Automatic driving decision method and device, electronic equipment and storage medium

Abstract

The application relates to the technical field of automatic driving, and discloses an automatic driving decision method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring perception information, wherein the perception information comprises self-vehicle environment information, self-vehicle state information and an initially set initial driving route; determining a real-time driving route and a real-time driving area based on the perception information; performing driving mode decision according to a driving space of the real-time driving area to obtain a first decision result, wherein the first decision result is a lane keeping mode, a lane changing driving mode or an intersection turning mode; and controlling the self-vehicle to drive along the real-time driving route based on the first decision result. The embodiment of the application can improve the synchronization and safety of low-speed automatic driving in a specific area range.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to an autonomous driving decision-making method, device, electronic device, and storage medium. Background Technology

[0002] As a crucial scenario for realizing autonomous driving, low-speed autonomous driving within specific areas (such as parks, scenic spots, or large buildings) has enormous commercial application potential. Among these, decision-making, path planning, and control systems are key technologies for autonomous driving.

[0003] Currently, algorithms for behavioral decision-making mainly include rule-based and learning-based algorithms. Rule-based methods have advantages in the breadth of scenario traversal, strong logical interpretability, and ease of modular design based on scenarios. However, their system architecture limits their ability to handle complex situations and results in low control precision due to certain bottlenecks in scenario traversal depth and decision accuracy. Learning-based behavioral decision-making systems offer the advantage of deep scenario traversal; for a specific scenario, big data systems can more easily cover all conditions. However, learning algorithms lack the advantage of broad scenario traversal, requiring a large amount of experimental data as learning samples and exhibiting poor real-time performance. Summary of the Invention

[0004] The purpose of this invention is to provide an autonomous driving decision-making method, device, electronic device, and storage medium, which aims to improve the synchronization and safety of low-speed autonomous driving within a specific area.

[0005] Firstly, an autonomous driving decision-making method is provided, including:

[0006] Acquire perception information, including vehicle environment information, vehicle status information, and a pre-set initial driving route;

[0007] Based on perceived information, determine the real-time driving route and real-time driving area;

[0008] Based on the driving space of the real-time driving area, a driving mode decision is made to obtain a first decision result, which is either lane keeping mode, lane changing driving mode, or intersection turning mode.

[0009] Based on the initial decision result, the vehicle is controlled to travel along the real-time driving route.

[0010] In some embodiments, determining the real-time driving route and real-time driving area based on perception information includes:

[0011] The perceived information is processed to extract the results.

[0012] Based on the extraction results, the distance between the vehicle and the obstacle is determined, and the drivable area of ​​the vehicle is determined according to the distance between the vehicle and the obstacle, thus obtaining the real-time driving area;

[0013] The initial driving route is adjusted based on the real-time driving area to obtain the real-time driving route.

[0014] In some embodiments, when there are no obstacles within the first distance in front of the vehicle or when there is a vehicle within the first distance in front of the vehicle, the first decision result is lane keeping mode.

[0015] The step of controlling the vehicle to travel along the real-time driving route based on the first decision result includes:

[0016] Determine if there are any vehicles within the first distance range in front of your vehicle;

[0017] If so, the real-time driving area is divided into a following area and a collision avoidance area. If the vehicle is in the following area, the target speed of the vehicle is adjusted according to the actual distance between the vehicle and the vehicle in front, so that the target speed is negatively correlated with the actual distance between the vehicle and the vehicle in front. If the vehicle is in the collision avoidance area, the target speed is adjusted to zero.

[0018] If not, control the vehicle to travel at a constant speed along the initial driving route.

[0019] In some embodiments, when there is an obstacle within the second distance in front of the vehicle, the first decision result is a lane-changing driving mode;

[0020] The step of controlling the vehicle to travel along the real-time driving route based on the first decision result includes:

[0021] Based on the distance between the vehicle and the obstacle, find several aiming points, and use the Bézier curves constructed from the aiming points to draw the route change;

[0022] Obstacles are subjected to boundary expansion processing, and the expansion radius of the obstacles is set based on the current road change route and the vehicle's environmental information;

[0023] Determine whether the expansion radius of the obstacle is greater than the width of the obstacle;

[0024] If so, control the vehicle to continue along the current change route;

[0025] If not, reset the position of the preview point, redraw the change route using the Bézier curve constructed with the reset preview point, and if the steering angle of the vehicle corresponding to the redrawn change route is greater than the steering angle of the vehicle corresponding to the previously drawn change route, return to the step of performing boundary expansion processing on the obstacle.

[0026] In some embodiments, controlling the vehicle to travel along the current change route includes:

[0027] When no continuous obstacles are detected, the vehicle after lane change is controlled to pass through the obstacles and then change back to the lane of the initial driving route; when continuous obstacles are detected, the vehicle after lane change is controlled to pass through the continuous obstacles and then change back to the lane of the initial driving route.

[0028] The continuous obstacles are those whose distance from the previous obstacle in the same lane is less than a preset obstacle distance threshold.

