Vehicle control method and device, storage medium and computer equipment

Abstract

The invention discloses a vehicle control method and device, a storage medium and computer equipment. Comprising the steps of constructing an envelope circle of a vehicle according to structure data of the vehicle; according to the obstacle data of the target road where the vehicle is located, the safe distance between the center of an envelope circle of the vehicle and an obstacle in the target road and the avoidance area of the obstacle are determined; predicting a forward path of the vehicle on the target road according to a comparison result of the safe distance and the radius of the envelope circle and the avoidance area; controlling the vehicle to travel on the target road according to the advancing path. According to the embodiment of the invention, a vehicle collision detection problem is converted into a problem that the vehicle envelope circle is restrained in the convex polygon, the complexity of path planning solution is effectively reduced, accurate vehicle advancing control can be timely performed even if an upstream environment sensing blind area exists on a target road, the braking waiting time is shortened, and the driving safety of a vehicle is improved. The continuity of an automatic driving path is improved, and the requirements of a driver for automatic driving time and safety are met.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a vehicle control method, apparatus, storage medium, and computer equipment based on a convex polygonal road structure. Background Technology

[0002] With the development of artificial intelligence technology and the continuous progress of the vehicle industry, autonomous driving technology is becoming increasingly mature. In narrow road scenarios, due to the uncertainty of upstream blind spots on the target road, the system cannot effectively plan routes in a timely manner to address these blind spots.

[0003] In related technologies, when the current lane is near the end of the downstream section, it can only passively brake and decelerate, and can only continue to plan the upstream route after passing the insurmountable physical boundary. However, the automatic driving method of braking and decelerating while waiting can easily cause traffic congestion and affect the passage of other vehicles. Moreover, if other vehicles cannot anticipate the unreasonable braking of the vehicle in front in time, it is easy to cause rear-end collisions, which cannot meet the driver's requirements for the time and safety of automatic driving. Summary of the Invention

[0004] In view of this, this application provides a vehicle control method, apparatus, storage medium, and computer device based on a convex polygonal road structure, to solve the problems described in the background art.

[0005] According to one aspect of this application, a vehicle control method based on a convex polygonal road structure is provided, comprising:

[0006] Construct the vehicle's envelope circle based on the vehicle's structural data;

[0007] Based on the obstacle data of the target road where the vehicle is located, determine the safe distance between the center of the vehicle's envelope circle and the obstacles in the target road, as well as the obstacle avoidance zone;

[0008] Based on the comparison results of the safe distance and the radius of the envelope circle and the avoidance area, the vehicle's forward path on the target road is predicted;

[0009] The previous path control system guides the vehicle to travel on the target road.

[0010] Furthermore, based on obstacle data of the target road where the vehicle is located, the safe distance between the center of the vehicle's envelope circle and obstacles in the target road is determined, including:

[0011] Based on obstacle data, at least one set of obstacle points for the target road is determined, wherein the set of obstacle points includes less than or equal to a preset number of points to be processed, and the points to be processed are adjacent obstacle points in a distance sequence, which is obtained by sorting the obstacle points according to their x-coordinates;

[0012] Triangulation of the obstacle point set yields a Deloni triangulation network.

[0013] Connect the circumcenters of adjacent triangles in the Deloni triangulation to generate a Veno diagram, wherein the Veno diagram includes at least one Veno polygon;

[0014] The safe distance is determined based on the distance between the grid within the Vino polygon and the center point of the Vino polygon.

[0015] Furthermore, based on obstacle data of the target road where the vehicle is located, the obstacle avoidance zone is determined, including:

[0016] Determine the obstacle outline based on obstacle data;

[0017] Determine the tangent line between the boundary of the obstacle near the envelope circle and the bulging circle containing the center of the envelope circle in the obstacle profile;

[0018] The avoidance zone is generated based on the tangent.

[0019] Furthermore, based on the comparison results of the safe distance and the radius of the envelope circle, and the avoidance area, the vehicle's forward path on the target road is predicted, including:

[0020] Based on the obstacle avoidance zone, a heuristic search algorithm is used to search for the vehicle's initial path on the target road;

[0021] The initial path is updated based on the comparison results to obtain the forward path.

[0022] Furthermore, based on the obstacle avoidance area, a heuristic search algorithm is used to search for the vehicle's initial path on the target road, including:

[0023] The target road is discretized based on obstacle data to obtain a grid map of the target road;

[0024] Determine the starting and target nodes for the vehicle's path search in the grid map;

[0025] Based on the obstacle avoidance zone, a heuristic function is used to calculate the path cost of the starting node and the target node;

[0026] The initial path is obtained by performing a discrete search based on the path cost.

[0027] Furthermore, vehicle control methods based on convex polygon road structures also include:

[0028] The vehicle dynamics equations are established based on the vehicle's pose, and then discretized to obtain dynamic nonlinear constraints.

[0029] Based on the avoidance area and the path of travel, construct linear constraints for obstacles;

[0030] Based on the nonlinear constraints of dynamics and the linear constraints of obstacles, a pre-defined optimization model is established;

[0031] The forward path is optimized based on a preset optimization model.

[0032] Furthermore, vehicle control methods based on convex polygon road structures also include:

[0033] Construct multiple polynomial curves based on preset factors;

[0034] The obstacle avoidance area is expanded based on multiple polynomial curves.

[0035] According to another aspect of this application, a vehicle control device based on a convex polygonal road structure is provided, comprising:

[0036] The module is used to construct the vehicle's envelope circle based on the vehicle's structural data; and,

[0037] Based on the obstacle data of the target road where the vehicle is located, determine the safe distance between the center of the vehicle's envelope circle and the obstacles in the target road, as well as the obstacle avoidance zone;

[0038] The path generation module is used to predict the vehicle's forward path on the target road based on the comparison results of the safety distance and the radius of the envelope circle and the avoidance area.

