the user experience.

[0074] Example 2

[0075] Furthermore, embodiments of this application provide a trajectory optimization method, referring to... Figure 8 Based on the above embodiments of this application, in the second embodiment of this application, the same or similar content as the above embodiments can be referred to the above description, and will not be repeated hereafter. Based on this, the step of obtaining the first and second forward trajectories to be optimized includes:

[0076] Step S11: Obtain the initial trajectory and perform trajectory segmentation on the initial trajectory to determine multiple forward trajectories;

[0077] In this embodiment, it should be noted that during the mapping process, the initial trajectory of the vehicle is different in different scenarios. Not all parts of the initial trajectory need to be processed and can be processed. Excessive processing may lead to vehicle collisions.

[0078] The initial trajectory is the unprocessed driving trajectory collected by the vehicle. During the vehicle's journey, it continuously collects position and time information. Based on the collected position and time information, multiple initial trajectory points can be determined, thereby obtaining an initial trajectory composed of each of the initial trajectory points.

[0079] As an example, step S11 includes: acquiring the unprocessed initial trajectory collected by the vehicle, segmenting the initial trajectory to obtain multiple forward trajectories and at least one reverse trajectory. In one feasible implementation, the reverse trajectory may be deleted.

[0080] Optionally, the step of segmenting the initial trajectory to determine multiple forward trajectories includes:

[0081] Step S111: Obtain vehicle driving information corresponding to the initial trajectory;

[0082] Step S112: Based on the vehicle driving information and the shape features of the initial trajectory, the initial trajectory is segmented to obtain multiple segmented trajectories;

[0083] Step S113: Determine multiple forward trajectories that satisfy the preset forward trajectory conditions from each of the segmented trajectories.

[0084] In this embodiment, it should be noted that the initial trajectory collected by the vehicle consists of multiple initial trajectory points containing time information, but the specific operation of the vehicle at each initial trajectory point is unknown. Therefore, it is impossible to know whether the vehicle moves forward or reverses to the next initial trajectory point at each initial trajectory point, which makes it difficult to identify the forward trajectory from the initial trajectory.

[0085] The vehicle driving information refers to information related to controlling vehicle movement, including at least one of steering wheel angle, vehicle speed, and gear information. The shape features of the initial trajectory include at least one of shape, curvature, and radius. At the transition point from forward to reverse, the vehicle driving information typically exhibits typical characteristics, such as a drop in speed to 0 or a change in gear information. The shape features of the resulting initial trajectory are usually quite distinctive; for example, it may exhibit a zigzag or "Z"-shaped trajectory, or abrupt changes in curvature. Therefore, using the vehicle driving information and the shape features of the initial trajectory as features for identifying the forward trajectory can accurately identify the forward trajectory within the initial trajectory.

[0086] The preset forward trajectory conditions can be either forward gear information or segmented trajectories arranged in chronological order with odd-numbered sequences. If the vehicle immediately shifts into reverse upon starting, a new map can be created, which is less costly and resource-intensive than trajectory optimization.

[0087] As an example, steps S111-S113 include: obtaining vehicle driving information corresponding to each trajectory point in the initial trajectory; identifying the dividing point between the forward trajectory and the reverse trajectory from the initial trajectory based on the vehicle driving information and the shape features of the initial trajectory; segmenting the initial trajectory based on the dividing point to obtain multiple segmented trajectories; and determining multiple forward trajectories that meet the preset forward trajectory conditions from each segmented trajectory.

[0088] Optionally, the initial trajectory consists of at least one initial trajectory point arranged in chronological order, the vehicle driving information includes the gear information corresponding to each of the initial trajectory points, and the shape feature of the initial trajectory includes the curvature corresponding to each of the trajectory points.

[0089] The step of segmenting the initial trajectory based on the vehicle driving information and the shape features of the initial trajectory to obtain multiple segmented trajectories includes:

[0090] Step S1121: Detect whether there is a first target initial trajectory point among the initial trajectory points with a curvature greater than a preset curvature threshold;

[0091] Step S1122: If yes, then detect whether there is a second target initial trajectory point among each of the first target initial trajectory points that has different gear information from the adjacent initial trajectory points;

[0092] Step S1123: If yes, then the initial trajectory is segmented at each of the second target initial trajectory points to obtain multiple segmented trajectories.