[0029] In some embodiments, when the vehicle arrives at the intersection, the first decision result is an intersection turning mode;

[0030] The step of controlling the vehicle to travel along the real-time driving route based on the first decision result includes:

[0031] The system detects other vehicles at the current intersection. If no other vehicles are detected within the detection range, the system waits for other vehicles within the detection range to leave the detection range before controlling the vehicle to turn at the current intersection.

[0032] In some embodiments, the autonomous driving decision-making method further includes:

[0033] The driver's braking intention is obtained by collecting electroencephalogram (EEG) data, and brainwave information is obtained.

[0034] The brainwave information is preprocessed, and the preprocessed brainwave information is classified and tested using a trained convolutional neural network to obtain the classification results.

[0035] Based on the classification results, the driver's driving intention is identified, and the identification result is converted into a control command to obtain the second decision result.

[0036] Based on the second decision result, the vehicle is controlled to enter a braking state.

[0037] Secondly, an autonomous driving decision-making device is provided, the device comprising:

[0038] The acquisition module is used to acquire perception information, which includes vehicle environment information, vehicle status information, and a pre-set initial driving route.

[0039] The determination module is used to determine the real-time driving route and real-time driving area based on the perceived information;

[0040] The decision module is used to make a driving mode decision based on the driving space of the real-time driving area and obtain a first decision result, which is lane keeping mode, lane changing driving mode or intersection turning mode.

[0041] The control module is used to control the vehicle to travel along the real-time driving route based on the first decision result.

[0042] Thirdly, an electronic device is provided, the electronic device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the autonomous driving decision-making method of the first aspect.

[0043] Fourthly, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements the autonomous driving decision-making method of the first aspect.

[0044] The beneficial effects of this invention are as follows: by acquiring perception information to determine the vehicle environment, vehicle state, and pre-set initial driving route, the real-time driving route and real-time driving area are determined. When the vehicle is driving along the real-time driving route, the driving mode decision is made based on the driving space of the real-time driving area. The driving mode of autonomous driving is determined by the decision, thereby controlling the vehicle to drive along the real-time driving route based on the first decision result, thereby improving the synchronization and safety of low-speed autonomous driving in a specific area. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating an autonomous driving decision-making method as shown in the first embodiment.

[0046] Figure 2 yes Figure 1 A flowchart illustrating step S102 in the process.

[0047] Figure 3 yes Figure 1 The first flowchart of step S104 is shown.

[0048] Figure 4 yes Figure 1 The second process diagram for step S104 is shown.

[0049] Figure 5 yes Figure 1 The third process diagram for step S104 in the process.

[0050] Figure 6 This is a schematic diagram of the structure of the autonomous driving decision-making device provided in the embodiments of this application.

[0051] Figure 7 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application.

[0052] Figure 8 This is a schematic diagram of drawing a route change path shown in the first embodiment.

[0053] Figure 9 This is a schematic diagram of drawing a route change path, as shown in the second embodiment.

[0054] Figure 10 This is a schematic diagram illustrating the drawing of a route change path, as shown in the third embodiment. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0056] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0058] First, let's analyze some of the terms used in this application:

[0059] 1) Artificial Intelligence (AI)

[0060] Artificial intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess the functions of perception, reasoning, and decision-making.

[0061] Artificial intelligence (AI) is a comprehensive discipline encompassing a wide range of fields, including both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies primarily include computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0062] 2) Machine Learning (ML)

[0063] Machine learning is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory, among others. It specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence; its applications span all areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and learn-by-doing.

[0064] In related technologies, behavioral decision-making algorithms mainly include rule-based and learning-based algorithms. Rule-based decision-making methods have advantages in the breadth of scenario traversal, strong logical interpretability, and ease of modular design according to scenarios. There are many examples of decision systems using finite state machines both domestically and internationally. However, their system structure determines that they have certain bottlenecks in terms of scenario traversal depth and decision accuracy, making it difficult to handle complex working conditions. The advantage of learning-based behavioral decision-making systems is their depth of scenario traversal. For a specific scenario, big data systems can more easily cover all working conditions. However, learning algorithms do not have the advantage of broad scenario traversal, and the learning models required for different scenarios may be completely different. Machine learning requires a large amount of experimental data as learning samples. The decision-making effect depends on data quality. Insufficient samples, poor data quality, and unreasonable network structure can lead to problems such as overlearning and underlearning.

[0065] Based on this, embodiments of this application provide an autonomous driving decision-making method, device, electronic device, and storage medium, aiming to improve the synchronization and safety of low-speed autonomous driving within a specific area.

[0066] The autonomous driving decision-making method, device, electronic device, and storage medium provided in this application are specifically described through the following embodiments. First, the recommended method in the embodiments of this application is described.

[0067] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0068] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0069] The autonomous driving decision-making method provided in this application relates to the field of artificial intelligence technology. The autonomous driving decision-making method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing a table information extraction method, etc., but is not limited to the above forms.

[0070] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0071] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards of the relevant countries and regions. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirects to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data for the proper functioning of the embodiments of this application obtained.