[0039] The control module is used to control the vehicle's movement on the target road according to the previous path.

[0040] Further, the determination module is specifically used to determine at least one set of obstacle points for the target road based on obstacle data. The set of obstacle points includes a number of points to be processed that are less than or equal to a preset number. The points to be processed are adjacent obstacle points in a distance sequence, which is obtained by sorting the obstacle points according to their x-coordinates. The obstacle point set is then triangulated to obtain a Deloni triangulation network. The centers of the circumcircles of adjacent triangles in the Deloni triangulation network are connected to generate a Vinno diagram, which includes at least one Vinno polygon. A safe distance is determined based on the distance between the grid within the Vinno polygon and the center point of the Vinno polygon.

[0041] Furthermore, the determination module is specifically used to determine the obstacle outline based on obstacle data; determine the tangent line between the obstacle boundary near the envelope circle and the expansion circle containing the center of the envelope circle in the obstacle outline; and generate the avoidance area based on the tangent line.

[0042] Furthermore, the path generation module is specifically used to search for the initial path of the vehicle on the target road using a heuristic search algorithm based on the obstacle avoidance area; and to update the initial path based on the comparison results to obtain the forward path.

[0043] Furthermore, the path generation module is specifically used to discretize the target road based on obstacle data to obtain a grid map of the target road; determine the starting node and target node for the vehicle's path search in the grid map; calculate the path cost of the starting node and target node using a heuristic function based on the obstacle avoidance area; and perform a discrete search based on the path cost to obtain the initial path.

[0044] Furthermore, the vehicle control device based on the convex polygon road structure also includes:

[0045] The optimization module is used to establish vehicle dynamics equations based on the vehicle's pose, and to discretize the vehicle dynamics equations to obtain dynamic nonlinear constraints; to construct obstacle linear constraints based on the avoidance area and the forward path; to establish a preset optimization model based on the dynamic nonlinear constraints and obstacle linear constraints; and to optimize the forward path based on the preset optimization model.

[0046] Furthermore, the vehicle control device based on the convex polygon road structure also includes:

[0047] The extension module is used to construct multiple polynomial curves based on preset factors; and to expand the obstacle avoidance area based on the multiple polynomial curves.

[0048] According to another aspect of this application, a computer-readable storage medium is provided having a program or instructions stored thereon, which, when executed by a processor, implement the steps of the vehicle control method based on a convex polygon road structure described above.

[0049] According to another aspect of this application, a computer device is provided, including at least one processor coupled to a memory storing a computer program that runs on the processor, characterized in that the processor executes the program to implement the steps of the above-described vehicle control method based on a convex polygon road structure.

[0050] By employing the aforementioned technical solution, the likelihood of a vehicle colliding with an obstacle is detected based on a comparison between the radius of the vehicle's outer contour envelope circle and the safe distance between the center of the vehicle's envelope circle and obstacles in the target road. If the obstacle avoidance requirement is met, the avoidance zone of the obstacle in the target road is used to determine the vehicle's drivable range. The vehicle's forward path on the target road is predicted based on its drivable range and the collision detection results. Finally, the vehicle is controlled to travel on the target road according to the predicted forward path. This transforms the vehicle collision detection problem into a problem where the vehicle's outer contour envelope circle is constrained within a convex polygon, simplifying the mathematical description of the vehicle's outer contour envelope circle collision detection problem, effectively reducing the complexity of path planning, and thus enabling safe, efficient, and robust narrow-lane passage trajectories under time constraints. Even in target roads with upstream environmental perception blind spots, timely and accurate vehicle movement control can be made, reducing braking waiting time, improving the continuity of the autonomous driving path, and meeting the driver's requirements for autonomous driving time and safety.

[0051] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0052] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0053] Figure 1 A flowchart illustrating the vehicle control method based on a convex polygonal road structure provided in an embodiment of this application is shown.

[0054] Figure 2 One of the schematic diagrams of the avoidance area according to an embodiment of this application is shown;

[0055] Figure 3 A second schematic diagram of the avoidance area according to an embodiment of this application is shown;

[0056] Figure 4 The third schematic diagram of the avoidance area according to an embodiment of this application is shown;

[0057] Figure 5 The fourth schematic diagram of the avoidance area according to an embodiment of this application is shown;

[0058] Figure 6 The fifth schematic diagram of the avoidance area according to an embodiment of this application is shown;

[0059] Figure 7This is shown as the sixth schematic diagram of the avoidance area according to an embodiment of this application;

[0060] Figure 8 The seventh schematic diagram of the avoidance area according to an embodiment of this application is shown;

[0061] Figure 9 A schematic diagram of the optimized forward path provided in an embodiment of this application is shown;

[0062] Figure 10 A structural block diagram of a vehicle control device based on a convex polygon road structure provided in an embodiment of this application is shown. Detailed Implementation

[0063] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0064] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0065] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this application means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “attached” to another element, it can be directly connected or attached to the other element, or there may be intermediate elements. Furthermore, “connected” or “attached” as used herein can include wireless connections or wireless interconnections. The term “and / or” as used herein includes all or any unit and all combinations of one or more associated listed items.

[0066] Exemplary embodiments according to this application will now be described in more detail with reference to the accompanying drawings. However, these exemplary embodiments may be implemented in many different forms and should not be construed as being limited to the embodiments set forth herein. It should be understood that these embodiments are provided so that the disclosure of this application is thorough and complete, and that the concept of these exemplary embodiments is fully conveyed to those skilled in the art.

[0067] This embodiment provides a vehicle control method based on a convex polygon road structure, such as... Figure 1 As shown, the method includes:

[0068] Step 110: Construct the vehicle's envelope circle based on the vehicle's structural data.