[0093] In this embodiment, it should be noted that when the curvature of the trajectory is large, the vehicle's direction of travel changes. Therefore, it can usually be considered that the vehicle is switching from forward to reverse or vice versa. However, there are rare cases where the vehicle malfunctions or rolls backward. If the judgment is based solely on gear information, since users frequently change gears while driving, it is common to shift into the wrong gear or not shift into the correct gear in one go. This can easily lead to incorrect trajectory segmentation due to user errors. Therefore, curvature and gear information can be combined. If the gear information also changes before and after a point with large curvature, this point can be considered the dividing point between the forward and reverse trajectories.

[0094] As an example, steps S1121-S1123 include: comparing the curvature of each initial trajectory point with a preset curvature threshold to find a first target initial trajectory point with a curvature greater than the preset curvature threshold; then, comparing the gear information of each first target initial trajectory point with the gear information of its respective adjacent initial trajectory points; determining the first target initial trajectory point with gear information different from that of the adjacent initial trajectory points as a second target initial trajectory point; and performing trajectory segmentation on the initial trajectory at each second target initial trajectory point to obtain multiple segmented trajectories.

[0095] In one feasible implementation, if there are identical or adjacent initial trajectory points among the initial trajectory points of each of the second targets, any one of them can be deleted, or they can be merged into one.

[0096] Optionally, after determining multiple forward trajectories that satisfy the preset forward trajectory conditions from each of the segmented trajectories, the method further includes:

[0097] Step S114: Detect whether there are target trajectory points with curvature greater than a preset curvature threshold in each of the forward trajectories;

[0098] Step S115: If yes, then the target forward trajectory to which the target forward trajectory point belongs is divided at the target forward trajectory point to obtain multiple forward trajectory segments.

[0099] Step S116: Determine the forward trajectory segments with an odd number of positions in the order as the forward trajectory.

[0100] In this embodiment, it should be noted that during vehicle operation, there may be instances of vehicle slippage or abnormal reversal. That is, the vehicle may be in forward gear but reverse, for example, when going uphill, insufficient forward power may cause the vehicle to slide backward. In this case, the vehicle's gear information remains unchanged, but a reversing trajectory will still be generated. Therefore, it is also necessary to detect whether there are any undetected reversing trajectories within the forward trajectory due to this situation.

[0101] As an example, steps S114-S116 include: after determining multiple forward trajectories, it is possible to further detect whether there is a target forward trajectory point with a curvature greater than a preset region threshold in each forward trajectory; after detecting the target forward trajectory point, the target forward trajectory of the target forward trajectory point can be divided into multiple forward trajectory segments from the target forward trajectory point; the forward trajectory segments are arranged in chronological order, and the forward trajectory segments with an odd number of arrangement orders are determined as forward trajectories.

[0102] Step S12: Obtain the adjacent first forward trajectory and second forward trajectory to be optimized from each of the forward trajectories.

[0103] As an example, step S12 includes: obtaining two adjacent forward trajectories from each of the forward trajectories as a first forward trajectory to be optimized and a second forward trajectory to be optimized.

[0104] In this embodiment, by segmenting the initial trajectory, the forward trajectory separated by the reversing trajectory is found, thereby optimizing the trajectory points that are repeatedly driven due to reversing. On the one hand, this avoids the vehicle repeatedly driving this meaningless trajectory during the cruise phase, which would otherwise waste time and resources. It also avoids situations where this part of the trajectory might cause vehicle vibration during autonomous driving, improving the smoothness and fluidity of vehicle driving during the cruise phase. On the other hand, it eliminates the need for trajectory smoothing of the entire initial trajectory, resulting in less computation, higher trajectory optimization efficiency, and less likelihood of over-optimization, thus reducing the probability of collisions and enhancing safety.

[0105] Example 3

[0106] Furthermore, embodiments of this application also provide a trajectory optimization device, referring to... Figure 9 The trajectory optimization device includes:

[0107] The acquisition module 10 is used to acquire the first forward trajectory to be optimized and the second forward trajectory to be optimized;

[0108] The trajectory smoothing module 20 is used to determine a set of points to be optimized from the first forward trajectory to be optimized and the second forward trajectory to be optimized, and to perform trajectory smoothing on the set of points to be optimized to obtain at least one first target forward trajectory point.