[0072] Figure 1 This is a flowchart illustrating an autonomous driving decision-making method as shown in the first embodiment. Figure 1 The method may include, but is not limited to, steps S101 to S104.

[0073] Step S101: Obtain sensory information.

[0074] Perception information includes vehicle environment information, vehicle status information, and a pre-set initial driving route. In essence, perception information refers to external data and vehicle status data that can affect safety during driving, such as vehicle environment information and vehicle status information.

[0075] In one possible implementation, the perceived information is vehicle environment information and vehicle status information collected by a device with a perception module, including LiDAR, camera, navigation system, and vehicle infotainment system, etc.

[0076] In one possible implementation, vehicle environment information can refer to all data related to vehicle driving safety collected by sensing modules such as LiDAR or cameras in the environment where the vehicle is located. For example, vehicle environment information can be the driving status of all vehicles in the environment where the vehicle is located, or the driving environment in the environment where the vehicle is located (such as lane line data, obstacle data, real-time positioning data, and map data). This application embodiment does not make any special limitation on this.

[0077] In one possible implementation, vehicle status information can refer to all data related to the vehicle's own driving status collected by sensing modules such as the vehicle system, such as vehicle status information, actual vehicle speed data, and sensor data from the steering wheel, brake pedal, accelerator pedal, and clutch. This application does not impose any special limitations on this.

[0078] In one possible implementation, the pre-set initial driving route can be obtained by the navigation software through the vehicle's location information. Before driving begins, the destination is entered, and the navigation software generates several driving routes based on the vehicle's current location and the destination location. The driver or passenger can select one of the driving routes as the initial driving route by interacting with the vehicle's infotainment system.

[0079] Step S102: Based on the perception information, determine the real-time driving route and the real-time driving area.

[0080] In one possible implementation, the driving route and real-time driving area can be determined by autonomous vehicle driving control planning based on vehicle environmental information and vehicle status information. The current real-time driving area is determined by analyzing the vehicle environmental information, and then the initial driving route is locally adjusted based on the current real-time driving area on the premise of ensuring driving safety, so as to obtain the real-time driving route.

[0081] The driving area can be understood as the area where vehicles can stay while ensuring driving safety. The driving route can be understood as the lane that indicates the direction that vehicles should travel when entering a section at an intersection. The driving route can be confirmed by lane lines. The purpose is to clarify the direction of travel, so that each vehicle stays in its lane and reduces traffic pressure. Lane lines include white dashed lines, white solid lines, directional lines, deceleration warning lines, etc.

[0082] Specifically, when vehicle environment information is detected, the system acquires the corresponding image resolution data, transmission cycle data, timestamp data, and color channel information. This data is then associated with the vehicle driving information in image form. Similarly, when vehicle status information is detected, the system acquires the corresponding acquisition parameter data, acquisition cycle data, and timestamp data. This data is then associated with the vehicle status information. Finally, the system associates the vehicle environment information and vehicle status information at the same time point, thereby improving the device's calculation speed, enhancing data accuracy, and ensuring the precision of real-time driving routes and areas.

[0083] Step S103: Make a driving mode decision based on the driving space of the real-time driving area to obtain the first decision result.

[0084] The first decision result is either lane keeping mode, lane changing mode, or intersection turning mode. It is understood that lane keeping mode, lane changing mode, and intersection turning mode are all preset driving modes. The driving mode decision is based on the real-time driving area's driving space to determine the vehicle's current road conditions. When the real-time driving area's driving space meets the preset conditions of the corresponding driving mode, the first decision result for that driving mode is output. In this embodiment, lane keeping mode can be understood as the driving mode where the vehicle maintains its current lane; lane changing mode can be understood as the driving mode where the vehicle switches from its current lane to another lane; and intersection turning can be understood as the driving mode where the vehicle turns based on its initial or real-time driving route when it reaches an intersection.

[0085] In one possible implementation, the vehicle receives real-time information about its surrounding environment, extracts information about obstacles to the vehicle, determines the driving space of the real-time driving area based on the distance and relative speed of obstacles ahead in the current lane, and then makes a judgment and decision based on the vehicle's expected distance between vehicles and speed information to obtain the first decision result.

[0086] Specifically, when there are no obstacles in the current lane that affect the vehicle's driving range, or when the distance between the obstacles is greater than the minimum safe following distance, the vehicle enters lane keeping mode; when the distance between the obstacles in the current lane and the vehicle is less than the safe following distance, and the relative speed between the obstacles and the vehicle is less than zero and the difference is not greater than a certain range, lane keeping mode is entered to improve vehicle safety; when the distance between the obstacles in the current lane and the vehicle is less than the safe following distance, and the speed difference between the obstacles and the vehicle is large, affecting traffic efficiency but there is no alternative route, lane keeping mode is entered. In this case, if there is an alternative route, lane keeping mode is switched out and lane changing mode is switched to improve driving efficiency.

[0087] Step S104: Based on the first decision result, control the vehicle to travel along the real-time driving route.