[0069] The vehicle's structural data includes at least one of the following: vehicle dimensions, vehicle weight, load capacity, chassis structure, fuel type, vehicle type and specifications, and configured vehicle safety equipment. Of course, it may also include other information that can be used to distinguish the vehicle structure, which will not be listed one by one in this application embodiment.

[0070] In specific application scenarios, step 110, which involves constructing the vehicle's envelope circle based on the vehicle's structural data, specifically includes:

[0071] Step 111: Determine the circumscribed rectangle of the vehicle's outline based on the vehicle's effective tire length and effective tire width.

[0072] Step 112: Generate at least one envelope circle on the circumscribed rectangle of the contour according to the preset circle radius.

[0073] The envelope circle is located at the vertex of the circumscribed rectangle of the contour, and the center of the envelope circle is located inside the circumscribed rectangle of the contour.

[0074] In this embodiment, by determining the circumscribed rectangle of the vehicle's outline, the spatial range occupied by the vehicle can be accurately determined. An envelope circle at each vertex is then generated based on the circumscribed rectangle. This envelope circle describes the maximum turning radius of the vehicle when turning, ensuring the accuracy of vehicle collision detection on target roads with convex polygonal curves. This helps ensure that the vehicle does not collide with obstacles or other vehicles when turning, improving driving safety.

[0075] It is understood that the preset circle radius can be reasonably set according to the required detection accuracy and vehicle model, and this application embodiment does not impose specific limitations. The number of envelope circles that can be drawn for a vehicle can be determined according to its different structural shapes, and the number of envelope circles that can be drawn for vehicles with different shapes can be the same or different. Figure 2 As shown, two envelope circles with origins O1 and O2 are formed based on the circumscribed rectangle of the vehicle's outer contour.

[0076] Step 120: Based on the obstacle data of the target road where the vehicle is located, determine the safe distance between the center of the vehicle's envelope circle and the obstacles in the target road, as well as the obstacle avoidance area.

[0077] It is understandable that obstacles in the target road include not only curbs and barriers that serve as road boundaries, but also dynamic and static individuals, such as walls, bollards, temporarily parked vehicles, and pedestrians. The target road can be a regular traffic road, or a driving road in complex traffic scenarios such as parking scenarios or narrow passage scenarios.

[0078] In this embodiment, determining the safe distance between the center of the vehicle's envelope circle and obstacles on the target road using obstacle data ensures sufficient clearance between the vehicle and obstacles to prevent collisions or scrapes. Determining the avoidance zone using obstacle data from the target road clarifies the vehicle's drivable range, allowing the system to develop more personalized feasible paths while ensuring the vehicle can avoid obstacles, thus maintaining smooth and stable driving.

[0079] The vehicles in this application embodiment can be cars, trucks, motorcycles, public vehicles, agricultural vehicles, recreational vehicles, amusement park vehicles, construction equipment vehicles, trams, golf carts, etc., and this application embodiment does not make specific limitations.

[0080] The vehicle may include various subsystems, such as a mobility system, a sensing system, a control system, one or more peripheral devices, a power supply, a computer system, and a user interface. The vehicle may also include more or fewer subsystems, and each subsystem may include multiple components. Furthermore, each subsystem and component of the vehicle can be interconnected via wired or wireless means. The mobility system may include components for providing powered motion to the vehicle. In one embodiment, the mobility system may include an engine, a transmission, an energy source, and wheels / tires. The engine may be an internal combustion engine, an electric motor, an air-compressed engine, or other combinations of engines; for example, a hybrid engine consisting of a gasoline engine and an electric motor, or a hybrid engine consisting of an internal combustion engine and an air-compressed engine. The engine converts the energy source into mechanical energy. The sensing system may include several sensors for sensing information about the environment surrounding the vehicle.

[0081] For example, energy sources may include gasoline, diesel, other petroleum-based fuels, propane, other compressed gas-based fuels, ethanol, solar panels, batteries, and other sources of electricity. Energy sources may also provide power to other systems within the vehicle. Sensing systems may include positioning systems, such as GPS, BeiDou, or other positioning systems, inertial measurement units (IMUs), radar components, laser rangefinders, and camera components. Sensing systems may also include sensors for the monitored internal systems of the vehicle, such as in-vehicle air quality monitors, fuel gauges, oil temperature gauges, etc. Sensor data from one or more of these sensors can be used to detect objects and their corresponding characteristics (position, shape, orientation, speed, etc.).

[0082] In one embodiment, step 120, namely determining the safe distance between the center of the vehicle's envelope circle and the obstacles in the target road based on the obstacle data of the target road where the vehicle is located, specifically includes:

[0083] Step 121-1: Determine at least one set of obstacle points for the target road based on obstacle data.

[0084] The obstacle point set includes a preset number of points to be processed. This preset number can be reasonably set based on the quality of the generated triangulated network. For example, the preset number can be set to 3, so that each obstacle point set contains no more than 3 obstacle points, facilitating the formation of a refined Delonix triangulated network. The points to be processed are adjacent obstacle points in a distance sequence, which is obtained by sorting the obstacle points according to their x-coordinates. Specifically, obstacle points can be pixels or obstacle contour points extracted from satellite remote sensing images or aerial photographs. They can also be detected by vehicle sensors (e.g., cameras or LiDAR), using computer vision or point cloud processing algorithms to detect and extract scattered obstacle points.

[0085] Step 121-2: Triangulate the obstacle point set to obtain the Deloni triangulation.

[0086] Delaunay triangulation is a common data structure and algorithm in computational geometry used to connect a set of discrete points into a grid of triangles that is as irregular as possible. In a Delaunay triangulation, triangles are adjacent to each other and do not overlap, and the circumcircle of each triangle does not contain any other points; that is, no point is inside the circumcircle of a triangle.