[0109] The generation module 30 is used to generate a target trajectory based on the second target trajectory points that do not belong to the set of points to be optimized in the first and second target trajectories to be optimized, as well as each of the first target trajectories to be optimized.

[0110] Optionally, the acquisition module 10 is further configured to:

[0111] Obtain an initial trajectory, and segment the initial trajectory to determine multiple forward trajectories;

[0112] Obtain the adjacent first and second forward trajectories to be optimized from each of the aforementioned forward trajectories.

[0113] Optionally, the acquisition module 10 is further configured to:

[0114] Obtain the vehicle driving information corresponding to the initial trajectory;

[0115] Based on the vehicle driving information and the shape features of the initial trajectory, the initial trajectory is segmented to obtain multiple segmented trajectories;

[0116] Multiple forward trajectories that satisfy the preset forward trajectory conditions are determined from each of the segmented trajectories.

[0117] Optionally, the initial trajectory consists of at least one initial trajectory point arranged sequentially in chronological order, the vehicle driving information includes the gear information corresponding to each of the initial trajectory points, and the shape features of the initial trajectory include the curvature corresponding to each of the trajectory points; the acquisition module 10 is further configured to:

[0118] Detect whether there is a first target initial trajectory point among the initial trajectory points with a curvature greater than a preset curvature threshold;

[0119] If so, then detect whether there is a second target initial trajectory point among each of the first target initial trajectory points that has different gear information from the adjacent initial trajectory points;

[0120] If so, the initial trajectory is segmented at each of the second target initial trajectory points to obtain multiple segmented trajectories.

[0121] Optionally, after the operation of determining multiple forward trajectories that satisfy the preset forward trajectory conditions from each of the segmented trajectories, the acquisition module 10 is further configured to:

[0122] Detect whether there are target trajectory points with curvature greater than a preset curvature threshold in each of the aforementioned forward trajectories;

[0123] If so, the target forward trajectory to which the target forward trajectory point belongs is divided at the target forward trajectory point to obtain multiple forward trajectory segments;

[0124] The forward trajectory segment with an odd number of positions is determined as the forward trajectory.

[0125] Optionally, the trajectory smoothing processing module 20 is further configured to:

[0126] Calculate the distance between each first forward trajectory point in the first forward trajectory to be optimized and each second forward trajectory point in the second forward trajectory to be optimized;

[0127] The first forward trajectory point and the second forward trajectory point of the target, which have the smallest distances to each other's trajectory points, are formed into a matching point pair;

[0128] The set of points to be optimized is determined based on the matching point pairs and the preset distance threshold.

[0129] Optionally, the trajectory smoothing processing module 20 is further configured to:

[0130] The connection range is determined based on each of the aforementioned matching point pairs;

[0131] Delete the first forward trajectory point and / or the second forward trajectory point that belong to the connection range;

[0132] The matching point pairs and the forward trajectory points whose distance to the matching point pairs is less than a preset distance threshold are determined as the set of points to be optimized.

[0133] Optionally, the trajectory smoothing processing module 20 is further configured to:

[0134] The midpoint of the line connecting each pair of matching points is determined as the reference point. The reference point is used to replace the corresponding pair of matching points to obtain a new set of points to be optimized.

[0135] The set of points to be optimized is subjected to trajectory smoothing processing to obtain at least one first target forward trajectory point.

[0136] Optionally, the trajectory smoothing processing module 20 is further configured to:

[0137] Based on the preset cost function and the position information of each point to be optimized in the set of points to be optimized, the position information of the first target forward trajectory point that minimizes the cost for each point to be optimized is solved. The preset cost function includes a smooth cost sub-function, a compact cost sub-function, and a geometric similarity cost sub-function.

[0138]

[0139] Where n is the number of points to be optimized in the set of points to be optimized, and P k-ref Let P be the coordinates of the k-th point to be optimized. k P represents the coordinates of the first target trajectory point obtained after optimizing the k-th point to be optimized. k-1 P represents the coordinates of the first target trajectory point obtained after optimizing the (k-1)th point to be optimized. k+1 The coordinates of the first target trajectory point obtained after optimizing the (k+1)th point are α, β, and γ. k For weights;

[0140] The reference points include the earliest and latest endpoints, γ. k With P k Point and distance P k The distance between two endpoints that are relatively close to each other is positively correlated.