[0088] In one possible implementation, a first decision result is sent to the vehicle's control system through interaction with the vehicle's control system. The vehicle's control system then converts the driving mode indicated by the received first decision result into a control command for the vehicle, controlling the vehicle to drive according to the driving mode indicated by the first decision result.

[0089] Please see Figure 2 In some embodiments, step S102 may include, but is not limited to, steps S201 to S203.

[0090] Step S201: Perform information extraction processing on the perceived information to obtain the extraction result.

[0091] In one possible implementation, information extraction processing of perceived information can be understood as calculating the corresponding accuracy, latency, and feedback action index values ​​based on the raw data of the collected perceived information.

[0092] Specifically, vehicle environmental information is extracted to obtain indicators for accuracy, time delay, and feedback actions. Accuracy indicators can be vehicle environmental information recognition accuracy and recall. To extract these, the number of correct and incorrect recognitions needs to be determined based on the recognition results. The accuracy is obtained by dividing the number of correct recognitions by the total number of recognitions. Recall can be calculated using a similar method. Time delay indicators can be recognition performance (recognition speed), calculated based on the total number of recognitions and the time taken to complete each or all recognitions. Feedback actions indicators can include brake / accelerator pedal changes, steering wheel angle, and steering wheel direction. If feedback indicators exist in the vehicle environmental information, they can be directly filtered from the target driving data.

[0093] Step S202: Determine the distance between the vehicle and the obstacle based on the extraction results, and determine the drivable area of ​​the vehicle based on the distance between the vehicle and the obstacle to obtain the real-time driving area.

[0094] In one possible implementation, the drivable area of ​​the vehicle is determined based on the distance between the vehicle and the obstacle in front and the distance between the vehicle and the obstacles on its left and right sides. This is achieved by using radar ranging or camera imaging. The drivable area of ​​the vehicle is divided based on the distance between the vehicle and the obstacles in each direction determined by radar ranging or by recognizing the distance between the vehicle and the obstacles in each direction in the image captured by the camera, thus obtaining the real-time drivable area.

[0095] Step S203: Adjust the initial driving route based on the real-time driving area to obtain the real-time driving route.

[0096] In one possible implementation, adjusting the initial driving route based on the real-time driving area can involve choosing to continue driving in the original lane or change lanes based on the driving space in the real-time driving area. If the driving space in the real-time driving area is sufficient, the initial driving route is not adjusted; if the driving space in the real-time driving area is insufficient, the initial driving route is adjusted. For example, when the distance between the vehicle and the vehicle in front is less than a preset following distance value, but the distance between the vehicle and an obstacle on its left is greater than a preset adjacent distance value, the vehicle is controlled to change lanes, generating a real-time driving route. The real-time driving route can be the left lane selected as the driving route, and the vehicle changes lanes to the left lane while traveling along the real-time driving route.

[0097] Please see Figure 3 In some embodiments, step S104 may include, but is not limited to, steps S301 to S303.

[0098] In this embodiment, if there are no obstacles within a first distance in front of the vehicle or if there are vehicles within a first distance in front of the vehicle, the first decision result is lane keeping mode.

[0099] Step S301: Determine whether there is a vehicle within a first distance range in front of the vehicle. If yes, proceed to step S302; otherwise, proceed to step S303.

[0100] Step S302: Divide the real-time driving area into a following area and a collision avoidance area. If the vehicle is in the following area, adjust the target speed of the vehicle according to the actual distance between the vehicle and the vehicle in front, so that the target speed is negatively correlated with the actual distance between the vehicle and the vehicle in front. If the vehicle is in the collision avoidance area, adjust the target speed to zero.

[0101] In one possible implementation, the real-time driving area can be divided into zones by pre-setting the range of the collision avoidance zone, including the distance in the driving direction and the distance in the turning direction. After the collision avoidance zone is determined, the remaining area of ​​the driving area is divided into the following zone.

[0102] For the area where the vehicle is following, the following relationship must be satisfied:

[0103] d stop <d actual ≤d stop +d follow ;

[0104] For the collision avoidance zone, the following relationship is satisfied:

[0105] d actual <d stop ;

[0106] Where, d stop d represents the distance of the collision avoidance zone in the direction of travel. actual d represents the actual distance between the vehicle and the vehicle in front. follow It indicates the distance of the area where the vehicle is following in the direction of travel.

[0107] For example, in the following mode, the preset collision avoidance zone is 5m away in the direction of travel. When there is a vehicle within 20m in front of the vehicle, the following zone is set 15m away from the collision avoidance zone in the direction of travel.

[0108] The target speed of the vehicle is adjusted based on the actual distance between the vehicle and the vehicle in front, so that the target speed is negatively correlated with the actual distance between the vehicle and the vehicle in front. This can be achieved by periodically collecting the actual speed of the vehicle and the actual distance between the vehicle and the vehicle in front. When the actual distance between the vehicle and the vehicle in front collected in the current collection period is less than that collected in the previous collection period, the target speed of the vehicle is increased, and the increased target speed is used as the actual speed of the vehicle. Conversely, when the actual distance between the vehicle and the vehicle in front collected in the current collection period is greater than that collected in the previous collection period, the target speed of the vehicle is decreased, and the decreased target speed is used as the actual speed of the vehicle.