[0087] In this embodiment, by grouping obstacle points that are close together, a mesh of non-intersecting, gap-free triangles can be generated based on the obstacle point set, thereby reducing the number of repeated triangle optimizations and improving the generation efficiency of the Deloni triangulation. Furthermore, using obstacle points as constraint points in the Deloni triangulation ensures that the final generated triangular mesh approximates the boundaries of obstacles in the target road, allowing triangular meshes to be generated within the obstacle boundary contours and preventing triangles from exceeding or being missing from the obstacle boundaries. Moreover, due to the characteristics of the Deloni triangulation, it can include as many triangles connecting obstacle points as possible, ensuring good connectivity between generated obstacles, improving the accuracy of obstacle description, and thus enabling the Deloni triangulation to better reflect the shape and contour of obstacles on the actual road.

[0088] Step 121-3: Connect the circumcenters of adjacent triangles in the Delonix triangulation to generate the Veno diagram.

[0089] A Voronoi diagram includes at least one Voronoi polygon, which can be considered as the boundary of an obstacle. Voronoi diagrams, also known as Thiessen polygons, divide a plane (or higher-dimensional space) into regions, each containing a point that is the unique nearest neighbor of all points within that region. Through Voronoi diagrams, the nearest neighbor relationships between points and the neighborhood region of each point can be determined.

[0090] Step 121-4: Determine the safety distance based on the distance between the grid within the Vino polygon and the center point of the Vino polygon.

[0091] In this embodiment, by triangulation and generating a Veno map, the spatial relationships of obstacle points can be analyzed in detail. The Veno map is rasterized, and by measuring the distance between the grid cells within the Veno polygons and the center point of the Veno polygons, a safe distance can be determined. This safe distance represents the minimum distance between a vehicle and an obstacle on the road, allowing for the assessment of the likelihood of a collision and the planning of paths to avoid collisions or scrapes.

[0092] For specific examples, such as Figure 2 As shown, circles O1 and O2 enclose the vehicle profile. All given obstacle points are sorted in ascending order by their x-coordinates. The sorted obstacle points are then divided into subsets of no more than three points each (obstacle point sets), and these subsets are triangulated. The subsets are then merged sequentially to obtain a Delonix triangulation. Connecting the centers of the circumcircles of adjacent triangles in the Delonix triangulation yields a Veno diagram. By definition, all grid cells within a Veno polygon are closer to their own Veno center than all grid cells within a Veno polygon are closer to other Veno centers. Therefore, by calculating the distance from each grid cell within a Veno polygon to its center using a row-by-row scanning algorithm, the distance from the grid cell to the nearest obstacle point, i.e., the critical safe distance for collision, can be obtained.

[0093] In one embodiment, step 120, namely determining the obstacle avoidance zone based on obstacle data of the target road where the vehicle is located, specifically includes:

[0094] Step 122-1: Determine the obstacle outline based on obstacle data.

[0095] Specifically, the obstacle outline can be extracted from satellite remote sensing imagery or aerial photography imagery, or it can be obtained by connecting collected obstacle points, or it can be determined using the Deloni triangulation network constructed in the above embodiments. This application does not specifically limit the method for determining the obstacle outline.

[0096] Step 122-2: Determine the tangent line between the obstacle boundary near the envelope circle and the expansion circle containing the center of the envelope circle in the obstacle profile.

[0097] Step 122-3: Generate the avoidance area based on the tangent.

[0098] In this embodiment, the edges of the obstacle contour are determined by connecting obstacle points, and obstacles close to the envelope circle are identified. The tangent lines intersecting the boundary of the obstacle near the envelope circle and the expanded circle containing the center of the envelope circle are calculated. Since the half-space formed by the intersection of the tangent lines and the edge of the obstacle contour can avoid obstacles, connecting the tangent lines obtained based on different obstacles forms an avoidance zone. The space enclosed by the avoidance zone is the drivable space where the vehicle can avoid obstacles. This provides precise avoidance zones for irregularly shaped roads, ensuring that the planned path meets the requirements for axial movement, significantly improving the avoidance and real-time adaptability of multiple irregularly distributed obstacles.

[0099] For example, such as Figure 3 As shown, the local rectangular frame A1A2A3A4 of the target road includes five obstacles D1, D2, D3, D4, and D5. An expanded circle centered at point O1 intersects obstacle D1, forming a tangent B1B2. This tangent B1B2 creates a half-space to isolate obstacle D1 and part of obstacle D2 from the envelope circle centered at O1. Figure 4 As shown, consider the portion of obstacle D2 within the rectangular frame A1A2A3A4 that is not isolated by the half-space formed by tangents B1B2. The expanded circle centered at point O1 intersects obstacle D2 to form tangent B2B3. The remaining portion of obstacle D2 and part of obstacle D3 are isolated from the envelope circle centered at point O1 based on the half-space formed by tangent B2B3. Figure 5 As shown, consider the portion of obstacle D3 within the rectangular frame A1A2A3A4 that is not isolated by the half-space formed by tangents B2B3. The expanded circle centered at point O1 intersects obstacle D5, forming tangent B5B6. Based on tangent B5B6, a half-space is formed to isolate obstacle D5, part of obstacle D4, and the envelope circle centered at point O1. For example... Figure 6 As shown, considering the unisolated obstacles within the rectangular frame A1A2A3A4, the expanded circle centered at point O1 intersects obstacle D4 to form a tangent line B4B5. Based on tangent line B4B5, a half-space is constructed to isolate part of obstacle D4 from the envelope circle centered at point O1. For example... Figure 7As shown, considering the unisolated obstacles within the rectangular frame A1A2A3A4, the expanded circle centered at point O1 intersects obstacle D3, forming a tangent B3B4. This tangent B3B4 creates a half-space to partially isolate obstacle D3 from the envelope circle centered at O1. Tangents B1B2, B2B3, B3B4, B4B5, and B5B6 form a convex polygon B1B2B3B4B5B6 (the avoidance area).