[0141] The trajectory optimization device provided by this invention, employing the trajectory optimization method in the above embodiments, solves the technical problem of poor performance of existing automatic parking systems based on memory. Compared with the prior art, the beneficial effects of the trajectory optimization device provided by this invention are the same as those of the trajectory optimization method provided in the above embodiments, and other technical features of this trajectory optimization device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0142] Example 4

[0143] Furthermore, embodiments of the present invention provide an electronic device, the electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the trajectory optimization method in the above embodiments.

[0144] The following is for reference. Figure 10 The diagram illustrates a structural schematic of an electronic device suitable for implementing embodiments of the present disclosure. The electronic devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as Bluetooth headsets, mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 10 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0145] like Figure 10 As shown, an electronic device may include a processing unit (such as a central processing unit, graphics processing unit, etc.) that can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) or a program loaded from a storage device into random access memory (RAM). The RAM also stores various programs and arrays required for the operation of the electronic device. The processing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0146] Typically, the following systems can be connected to the I / O interface: input devices including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices including, for example, magnetic tapes, hard disks, etc.; and communication devices. Communication devices allow electronic devices to communicate wirelessly or wiredly with other devices to exchange arrays. Although electronic devices with various systems are shown in the figures, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems may be implemented alternatively.

[0147] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processing device, it performs the functions defined above in the methods of embodiments of this disclosure.

[0148] The electronic device provided by this invention, employing the trajectory optimization method in the above embodiments, solves the technical problem of poor performance of existing automatic parking systems based on memory. Compared with the prior art, the beneficial effects of the electronic device provided by this invention are the same as those of the trajectory optimization method provided in the above embodiments, and other technical features of this electronic device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0149] It should be understood that various parts of this disclosure can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0150] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0151] Example 5

[0152] Furthermore, this embodiment provides a computer-readable storage medium having computer-readable program instructions stored thereon, the computer-readable program instructions being used to execute the trajectory optimization method in the above embodiment.

[0153] The computer-readable storage medium provided in this embodiment of the invention may be, for example, a USB flash drive, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or any combination thereof. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0154] The aforementioned computer-readable storage medium may be included in an electronic device or may exist independently without being assembled into an electronic device.

[0155] The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by an electronic device, the electronic device causes the electronic device to: acquire a first forward trajectory to be optimized and a second forward trajectory to be optimized; determine a set of points to be optimized from the first forward trajectory to be optimized and the second forward trajectory to be optimized; perform trajectory smoothing processing on the set of points to be optimized to obtain at least one first target forward trajectory point; and generate a target trajectory based on the second target forward trajectory point in the first forward trajectory to be optimized and the second forward trajectory to be optimized that does not belong to the set of points to be optimized, as well as each of the first target forward trajectory points.

[0156] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0157] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0158] The modules described in the embodiments of this disclosure can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0159] The computer-readable storage medium provided by this invention stores computer-readable program instructions for executing the above-described trajectory optimization method, thus solving the technical problem of poor memory performance in existing automatic parking systems. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this invention are the same as those of the trajectory optimization method provided in the above-described embodiments, and will not be repeated here.

[0160] Example 6

[0161] Furthermore, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the trajectory optimization method as described above.

[0162] The computer program product provided in this application solves the technical problem of poor performance of existing automatic parking memory. Compared with the prior art, the beneficial effects of the computer program product provided in this embodiment are the same as those of the trajectory optimization method provided in the above embodiments, and will not be repeated here.

[0163] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent scope of this application.

Claims

1. A trajectory optimization method, characterized in that, The trajectory optimization method includes the following steps: Obtain the first and second forward trajectories to be optimized; Calculate the distance between each first forward trajectory point in the first forward trajectory to be optimized and each second forward trajectory point in the second forward trajectory to be optimized; The first forward trajectory point and the second forward trajectory point of the target, which have the smallest distances to each other's trajectory points, are formed into a matching point pair; Based on the matching point pairs and the preset distance threshold, a set of points to be optimized is determined. The midpoint of the line connecting each pair of matching points is determined as a reference point. The reference point is used to replace the corresponding matching point pair to obtain a new set of points to be optimized. The new set of points to be optimized is subjected to trajectory smoothing processing to obtain at least one first target forward trajectory point; Based on the second target forward trajectory points that do not belong to the set of points to be optimized in the first and second forward trajectories to be optimized, and each of the first target forward trajectory points, a target trajectory that does not include a reversing trajectory is generated.