[0109] For example, when the actual distance between the vehicle and the vehicle in front collected in the current collection cycle is less than the actual distance between the vehicle and the vehicle in front collected in the previous collection cycle, the target speed satisfies:

[0110] v d =v d-1 +1;

[0111] When the actual distance between the vehicle and the vehicle in front, as measured in the current acquisition cycle, is greater than the actual distance between the vehicle and the vehicle in front, as measured in the previous acquisition cycle, the target speed satisfies:

[0112] v d =v d-1 -1;

[0113] Among them, v d v represents the target velocity in the previous acquisition cycle. d-1 v represents the target velocity in the previous acquisition cycle. d <v0, where v0 represents the following speed threshold.

[0114] Step S303: Control the vehicle to travel at a constant speed along the initial driving route.

[0115] In one possible implementation, when the driving mode is constant speed driving, the actual distance between the vehicle and the vehicle in front is within the preset range of constant speed driving, and the vehicle is controlled to drive at a constant speed along the initial driving route according to the preset constant speed driving speed.

[0116] Please see Figure 4 In some embodiments, step S104 may include, but is not limited to, steps S401 to S405.

[0117] In this embodiment, when there is an obstacle within a second distance in front of the vehicle, the first decision is to change lanes.

[0118] Step S401: Find several aiming points based on the distance between the vehicle and the obstacle, and use the Bézier curve constructed from the aiming points to draw the route change.

[0119] A Bézier curve, also known as a Bézier curve or Bézier line, is a mathematical curve used in two-dimensional graphics applications. Bézier curves are created and edited by controlling four points on the curve (a start point, an end point, and two separate midpoints), with the control line located in the center of the curve playing a crucial role. This control line is virtual, intersecting the Bézier curve in the middle, and its two ends are control endpoints. Moving the endpoints changes the curvature (degree of bending) of the Bézier curve; moving the midpoints (i.e., moving the virtual control line) causes the Bézier curve to move uniformly while the start and end points remain locked.

[0120] Bézier curves have advantages such as continuous curvature, ease of tracking, and satisfaction of vehicle dynamics constraints. Therefore, this embodiment of the invention selects a fifth-order Bézier curve for route changing.

[0121] A fifth-order Bézier curve can be represented as:

[0122]

[0123] It can be represented as:

[0124]

[0125] Where B(t) represents the Bézier curve, P0, P1, P2, P3, P4 and P5 are all aiming points, t∈[0,1], n∈[0,5], i∈[0,5].

[0126] Several preview points can be found based on the distance between the vehicle and obstacles. These points can be selected within the real-time driving area, and the selected preview points can be connected to form a Bézier curve, which serves as the route for changing lanes. For example... Figure 8 As shown, the coordinates P of the six aiming points in the fifth-order Bézier curve are... i =(X i ,Y i The distance between the selected autonomous vehicle and the obstacle is determined, specifically:

[0127] First, determine the coordinates of the aiming points P0 and P5:

[0128] P0 = (X0, Y0);

[0129] P5 = (X5, Y5) = (L, w r );

[0130] Divide the distance between the vehicle and the obstacle into four parts to obtain the coordinates P1, P2, P3, and P4:

[0131] P1 = (X1, Y1) = (X5 / 4, Y0);

[0132] P2 = (X2, Y2) = (X5 / 2, Y0);

[0133] P3 = (X3, Y3) = (X5 / 2, Y5);

[0134] P4 = (X4, Y4) = (3X5 / 4, Y5);

[0135] Where L is the distance between the vehicle and the obstacle in the Frenet coordinate system detected by the perception system; w r Lane width information provided for lane line detection in the perception system.

[0136] Step S402: Perform boundary expansion processing on the obstacle, and set the expansion radius of the obstacle based on the current road change route and the vehicle's environmental information.

[0137] When using Bézier curves to draw road change routes, since the vehicle is treated as a point mass and its width is ignored, there is a possibility of collision with obstacles as the vehicle travels along the road change route. By applying boundary expansion to the obstacles, the distance between the vehicle and the obstacle while traveling along the road change route is compared with the radius of the collision with the expanded obstacle to perform a collision assessment. The formula for collision assessment can be expressed as:

[0138]

[0139] Where r is the radius of expansion of the obstacle. d is the width of the vehicle itself. min y represents the minimum safe distance between the desired obstacle and the vehicle. ob The distance between the center of the obstacle and the lane line in the Frenet coordinate system. The width of the obstacle detected by the sensing system.

[0140] Step S403: Determine whether the expansion radius of the obstacle is greater than the width of the obstacle. If yes, proceed to step S403; otherwise, proceed to step S404.