[0100] Similarly, such as Figure 8 As shown, tangents C1C2, C2C3, C3C4, C4C5, and C5C6 based on circle O2 are obtained using the same method. These tangents form a convex polygon C1C2C3C4C5C6 (avoidance area). Circles O1 and O2 enclose the vehicle's outer contour, transforming the vehicle collision detection problem into a collision detection problem involving enclosing circles O1 and O2. Processing enclosing circles O1 and O2 respectively yields convex polygons B1B2B3B4B5B6 and C1C2C3C4C5C6 that isolate the passable area and obstacle area. Comparing the relative positions of the convex polygons and obstacles reveals that the enclosing circle has high space utilization in the axial direction, meeting the requirement of axial movement during the narrow passage trajectory optimization process.

[0101] Furthermore, in one embodiment, before steps 122-3, the vehicle control method based on the convex polygonal road structure further includes: constructing a polynomial curve based on preset factors; and expanding the obstacle avoidance area based on the polynomial curve.

[0102] In this embodiment, after confirming the obstacle boundary near the envelope circle, multiple polynomial curves can be flexibly constructed based on preset factors such as dynamic parameters, obstacle position, and shape to adapt to different obstacles and road conditions. The obstacle boundary is then expanded according to these polynomial curves. This expands the obstacle avoidance area, ensuring that the vehicle has sufficient space to turn and adjust while proceeding along the planned path, reducing the risk of collisions and accidents.

[0103] For example, such as Figure 8 As shown, a series of fifth-order polynomial curves L1, L2, L3, and L4 are constructed by comprehensively considering factors such as safety, comfort, and narrow passage passability. Each fifth-order polynomial curve is scored. The fifth-order polynomial curve with the highest score is selected as the basis for obstacle boundary expansion and is then processed to obtain the expanded obstacle boundary of obstacle E1.

[0104] Step 130: Based on the comparison results of the safety distance and the radius of the envelope circle and the avoidance area, predict the vehicle's forward path on the target road.

[0105] In this embodiment, the potential collision between the vehicle and an obstacle can be determined by comparing the safe distance and the radius of the enveloping circle. If the safe distance is greater than the circle radius, the vehicle is considered to have avoided a collision. Thus, the vehicle collision detection problem is transformed into a comparison problem between the distance (safe distance) from the grid cell containing the center of the enveloping circle to the nearest obstacle and the circle radius. This simplifies the mathematical description of the vehicle's outer contour enveloping circle collision detection problem and significantly improves the efficiency of vehicle collision detection. Simultaneously, by combining the obstacle avoidance zone in the target road with the predicted vehicle path, more path planning possibilities are provided while ensuring the presence of obstacles, which helps improve the quality of path planning and achieve high-quality, fast path planning functionality.

[0106] In one embodiment, step 130, namely predicting the vehicle's path on the target road based on the comparison results of the safety distance and the radius of the envelope circle and the avoidance area, specifically includes:

[0107] Step 131: Based on the obstacle avoidance area, use a heuristic search algorithm to search for the initial path of the vehicle on the target road.

[0108] The heuristic search algorithms include: the traditional A* algorithm, Dijkstra's algorithm, and the hybrid A* algorithm. Preferably, the hybrid A* algorithm, as an improvement on the traditional A* algorithm, considers the vehicle's attitude angles during the search process, avoiding the blind search problem of Dijkstra's algorithm, and ensuring that the searched path meets the constraints of vehicle dynamics.

[0109] Specifically, the hybrid A* algorithm primarily aims to minimize the total cost, and its formula is expressed as follows:

[0110] f(n) = p(n) + q(n),

[0111] In the formula, f(n) represents the cost function from the starting node to the target node, p(n) represents the actual cost of the vehicle from the starting path point to the target node, and (n) represents the estimated cost of the vehicle from node n to the target node via the optimal path.

[0112] In this embodiment, a heuristic search algorithm is first used to search for the initial path of the vehicle on the target road based on the avoidance zone search. During the search process, heuristic function estimation and actual cost are comprehensively utilized to ensure the vehicle's orientation and position in continuous space, while also reducing computational load, increasing path search speed, and achieving higher search efficiency.

[0113] It is worth mentioning that a suboptimal initial path can be quickly found using a loose boundary based on a heuristic search algorithm. Then, if time permits, the boundary can be gradually narrowed to improve the quality of the suboptimal initial path. This can balance the time constraints of solving the initial path and the quality requirements of the initial path.

[0114] In specific application scenarios, step 131, which involves using a heuristic search algorithm to search for the vehicle's initial path on the target road based on the obstacle avoidance area, specifically includes:

[0115] Step 131-1: Discretize the target road based on the obstacle data to obtain a grid map of the target road.

[0116] Step 131-2: Determine the starting node and target node for the vehicle's path search in the grid map.

[0117] Step 131-3: Based on the obstacle avoidance area, use a heuristic function to calculate the path cost of the starting node and the target node.

[0118] The heuristic function can be Euclidean distance, Manhattan distance, or other suitable heuristic functions. Path cost may include the estimated cost from the starting node's attitude to the target node's attitude, the cost of the next turn-over, the distance cost, and the cumulative turn cost. Specifically, Euclidean distance can be used to calculate the distance cost.

[0119] Step 131-4: Perform a discrete search based on the path cost to obtain the initial path.