2. The trajectory optimization method as described in claim 1, characterized in that, The steps of obtaining the first and second forward trajectories to be optimized include: Obtain an initial trajectory, and segment the initial trajectory to determine multiple forward trajectories; Obtain the adjacent first and second forward trajectories to be optimized from each of the aforementioned forward trajectories.

3. The trajectory optimization method as described in claim 2, characterized in that, The step of segmenting the initial trajectory to determine multiple forward trajectories includes: Obtain the vehicle driving information corresponding to the initial trajectory; Based on the vehicle driving information and the shape features of the initial trajectory, the initial trajectory is segmented to obtain multiple segmented trajectories; Multiple forward trajectories that satisfy the preset forward trajectory conditions are determined from each of the segmented trajectories.

4. The trajectory optimization method as described in claim 3, characterized in that, The initial trajectory consists of at least one initial trajectory point arranged in chronological order, the vehicle driving information includes the gear information corresponding to each of the initial trajectory points, and the shape features of the initial trajectory include the curvature corresponding to each of the trajectory points. The step of segmenting the initial trajectory based on the vehicle driving information and the shape features of the initial trajectory to obtain multiple segmented trajectories includes: Detect whether there is a first target initial trajectory point among the initial trajectory points with a curvature greater than a preset curvature threshold; If so, then detect whether there is a second target initial trajectory point among each of the first target initial trajectory points that has different gear information from the adjacent initial trajectory points; If so, the initial trajectory is segmented at each of the second target initial trajectory points to obtain multiple segmented trajectories.

5. The trajectory optimization method as described in claim 3, characterized in that, After the step of determining multiple forward trajectories that satisfy the preset forward trajectory conditions from each of the segmented trajectories, the method further includes: Detect whether there are target trajectory points with curvature greater than a preset curvature threshold in each of the aforementioned forward trajectories; If so, the target forward trajectory to which the target forward trajectory point belongs is divided at the target forward trajectory point to obtain multiple forward trajectory segments; The forward trajectory segment with an odd number of positions is determined as the forward trajectory.

6. The trajectory optimization method as described in claim 5, characterized in that, The step of determining the set of points to be optimized based on the matching point pairs and the preset distance threshold includes: The connection range is determined based on each of the aforementioned matching point pairs; Delete the first forward trajectory point and / or the second forward trajectory point that belong to the connection range; The matching point pairs and the forward trajectory points whose distance to the matching point pairs is less than a preset distance threshold are determined as the set of points to be optimized.

7. The trajectory optimization method as described in claim 6, characterized in that, The step of performing trajectory smoothing processing on the set of points to be optimized to obtain at least one first target forward trajectory point includes: Based on the preset cost function and the position information of each point to be optimized in the set of points to be optimized, the position information of the first target forward trajectory point that minimizes the cost for each point to be optimized is solved. The preset cost function includes a smooth cost sub-function, a compact cost sub-function, and a geometric similarity cost sub-function.

8. The trajectory optimization method as described in claim 7, characterized in that, The preset cost function is: Where n is the number of points to be optimized in the set of points to be optimized, and P k-ref Let P be the coordinates of the k-th point to be optimized. k P represents the coordinates of the first target trajectory point obtained after optimizing the k-th point to be optimized. k-1 P represents the coordinates of the first target trajectory point obtained after optimizing the (k-1)th point to be optimized. k+1 The coordinates of the first target trajectory point obtained after optimizing the (k+1)th point are α, β, and γ. k For weights; The reference points include the earliest and latest endpoints, γ. k With P k Point and distance P k The distance between two endpoints that are relatively close to each other is positively correlated.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the steps of the trajectory optimization method according to any one of claims 1 to 8.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and the computer-readable storage medium stores a program that implements the trajectory optimization method. The program that implements the trajectory optimization method is executed by a processor to implement the steps of the trajectory optimization method as described in any one of claims 1 to 8.

Images