[0141] When determining whether the expansion radius of an obstacle is greater than the width of the obstacle, that is:

[0142]

[0143] Step S404: Control the vehicle to travel along the current change route.

[0144] Step S405: Reset the position of the preview point, and redraw the route change using the Bézier curve constructed with the reset preview point. The steering angle of the vehicle corresponding to the redrawn route change is greater than the steering angle of the vehicle corresponding to the previously drawn route change. Return to step S402.

[0145] Within the real-time driving area, a new aiming point is selected. First, the coordinates of aiming point P5 are redefined. By shortening the distance between P0 and P5, the vehicle's steering angle is increased, ensuring that the steering angle corresponding to the redrawn road change route is greater than that corresponding to the previously drawn road change route, thus reducing the risk of collision between the vehicle and obstacles. In this embodiment, the coordinates of aiming point P5 are changed using the following formula:

[0146] P5=(X5,Y5)=(4L / 5,w r ).

[0147] like Figure 9 As shown, after the vehicle travels along the current lane change route and bypasses the obstacle, it is controlled to return to the lane of the initial travel route. This can be achieved by using a mirrored Bézier curve to draw the route back to the lane of the initial travel route, thus returning to the lane of the initial travel route after bypassing the obstacle.

[0148] In some embodiments, such as Figure 10 As shown, controlling the vehicle to travel along the current lane-changing route can be done in two ways: when there are no continuous obstacles within the detection range, the vehicle is controlled to pass the obstacles after the lane change and then change back to the lane of the initial driving route; when there are continuous obstacles within the detection range, the vehicle is controlled to pass the continuous obstacles after the lane change and then change back to the lane of the initial driving route.

[0149] Among them, continuous obstacles are obstacles whose distance from the previous obstacle in the same lane is less than a preset obstacle distance threshold.

[0150] In one embodiment, when the vehicle arrives at the intersection, the first decision is to adopt the intersection turning mode. In this embodiment, other vehicles at the current intersection are detected. If no other vehicles appear within the detection range, the vehicle waits for other vehicles within the detection range to leave the detection range before controlling the vehicle to turn at the current intersection.

[0151] In the above embodiments, the control of the vehicle includes lateral control and longitudinal control.

[0152] Lateral control employs a pure tracking control algorithm. The formula for controlling the steering wheel angle during lane changes is:

[0153]

[0154] Where δ(t) is the front wheel steering angle of the vehicle calculated by the pure tracking control algorithm at time t, L′ is the wheelbase; k is the proportional coefficient, v x It represents the real-time speed of the autonomous vehicle, and l0 represents the minimum forward-looking distance.

[0155] Vertical control is based on the PID algorithm, specifically:

[0156] Construct a control output model based on the target velocity, wherein the control output model is as follows:

[0157]

[0158] Among them, U t This represents the control output quantity, e(t) = v d -v a v d V represents the target velocity. a K represents the current speed. P K represents the scaling factor. I K represents the integration time constant. D Represents the differential time constant;

[0159] The control output model is discretized to obtain a discrete control output model, which is as follows:

[0160]

[0161] Where T is the controller's instruction cycle, e j e represents the velocity deviation at time j. k e represents the velocity deviation at time k. k-1 This represents the velocity deviation at time k-1;

[0162] The control output is calculated using the discrete control output model, and the calculated control output is used to perform assisted driving control on the vehicle.

[0163] Figure 5 This is the first optional flowchart of the autonomous driving decision-making method provided in the embodiments of this application. Figure 1 Based on the implementation examples, Figure 5 The method may include, but is not limited to, steps S501 to S504.

[0164] Step S501: Obtain the driver's braking intention through EEG acquisition and obtain brainwave information.

[0165] In the embodiments of this application, brain nerve cells generate corresponding electrical waves when they react. These electrical waves are called brain waves. Collecting brain waves and expressing them in the form of signals is called brain electrical signals.

[0166] In one possible implementation, the EEG signal can be the driver's brainwave signal collected by a device with a brain-computer interface. The brain-computer interface has at least two electrodes, which are located in different areas of the driver's head during the signal collection process, so as to collect the brainwave signals generated in different areas of the driver.

[0167] Step S502: Preprocess the EEG information, and use the trained convolutional neural network to perform classification tests on the preprocessed EEG information to obtain classification results.

[0168] In one possible implementation, preprocessing of brainwave information can include signal processing procedures such as noise reduction, shaping, filtering, and signal amplification.

[0169] In one possible implementation, a trained convolutional neural network (CNN) is used to classify the preprocessed EEG information to obtain the classification result. Specifically, when EEG information is detected, the trained CNN can classify the EEG information and predict the driving intention type associated with the EEG signal. A cosine similarity matrix is ​​constructed based on the calculated cosine similarity. The trained CNN then predicts the type associated with this cosine similarity matrix to obtain the driving intention type associated with the EEG signal. For example, if the feature vectors of the EEG information all have 128 features, 128 cosine similarities can be calculated. A cosine similarity matrix can then be constructed based on these 128 similarities, and the trained CNN can predict the type associated with this cosine similarity matrix to obtain the driving intention type associated with the EEG signal.