[0120] In this embodiment, the target road is divided into a grid, with each grid representing a discrete region, ensuring that the generated initial path avoids obstacles. Based on the vehicle's current position and the target road's endpoint, the positions of the starting and target nodes are determined on the grid map to guide the path search direction and accelerate the search process. Using a heuristic search-based algorithm, starting from the starting node and aiming to minimize path cost, a discrete search is performed on the grid map until the initial path to the target node is reached or no feasible path can be found. Thus, by guiding the search through a priority queue and a heuristic function, unnecessary exploration is avoided, and the search quickly converges to the optimal or near-optimal solution, improving path search efficiency while ensuring path safety and smoothness.

[0121] Step 132: Update the initial path based on the comparison results to obtain the forward path.

[0122] In this embodiment, after determining the initial path, the initial path is fine-tuned by combining the comparison results of the safety distance and the radius of the envelope circle to find a forward path with good transition within the vehicle's driving range and that can avoid obstacles. This ensures that the planned forward path avoids obstacles and conforms to the actual operating limitations of the vehicle, ensuring that the vehicle can safely pass through obstacles and reach the target position.

[0123] Step 140: Control the vehicle to travel on the target road using the forward path.

[0124] The vehicle control method based on a convex polygon road structure provided in this application detects the likelihood of a vehicle colliding with an obstacle based on a comparison between the radius of the vehicle's outer contour envelope circle and the safe distance between the center of the vehicle's envelope circle and an obstacle in the target road. If the obstacle avoidance requirement is met, the avoidance area of ​​the obstacle in the target road is used to determine the vehicle's drivable range. The vehicle's forward path on the target road is predicted based on the drivable range and the collision detection results. Finally, the vehicle is controlled to travel on the target road according to the predicted forward path. This transforms the vehicle collision detection problem into a problem where the vehicle's envelope circle is constrained within a convex polygon, simplifying the mathematical description of the vehicle's outer contour envelope circle collision detection problem, effectively reducing the complexity of path planning, and thus enabling safe, efficient, and robust narrow-lane passage trajectories under planning time constraints. Even on target roads with upstream environmental perception blind spots, timely and accurate vehicle movement control can be made, reducing braking waiting time, improving the continuity of the autonomous driving path, and meeting the driver's requirements for autonomous driving time and safety.

[0125] Furthermore, in one embodiment, before step 140, the vehicle control method based on a convex polygonal road structure further includes: establishing a vehicle dynamics equation based on the vehicle's pose, and discretizing the vehicle dynamics equation to obtain dynamic nonlinear constraints; constructing obstacle linear constraints based on the avoidance area and the forward path; establishing a preset optimization model based on the dynamic nonlinear constraints and the obstacle linear constraints; and optimizing the forward path based on the preset optimization model.

[0126] In this embodiment, based on the obtained forward path, nonlinear constraints on vehicle dynamics are constructed using vehicle dynamics, and linear constraints on obstacles are constructed using the avoidance zone and the forward path. A preset optimization model is established based on the nonlinear constraints of dynamics and the linear constraints of obstacles. The optimized forward path is obtained by solving the preset optimization model. Thus, by comprehensively considering the constraints of multiple objectives, including the linear constraints of convex polygon obstacles and the nonlinear constraints of vehicle kinematics, a nonlinear constraint optimization problem is established and iteratively solved. This aims to make the route planning close to optimal in multiple aspects, and to plan a forward path that satisfies vehicle kinematic constraints and avoids obstacles on both sides of the road. This assists drivers of different experience levels, improves vehicle driving safety, effectively protects the driver's personal safety, and provides the driver with a good driving experience.

[0127] For example, using the longitudinal coordinate of the midpoint of the rear axle of the vehicle x Horizontal coordinates y Based on the yaw angle θ, the front wheel steering angle δ, and the velocity v, the system state vector is defined as z = [xy θ δ v] T The system control vector u = [γa] is defined based on the rate of change of the vehicle's front wheel steering angle γ and the acceleration a. T Then the vehicle kinematic equations It can be represented as:

[0128]

[0129] In the formula, L is the vehicle wheelbase.

[0130] Discretizing the vehicle kinematic equations using the fourth-order Runge-Kutta integral method yields the following equations:

[0131]

[0132] In the formula, h is the calculation step size; K1, K2, K3 and K4 are the coefficients of the fourth-order Runge-Kutta integral method.

[0133] For example, K1 can be represented as:

[0134]

[0135] K2 can be represented as:

[0136]

[0137] K3 can be represented as:

[0138]

[0139] K4 can be represented as:

[0140]

[0141] In the formula, x k Let y be the longitudinal coordinate of the midpoint of the rear axle of the vehicle at time k; k Let θ be the lateral coordinate of the midpoint of the rear axle of the vehicle at time k; k Let δ be the vehicle yaw angle at time k; k Let v be the front wheel steering angle of the vehicle at time k; k Let k be the vehicle speed at time k.

[0142] like Figure 9 As shown, considering both the linear constraints of the convex polygon obstacle and the nonlinear constraints of vehicle kinematics, the following preset optimization model is established:

[0143]

[0144] In the formula, E is the set of equality constraints; I is the set of inequality constraints; h i (z k ,u k ) = 0 is the equality constraint function; h i (z k ,u k )≤0 is the inequality constraint function; J N Represented as:

[0145]

[0146] J k Represented as:

[0147]

[0148] In the formula, W N,l W N,θ W N, δ and W N,v W represents the terminal state weighting coefficient. k,l W k,θ W k, δ, W k,v W k,γ and W k,a For process weighting coefficients; l k,ref x is the reference offset. k,ref y k,ref and θ k,ref The reference longitudinal coordinate, reference lateral coordinate, and reference yaw angle are used for the midpoint of the vehicle's rear axle.