[0170] The trained convolutional neural network is used to predict the driving intention type associated with the EEG signal, specifically:

[0171] The preprocessed EEG information is labeled to obtain labeled EEG information;

[0172] Labeled EEG information is input into a trained convolutional neural network in frame form for classification testing. The output of the trained convolutional neural network is then rearranged according to the labels of the labeled EEG information to serve as the classification result.

[0173] The trained convolutional neural network consists of five layers: 1) an input layer, 2) a first convolutional layer, 3) a second convolutional layer, 4) a fully connected layer, and 5) an output layer. The input EEG information is in the form of a 200*16 matrix, and the input to the fully connected layer is the output of the last convolutional layer.

[0174] y = sigmoid(WM + b);

[0175] Where y represents the output of the fully connected layer, W represents the weight matrix, b represents the bias, and M represents the output of the last convolutional layer.

[0176] After obtaining the label of each EEG signal segment in the labeled EEG information sample, the labeled EEG information is classified by a trained convolutional neural network to determine the driver's driving intention. A cosine similarity matrix is ​​constructed based on the calculated cosine similarity. The type of the cosine similarity matrix is ​​predicted by the convolutional neural network to obtain the driving intention type of the EEG signal sample. The output of the trained convolutional neural network is rearranged according to the pre-obtained label of each EEG signal segment to obtain the classification result.

[0177] Step S503: Based on the classification results, identify the driver's driving intention, obtain the identification result, convert the identification result into control commands, and obtain the second decision result.

[0178] Step S504: Based on the second decision result, control the vehicle to enter the braking state.

[0179] In one possible implementation, a device with brainwave information classification capabilities interacts with the vehicle's control system. The device sends the recognition results to the vehicle's control system, which then converts the received recognition results into control commands for the vehicle, thereby controlling the vehicle to enter a braking state based on a second decision result.

[0180] As can be seen from the above, in the autonomous driving decision-making method provided in this application embodiment, the vehicle environment, vehicle state and pre-set initial driving route are determined by acquiring perception information, thereby determining the real-time driving route and real-time driving area. When controlling the vehicle to drive along the real-time driving route, the driving mode decision is made based on the driving space of the real-time driving area. The driving mode of autonomous driving is determined by the decision, thereby controlling the vehicle to drive along the real-time driving route based on the first decision result, thereby improving the synchronization and safety of low-speed autonomous driving in a specific area.

[0181] To better implement the above methods, embodiments of the present invention also provide an assisted driving device, which can be integrated into electronic devices such as servers or terminals.

[0182] Please see Figure 6 This application also provides an autonomous driving decision-making device that can implement the autonomous driving decision-making method mentioned in the above embodiments. The device includes:

[0183] The acquisition module 601 is used to acquire perception information, including vehicle environment information, vehicle status information and a pre-set initial driving route;

[0184] The determination module 602 is used to determine the real-time driving route and real-time driving area based on the perception information;

[0185] The decision module 603 is used to make a driving mode decision based on the driving space of the real-time driving area and obtain a first decision result, which is lane keeping mode, lane changing driving mode or intersection turning mode.

[0186] The control module 604 is used to control the vehicle to travel along the real-time driving route based on the first decision result.

[0187] The specific implementation of this autonomous driving decision-making device is basically the same as the specific implementation of the above-described autonomous driving decision-making method, and will not be repeated here.

[0188] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned autonomous driving decision-making method. This electronic device can be any smart terminal, including a tablet computer, an in-vehicle computer, or similar device.

[0189] Please see Figure 7 , Figure 7 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0190] The processor 701 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0191] The memory 702 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 702 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 702 and is called and executed by the processor 701 using the autonomous driving decision-making method of the embodiments of this application.

[0192] The input / output interface 703 is used to implement information input and output;

[0193] The communication interface 704 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0194] Bus 705 transmits information between various components of the device (e.g., processor 701, memory 702, input / output interface 703, and communication interface 704);

[0195] The processor 701, memory 702, input / output interface 703, and communication interface 704 are connected to each other within the device via bus 705.

[0196] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described autonomous driving decision-making method.

[0197] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0198] The autonomous driving decision-making method, device, electronic device, and storage medium provided in this application obtain perception information to determine the vehicle environment, vehicle state, and a pre-set initial driving route, thereby determining the real-time driving route and real-time driving area. When controlling the vehicle to drive along the real-time driving route, the driving mode decision is made based on the driving space of the real-time driving area. The driving mode of autonomous driving is determined through the decision, thereby controlling the vehicle to drive along the real-time driving route based on the first decision result, thereby improving the synchronization and safety of low-speed autonomous driving within a specific area.

[0199] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0200] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0201] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0202] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0203] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0204] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0205] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0206] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0207] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0208] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0209] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. An autonomous driving decision-making method, characterized in that, include: Acquire perception information, including vehicle environment information, vehicle status information, and a pre-set initial driving route; Based on perceived information, determine the real-time driving route and real-time driving area; Based on the driving space of the real-time driving area, a driving mode decision is made to obtain a first decision result, which is either lane keeping mode, lane changing driving mode or intersection turning mode. Based on the first decision result, control the vehicle to travel along the real-time driving route; Where there are no obstacles within the first distance in front of the vehicle or there are vehicles within the first distance in front of the vehicle, the first decision result is lane keeping mode. The step of controlling the vehicle to travel along the real-time driving route based on the first decision result includes: Determine if there are any vehicles within the first distance range in front of your vehicle; If so, the real-time driving area is divided into a following area and a collision avoidance area. If the vehicle is in the following area, the target speed of the vehicle is adjusted according to the actual distance between the vehicle and the vehicle in front, so that the target speed is negatively correlated with the actual distance between the vehicle and the vehicle in front. If the vehicle is in the collision avoidance area, the target speed is adjusted to zero. If not, control the vehicle to travel at a constant speed along the initial driving route.

2. The autonomous driving decision-making method according to claim 1, characterized in that, The process of determining the real-time driving route and real-time driving area based on perception information includes: The perceived information is processed to extract the results. Based on the extraction results, the distance between the vehicle and the obstacle is determined, and the drivable area of ​​the vehicle is determined according to the distance between the vehicle and the obstacle, thus obtaining the real-time driving area; The initial driving route is adjusted based on the real-time driving area to obtain the real-time driving route.

3. The autonomous driving decision-making method according to claim 1, characterized in that, When there is an obstacle within the second distance in front of the vehicle, the first decision result is a lane-changing driving mode; The step of controlling the vehicle to travel along the real-time driving route based on the first decision result includes: Based on the distance between the vehicle and the obstacle, find several aiming points, and use the Bézier curves constructed from the aiming points to draw the route change; Obstacles are subjected to boundary expansion processing, and the expansion radius of the obstacles is set based on the current road change route and the vehicle's environmental information; Determine whether the expansion radius of the obstacle is greater than the width of the obstacle; If so, control the vehicle to continue along the current change route; If not, reset the position of the preview point, redraw the change route using the Bézier curve constructed with the reset preview point, and if the steering angle of the vehicle corresponding to the redrawn change route is greater than the steering angle of the vehicle corresponding to the previously drawn change route, return to the step of performing boundary expansion processing on the obstacle.

4. The autonomous driving decision-making method according to claim 3, characterized in that, The control of the vehicle to travel along the current changing route includes: When no continuous obstacles are detected, the vehicle after lane change is controlled to pass through the obstacles and then change back to the lane of the initial driving route; when continuous obstacles are detected, the vehicle after lane change is controlled to pass through the continuous obstacles and then change back to the lane of the initial driving route. The continuous obstacles are those whose distance from the previous obstacle in the same lane is less than a preset obstacle distance threshold.

5. The autonomous driving decision-making method according to claim 1, characterized in that, When the vehicle arrives at the intersection, the first decision result is the intersection turning mode; The step of controlling the vehicle to travel along the real-time driving route based on the first decision result includes: The system detects other vehicles at the current intersection. If no other vehicles are detected within the detection range, the system waits for other vehicles within the detection range to leave the detection range before controlling the vehicle to turn at the current intersection.

6. The autonomous driving decision-making method according to claim 1, characterized in that, Also includes: The driver's braking intention is obtained by collecting electroencephalogram (EEG) data, and brainwave information is obtained. The brainwave information is preprocessed, and the preprocessed brainwave information is classified and tested using a trained convolutional neural network to obtain the classification results. Based on the classification results, the driver's driving intention is identified, and the identification result is converted into a control command to obtain the second decision result. Based on the second decision result, the vehicle is controlled to enter a braking state.

7. An autonomous driving decision-making device, characterized in that, The device includes: The acquisition module is used to acquire perception information, which includes vehicle environment information, vehicle status information, and a pre-set initial driving route. The determination module is used to determine the real-time driving route and real-time driving area based on the perceived information; The decision module is used to make a driving mode decision based on the driving space of the real-time driving area and obtain a first decision result, which is lane keeping mode, lane changing driving mode or intersection turning mode. The control module is used to control the vehicle to travel along the real-time driving route based on the first decision result; Where there are no obstacles within the first distance in front of the vehicle or there are vehicles within the first distance in front of the vehicle, the first decision result is lane keeping mode. The step of controlling the vehicle to travel along the real-time driving route based on the first decision result includes: Determine if there are any vehicles within the first distance range in front of your vehicle; If so, the real-time driving area is divided into a following area and a collision avoidance area. If the vehicle is in the following area, the target speed of the vehicle is adjusted according to the actual distance between the vehicle and the vehicle in front, so that the target speed is negatively correlated with the actual distance between the vehicle and the vehicle in front. If the vehicle is in the collision avoidance area, the target speed is adjusted to zero. If not, control the vehicle to travel at a constant speed along the initial driving route.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the autonomous driving decision-making method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the autonomous driving decision-making method according to any one of claims 1 to 6.

Images