[0149] By restricting the range of variable values ​​through inequality and equality constraints, the solution space is narrowed down to the feasible region that satisfies the constraints. The solution that satisfies these constraints is thus the solution to the optimization problem. Finally, the augmented Lagrange method is used to solve the path optimization problem described by the pre-defined optimization model, generating the optimized forward path.

[0150] It should be noted that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0151] Furthermore, such as Figure 10 As shown, as a specific implementation of the above-mentioned vehicle control method based on convex polygon road structure, this application embodiment provides a vehicle control device 300 based on convex polygon road structure. The vehicle control device 300 based on convex polygon road structure includes: a determination module 301, a path generation module 302, and a control module 303.

[0152] The determining module 301 is used to construct the vehicle's envelope circle based on the vehicle's structural data; and to determine the safe distance between the center of the vehicle's envelope circle and the obstacles in the target road and the obstacle avoidance area based on the obstacle data of the target road where the vehicle is located.

[0153] The path generation module 302 is used to predict the vehicle's forward path on the target road based on the comparison results of the safety distance and the radius of the envelope circle and the avoidance area.

[0154] The control module 303 is used to control the vehicle's movement on the target road according to the previous path.

[0155] In this embodiment, the likelihood of a vehicle colliding with an obstacle is detected by comparing the radius of the vehicle's outer contour envelope circle and the safe distance between the center of the vehicle's envelope circle and an obstacle in the target road. If the obstacle avoidance requirement is met, the avoidance zone of the obstacle in the target road is used to determine the vehicle's drivable range. The vehicle's forward path on the target road is predicted based on the drivable range and the collision detection results. Finally, the vehicle is controlled to travel on the target road according to the predicted forward path. This transforms the vehicle collision detection problem into a problem where the vehicle's outer contour envelope circle is constrained within a convex polygon, simplifying the mathematical description of the vehicle's outer contour envelope circle collision detection problem, effectively reducing the complexity of path planning, and thus enabling safe, efficient, and robust narrow-lane passage trajectories under time constraints. Even in target roads with upstream environmental perception blind spots, timely and accurate vehicle movement control can be made, reducing braking waiting time, improving the continuity of the autonomous driving path, and meeting the driver's requirements for autonomous driving time and safety.

[0156] Further, the determining module 301 is specifically used to determine at least one set of obstacle points for the target road based on obstacle data, wherein the set of obstacle points includes less than or equal to a preset number of points to be processed, the points to be processed are adjacent obstacle points in a distance sequence, and the distance sequence is obtained by sorting the x-coordinates of the obstacle points; the obstacle point set is triangulated to obtain a Deloni triangulation network; the centers of the circumcircles of adjacent triangles in the Deloni triangulation network are connected to generate a Vinno diagram, wherein the Vinno diagram includes at least one Vinno polygon; and a safe distance is determined based on the distance between the grid within the Vinno polygon and the center point of the Vinno polygon.

[0157] Furthermore, module 301 is specifically used to determine the obstacle outline based on obstacle data; determine the tangent line between the obstacle boundary near the envelope circle and the expansion circle containing the center of the envelope circle in the obstacle outline; and generate an avoidance area based on the tangent line.

[0158] Furthermore, the path generation module 302 is specifically used to search for the initial path of the vehicle on the target road using a heuristic search algorithm based on the obstacle avoidance area; and to update the initial path based on the comparison results to obtain the forward path.

[0159] Furthermore, the path generation module 302 is specifically used to discretize the target road based on obstacle data to obtain a grid map of the target road; determine the starting node and target node of the vehicle's path search in the grid map; calculate the path cost of the starting node and target node using a heuristic function based on the obstacle avoidance area; and perform a discrete search based on the path cost to obtain the initial path.

[0160] Furthermore, the vehicle control device 300 based on the convex polygon road structure also includes: an optimization module (not shown in the figure), which is used to establish vehicle dynamics equations based on the vehicle's pose, and to discretize the vehicle dynamics equations to obtain dynamic nonlinear constraints; to construct obstacle linear constraints based on the avoidance area and the forward path; to establish a preset optimization model based on the dynamic nonlinear constraints and the obstacle linear constraints; and to optimize the forward path based on the preset optimization model.

[0161] Furthermore, the vehicle control device 300 based on the convex polygonal road structure also includes: an expansion module (not shown in the figure), which is used to construct a polynomial curve based on preset factors; and to expand the obstacle avoidance area based on the polynomial curve.

[0162] Specific limitations regarding the vehicle control device based on convex polygon road structures can be found in the above-described limitations of the vehicle control method based on convex polygon road structures, and will not be repeated here. Each module in the aforementioned vehicle control device based on convex polygon road structures can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0163] Based on the above Figure 1 Accordingly, embodiments of this application also provide a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. Figure 1 The vehicle control method based on a convex polygon road structure is shown.

[0164] Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, or portable hard drive) and includes several instructions to cause a computer device (such as a personal computer, server, or network device) to execute the methods described in the various implementation scenarios of this application.

[0165] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages and stores the hardware and software resources of a computer device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software within the physical device.

[0166] Based on the above Figure 1 The method shown, and Figure 10 To achieve the above objectives, the present application also provides a computer device, specifically a personal computer, server, network device, vehicle-mounted device, etc., as shown in the virtual device embodiment. This computer device includes at least one processor; the processor is coupled to a memory, which stores a computer program that runs on the processor; the processor executes the computer program to achieve the above-described... Figure 1 The vehicle control method based on a convex polygon road structure is shown.

[0167] Optionally, the computer device may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Bluetooth interfaces, Wi-Fi interfaces), etc.

[0168] Those skilled in the art will understand that the computer device structure provided in this embodiment does not constitute a limitation on the computer device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0169] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms, or by hardware implementation to construct the vehicle's envelope circle based on the vehicle's structural data; based on the obstacle data of the target road where the vehicle is located, determine the safe distance between the center of the vehicle's envelope circle and the obstacles in the target road and the obstacle avoidance area; based on the comparison results of the safe distance and the radius of the envelope circle and the avoidance area, predict the vehicle's forward path on the target road; and control the vehicle's movement on the target road according to the forward path. In the embodiments of this application, the vehicle collision detection problem is transformed into a problem of the vehicle's envelope circle constrained within a convex polygon, simplifying the mathematical description of the vehicle's outer contour envelope circle collision detection problem, effectively reducing the complexity of path planning solutions, and thus obtaining a safe, efficient, and robust narrow-lane passage trajectory under the premise of planning time constraints. Even if there are blind spots in the upstream environmental perception of the target road, the vehicle's movement control can be made in a timely and accurate manner, reducing braking waiting time, improving the continuity of the autonomous driving path, and meeting the driver's requirements for autonomous driving time and safety.

[0170] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application. Those skilled in the art will understand that the modules in the apparatus of the embodiment can be distributed within the apparatus of the embodiment as described, or can be modified to be located in one or more apparatuses different from this embodiment. The modules of the above-described embodiment can be combined into one module, or further divided into multiple sub-modules.

[0171] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of any particular implementation scenario. The above disclosures are merely a few specific implementation scenarios of this application; however, this application is not limited thereto, and any variations conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. A vehicle control method based on a convex polygon road structure, characterized in that, The method includes: Construct the vehicle's envelope circle based on the vehicle's structural data; Based on the obstacle data of the target road where the vehicle is located, determine the safe distance between the center of the vehicle's envelope circle and the obstacle in the target road, and the obstacle avoidance area; Based on the comparison results of the safety distance and the radius of the envelope circle and the avoidance area, the vehicle's forward path on the target road is predicted; The vehicle is controlled to travel on the target road according to the stated forward path.

2. The vehicle control method based on a convex polygon road structure according to claim 1, characterized in that, Determining the safe distance between the center of the vehicle's envelope circle and obstacles in the target road based on obstacle data of the target road includes: Based on the obstacle data, at least one set of obstacle points for the target road is determined, wherein the set of obstacle points includes less than or equal to a preset number of points to be processed, and the points to be processed are adjacent obstacle points in a distance sequence, the distance sequence being sorted according to the x-coordinates of the obstacle points; The obstacle point set is triangulated to obtain a Deloni triangulation network; Connect the circumcenters of adjacent triangles in the Deloni triangulation to generate a Vinno diagram, wherein the Vinno diagram includes at least one Vinno polygon; The safety distance is determined based on the distance between the grid within the Vino polygon and the center point of the Vino polygon.

3. The vehicle control method based on a convex polygonal road structure according to claim 1, characterized in that, The step of determining the obstacle avoidance zone based on obstacle data of the target road where the vehicle is located includes: The obstacle outline is determined based on the obstacle data; Determine the tangent line between the boundary of the obstacle near the envelope circle and the expansion circle containing the center of the envelope circle in the obstacle profile; The avoidance area is generated based on the tangent.

4. The vehicle control method based on a convex polygon road structure according to claim 1, characterized in that, The step of predicting the vehicle's path on the target road based on the comparison result of the safety distance and the radius of the envelope circle and the avoidance area includes: Based on the obstacle avoidance area, a heuristic search algorithm is used to search for the initial path of the vehicle on the target road; The initial path is updated based on the comparison results to obtain the forward path.

5. The vehicle control method based on a convex polygonal road structure according to claim 4, characterized in that, The step of searching the initial path of the vehicle on the target road using a heuristic search algorithm based on the obstacle avoidance area includes: The target road is discretized based on the obstacle data to obtain a grid map of the target road; Determine the starting node and target node for the vehicle's path search in the grid map; Based on the obstacle avoidance area, the path cost of the starting node and the target node is calculated using a heuristic function; The initial path is obtained by performing a discrete search based on the path cost.

6. The vehicle control method based on a convex polygonal road structure according to any one of claims 1 to 5, characterized in that, The method further includes: The vehicle dynamics equations are established based on the vehicle's pose, and the vehicle dynamics equations are discretized to obtain dynamic nonlinear constraints. Based on the avoidance area and the forward path, construct linear constraints for the obstacles; Based on the aforementioned nonlinear dynamic constraints and linear obstacle constraints, a preset optimization model is established; The forward path is optimized based on the preset optimization model.

7. The vehicle control method based on a convex polygonal road structure according to any one of claims 1 to 5, characterized in that, The method further includes: Construct multiple polynomial curves based on preset factors; The obstacle avoidance area is expanded based on multiple polynomial curves.

8. A vehicle control device based on a convex polygonal road structure, characterized in that, The device includes: The determination module is used to construct the envelope circle of the vehicle based on the vehicle's structural data; and, Based on the obstacle data of the target road where the vehicle is located, determine the safe distance between the center of the vehicle's envelope circle and the obstacle in the target road, and the obstacle avoidance area; The path generation module is used to predict the vehicle's forward path on the target road based on the comparison result of the safety distance and the radius of the envelope circle and the avoidance area; A control module is used to control the vehicle to travel on the target road along the said forward path.

9. A computer-readable storage medium having a program or instructions stored thereon, characterized in that, When the program or instructions are executed by the processor, they implement the steps of the vehicle control method based on a convex polygonal road structure as described in any one of claims 1 to 7.

10. A computer device comprising at least one processor, said processor being coupled to a memory storing a computer program running on said processor, characterized in that, When the processor executes the program, it implements the vehicle control method based on a convex polygonal road structure as described in any one of claims 1 to 7.

Images